Partition 36 Blog Feed

Cyberpunks

2010-07-22T00:00:00-06:00

Well, the community compilation CD that my friend here at work was putting on is finally released! The official title of it is Emergent Collective One, and is comprised of 11 tracks. Partition 36 is featured on it with my brand new track, Cyberpunks.

You can listen to it on the official CD, or download it separately below in a few extra formats (mp3, Ogg Vorbis, or FLAC. The version below is slightly different in that it has a greater dynamic range (the way I intended it to sound), so if you’re on the fence about which one to grab, download it off of this page. I highly recommend checking out the other tracks, however. There’s a lot of great talent on the CD!

Cyberpunks
(mp3| ogg| flac)

Can’t hear the embedded audio? Get a better browser, such as Google Chrome or Firefox.

As far as Cyberpunks goes, when I first started the track, I didn’t have any intention on writing about cyberpunk ideas at all. It actually started out as a sort of spiritual successor to I Love Penguins, where it would have again talked about the virtues and benefits of Linux. Another idea I tossed around early on was making it more generally about Free/Libre Software, focusing on the GNU General Public License. But I couldn’t come up with the right words, and I was also concerned that doing something like this could potentially alienate listeners who don’t know about the free software movement.

After I had a few of the core beats and themes of the song done, I had the idea of instead focusing on the idea of “cyberpunk”. The term is most often used to describe a subgenre of science fiction, which focuses on post-industrial dystopias where technology permeates everything. The Bubblegum Crisis, Ghost In The Shell, RoboCop, and Blade Runner are classic examples from anime and film that fall into this subgenre, while William Gibson’s Neuromancer is a classic example from literature.

I had heard a remark made quite a few years ago that pointed out something interesting. People who enjoy this genre tend to wonder what life would be like in a cyberpunk world, sometimes to the point that they even wish one existed. But what people don’t understand is that the current world is already beginning to resemble a cyberpunk world. One just need to look at the Internet and hacker culture for examples.

The remark always struck me as interesting. I would say that it’s not 100% true since our world doesn’t look exactly like what we see in cyberpunk media, but there are some early signs of us entering one. This was the root of my inspiration for the song. I searched online for some sort of cyberpunk-related text, eventually settling on something called The Cyberpunk Manifesto, originally written by Christian As. Kirtchev. It fit perfectly with the feel of the song, as well as my ideas for it, so I recorded myself reading section of it. Things just sort of fell in place from there.

On the technical side of things, Cyberpunks uses a few new tricks and ideas that I’ve had floating around my head. The opening background noise was generated in Reaktor using a built-in synth called Space Drone. I took one of the presets and modified it a bit to create a pulsing background pad that, I believe, is totally random and never repeats (I can’t confirm this yet, as I haven’t dug into the guts of this synth). From there it builds into the opening “pingy” sound, which was actually the starting point for the entire song.

The bass is actually the same patch that I used in I Love Penguins, and is generated using Arturia’s Jupiter-8V software. Meanwhile, the saw sound at the end was done using my PolyEvolver Keyboard. Aside from a little bit of reverb and some very sight EQing to keep the mix clean, it’s very close to the original sound.

The drums need special mentioning. These were highly inspired by the track Hi-Scores by Boards of Canada. To achieve the effect of a swirling, bouncy set of noisy high hats, I took various samples and passed them through a few effects plugins (at least one was a distortion unit). From there I used a delay unit to make the sound bounce back and forth before finally entering an EQ with an LFO on it. The EQ actually has a small peak that moves back and forth through most of the sound spectrum to give it just a little bit of movement. It’s probably easiest to hear this subtle effect right after the “speech section” as the drums come back in.

There are two separate vocal tracks in the song. One is obviously the “speech” in the middle, which I passed through a few effects, an EQ, and a reverb to achieve a sound that resembles something coming out of a radio. The other vocal line is, of course, the vocoded vocals that repeat the line, “We are cyberpunks”. The vocoding was done in Reaktor. I was going to use the VirSyn’s Matrix vocoder, but I was having some issues with it crashing Sonar at the time.

So there you have it. The song will also be appearing on my upcoming album Inside The Beat, which is now about 2/3rds of the way done. I’m shooting for a late November or early December release, but don’t hold me to that ;^)

It's free, too! (as in beer)

2010-06-09T00:00:00-06:00

For my next album, I think I’m going to switch up the license I use a bit. Up until now I’ve used the Creative Commons Attribution-ShareAlike 3.0 license, which meant that anyone could distribute the music, make derivative works, and sell it, provided anything derived from it used the same or similar license. This, in effect, made it work a bit like a program that uses the GPL license, which was perfect for me since I’m a big supporter of the GPL and free culture in general.

But after giving it more consideration, I think I’m going to switch to licensing all future works under the Creative Commons Attribution-Noncommercial-ShareAlike 3.0 license instead. It’s the same as before, except there’s an additional clause in there that prohibits any commercial sale of the music.

I want my stuff to be free as in beer too, not just free as in freedom. This will start with Inside The Beat once it’s released.

Community Compilation CD Project

2010-05-19T00:00:00-06:00

A friend of mine here at work recently had the idea of putting together a Community Compilation CD made up of tracks composed by people who work within the tech industry. All of the tracks will be released digitally under the Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 license. There’s even going to be a booklet included with it!

Being in the tech world myself, I thought it was a great idea, and so I’m going to contribute a track to it. The problem is… what track do I submit to it? One idea is that I take a recent track I finished that won’t be appearing on “Inside The Beat” and submit that. Another idea - the one I’m leaning more toward - is doing a 4th remix of “I Love Penguins”. I’ve had some really cool ideas for how to expand on the original concept, like taking the voice samples and giving them even more impact.

So we’ll see! If you’re a developer or someone who works within the tech world, you should check out the project. I think it’s going to turn out really well!

More Album Plans

2010-05-14T00:00:00-06:00

Last night I mentioned some plans for the upcoming album to a friend of mine. The plan is that the album will reflect the beats and music that constantly play in my head, hence the title “Inside The Beat”. There’s a huge range of them that I hear, so to reflect this the first couple of songs on the album will be more upbeat and synthpop-ish. As the tracks progress they’ll use darker sounds, as well as more mechanical rhythms.

I think it’s going to end up working on more levels than I initially planned. He mentioned the idea that going from light to dark reflects life in general, as we move from an innocent childhood into adulthood. I also see it reflecting how my music is changing as time goes on.

That said, here’s a new work-in-progress clip that may or may not turn into something. There’s absolutely no mastering or production work done on this. It’s straight from my head to the sequencer and synths.

New Work-in-Progress - Stagnation

2010-04-12T00:00:00-06:00

You may have already noticed this, but I haven’t posted any new songs lately. The reason for this is that I want to direct more attention to finished albums, and so it doesn’t make sense to post songs individually. After all, if you’ve already heard the song (or even downloaded it), why grab the album? This means that the New Music Podcast will probably not see any new full songs. It will, however, start seeing clips of new songs. Which brings me to the real reason I’m writing this post…

This is a work-in-progress clip from an upcoming song called “Stagnation”, which will appear on Inside The Beat. The clip hasn’t been mixed or mastered yet, so no final EQing, multiband compression, or stuff like that. Also no work has been done to make sure things are balanced. But it should give you an idea of where Partition 36 is heading with this next album.

Stagnation WIP:

Album Update

2010-03-16T00:00:00-06:00

Well, since news at www.partition36.com has been fairly quiet lately, I thought I’d give everyone a quick update on how the new album, Inside The Beat, is coming.

Work on the album is continuing at a slow but steady pace, and it should be increasing soon. Three songs are definitely “done” and have been selected for inclusion in the album. Three others have also been finished, but they’ll likely not be included in the final album. Their style is just too different from what I’ve envisioned for this album, and I’m wanting this to be a much more solid release than anything I’ve done so far. So, I’ll instead release these three on a future EP or on singles instead. The ones that are making it on the album are much more beat-driven, a bit darker (such as the title song), and will all tie into computer-related issues. After all, without computers Partition 36 wouldn’t exist, right?

I’ll probably not be posting any more completed full songs from the album until it’s entirely finished. I will, however, be posting short clips from songs as they get completed. So keep checking the homepage and the official Facebook Page (you are a fan there, right?)!

Possible Future Live Show?

2010-02-17T00:00:00-07:00

A live performance of my music is something I’ve wanted to do for a long time, but haven’t because I always ran into some issue that put it on hold. Some of the issues were technical in nature, such as finding the right software so I could play both hardware and software synths live while still processing their sound in real time. Others were more of a question of what the performance would consist of to keep things interesting since I would be the only one on stage. But in the past few weeks I’ve stumbled, almost accidentally, onto quite a few solutions. In short, I think the first live performance for Partition 36 might be in the semi-near future.

What I’m looking at doing is performing a piece of each song live against a backing track, singing lyrics where appropriate and modifying parts of the song in real time. So a super simple example might be where I play the melody while the rest of the track plays off the computer. Meanwhile a sort of real-time video show will be going on behind me on a screen (not sure if I can do much control of this in real time yet or not, but I know for sure it reacts to the music).

I think what I’ll end up doing first is a test performance of three to four songs in my basement among friends, that way I can see what works and what doesn’t (I guess I’m an engineer at heart). There’s still some hardware I need to purchase before I can do this though, and I’ll also need to prepare the songs for live use, but I hope to have this test performance done by the start of summer.

In the meantime, you can become a fan on Facebook for updates.

new Synths, Small Album Update

2010-01-13T00:00:00-07:00

No, I haven’t bought any more :^)

I think that whenever you buy a new synthesizer, there’s a period of time where it’s almost unusable. It’s not so much an inability produce sounds, but rather a general dumbfoundness relating to how it should be used. This is how it was with my Nord Rack 2X that I bought over the summer. Was it better for bass sounds, pads, or leads? What are the harmonics like on the oscillators? Do I use it in the background or foreground? I essentially had a very expensive red box sitting there and no idea what I was going to use it for. Sure, I had plans for how to use it before I bought it, but trying one out in a store and actually using it in your own studio are two very different things.

A little over month ago I finally had the “ah ha!” moments I had been waiting for. The first was when I discovered exactly what the Modulation Envelope section does, while the second was finding out just how flexible its two oscillators are. Will it replace my Poly Evolver or Prophet V software as my main workhorse synth? No, probably not. But it is definitely turning into a centerpiece in my music and will show up quite a bit in the next album.

Meanwhile, I found an excellent free software synthesizer the other night called Synth1, by Ichiro Toda. It claims to be modeled after the Nord Lead 2, which I can understand as far as its structure goes, but I think it ends up sounding different. For being freeware, it sounds out of this world… as long as you don’t use the built-in presets. This will probably also start showing up on some tracks in my next album.

Speaking of which, progress on Inside The Beat is coming along great. I’ve finished three songs and have been working on a fourth, although I’m thinking one of these may not make it into the final album for quality reasons. If it does get cut, it will probably show up on the EP that will follow the album, or as a B-side on a single.

Also, if you haven’t seen them yet, I’ve put up promotional images that were taken for Inside The Beat up on my website and on the Facebook Page. A little outside of my normal style of dress, but it was LOADS of fun shooting these and I really think it reflects the view I have for the album. A very special thanks to my friend Curt Richter for shooting these!

Sequencers and LFOs

2009-12-05T00:00:00-07:00

Wow, it’s been a while since I’ve written in this… gotta update it more often :^)

Lately I’ve been playing around with my Poly Evolver quite a bit to try and get better at programming it. I bought the synth about a year ago, but I’m still just scratching the surface of all the things that you can do with it. So, aside from creating some drum sounds, I spent some tonight time messing around with the sequencer. It’s a 16-step analog-style step sequencer, which means that you have up to 16 steps per cycle in which it can control other things.

Normally I approach sequencers as simply a device that one can use to control pitch over time. The most simple example would be to generate a simple arpeggio, while more complex examples can sound more like actual elements of a song. Both of these - which are just presets in my Poly Evolver - are programmed such that the sequencer controls the pitch of the oscillators. This means that, when the sequence played back, each step acts as if a new key is pressed at a certain pitch, thereby generating a melodic sound. In some ways it’s like the tape roll on an old player piano, except that it can only have up to sixteen holes.

However, step sequencers can be used to control a lot more than just notes. The experiment I did tonight had the sequencer controlling the low pass filter instead of the oscillators. This meant that sound and pitch did not get generated when I played back the sequence. Instead, sound got generated the traditional way, with me holding down one or more keys. As I held them down, the sequencer acted as a retrigger, creating a straight rhythm. At the same time, I also had the sequencer modulating the lowpass filter’s cutoff frequency and (very slightly) the resonance, so you end up with this pulsing pattern that alternates between partially "open" sounding and partially "closed. To add a bit of flavor to the sound, I then finished off with some additional modulation from the LFOs.

Now for those who don’t know what an LFO is, the term stands for "Low Frequency Oscillator". Just like the name implies, it generates wave forms at low frequencies. However, you never actually hear these waves directly. Instead they act as control signals that can be mapped to other areas of the synthesizer. A good example would be mapping an LFO that generates a sine wave to an oscillator’s pitch. As the LFO’s wave rises and falls, the pitch of the sound will "follow" this, creating a sort of pseudo police siren sound. Also, "low" is somewhat of a relative term. On the Poly Evolver, an LFO can be set so slow that it generates a single cycle only every 30 seconds, or as fast as 261 cycles per second (hertz), which is almost the equivalent of middle C on a piano.

Anyways, for the test sound I made tonight, I had two LFOs controlling the volume of oscillators 3 and 4. The LFOs were set to a pretty slow square wave, which made one complete cycle about every six seconds. This basically gave the oscillators an On/Off effect. So, the final sound ended up being a pulsing sound, alternating between open and closed, with its timbre intermittently changing. You can hear the final result here, where I first play a single note/pitch, followed by a simple power chord.

This isn’t the first time I’ve messed around with step sequencers to control a filter, but it is the first time I’ve actually figured out how to program an analog-style one. Since the destination possibilities for each step are huge, I should be able to create some pretty interesting sounds using it. What’s more is that it can also be used to control other synthesizers… or at least I think.

New Netbook

2009-11-01T00:00:00-06:00

This morning I picked up a new netbook, a Toshiba Mini nb200. Spec wise it sports 1GB of RAM, a 160GB hard drive, an Intel GMA950 for graphics, and an Intel Atom N280 processor. This CPU is the same speed as an N270 (1.6GHz), but has a faster front-side bus at 667MHz. The screen is a bit bigger than the one on my 900A model Eee PC, though it still has the same resolution of 1024x600. Also, unlike my Eee PC, this one has a built-in web cam. The laptop itself is extremely slim, more so than my Eee PC even, and the battery is nice and fairly low-profile.

The keyboard layout is pretty interesting, and somewhat reminiscent of my old Toshiba Satellite. There are dedicated Page Up and Page Down keys on the right hand side, next to the arrow keys, and even dedicated Home and End keys at the very top. The tilde key, which is important in Linux when you spend as much time at the command line as I do, is in between the space bar and alt key. Some may find this annoying, but I actually like it. Otherwise, the keyboard is extremely comfortable and I can type full speed on it. As for the mouse pad, it feels smooth, responsive, and I can slide my finger on the side for tracking. Multi-touch doesn’t work unfortunately, unlike my Eee PC, but this isn’t that big of a deal. Also, the buttons are actually quiet! This has always been one of my favorite things about Toshiba laptops, the mouse pad buttons don’t emit a loud click.

So yeah, the system of course came with Windows on it (XP Home specifically), so the very first thing I needed to do when I got home was remove this and put Linux on it. Earlier this week I had installed release candidate for Ubuntu 9.10 onto my Alienware laptop from a USB drive. Now, the installer I originally loaded up had some issues with the GUI installer (probably because I followed instructions for a previous Ubuntu release), and so I was limited to the alternate installer instead. Not a problem, just a minor annoyance, so since I still had the drive set up for that I figured I’d give it a try. Unfortunately, the installer decided that it couldn’t find the install ISO. Not wanting to recreate the USB installer again, I decided to stick with Slackware instead.

As usual I booted into the Slackware 13 installer using a USB stick, and then proceeded to install using an pre-mounted directory (through NFS this time) as the source. The install went great, but when I was most of the way through I realized that I had a bad symlink on my server and so I was actually installing Slackware 12.2. Oops. So, I fixed the symlink, rebooted the installer, and then redid everything. The install goes so quick anymore that it wasn’t really a problem.

During the final config I was of course given the chance to name my laptop. Keeping with my current scheme, female characters from anime or kaiju films, I decided to name it Kusanagi :^)

After rebooting, I did my usual setup:

Installed the newer 2.6.30.5 kernel from /testing
Installed a newer Intel driver (though I later compiled 2.9.0 from source anyways)
Installed some packages from /extra that I like
Installed various Slackbuilds packages that I had built over time and keep on my server
Setup my user account
Copied over data from my Rei (my other netbook)
Set up Xfce just how I like it
Set up my most often-used programs: Firefox, Pidgin, and the terminal

So how does everything work? So far, I can confirm that the web cam, SD card slot, VGA port, and both the wired and wireless connections work out-of-the-box with Slackware 13. Unfortunately the sound chip in it isn’t 100% supported by Alsa, and at first I had absolutely no sound. After I tinkered a bit (and installed a newer alsa-driver package from source), I at least got the headphone jack working. I can’t say if capture is working yet or not as I haven’t tried, but at least everything else (minus the speakers) is working perfectly.

As far as performance goes, it’s noticeably more responsive than my Eee PC. My guess is that it’s because it has a real hard drive in it rather than an earlier generation of SSD. I’m sure the faster FSB also helps, though ;^)

Now for some nice, large images. Click the links to open them.

How the thing looks when it’s closed
A comparison of the thickness to a Wiimote
Left side
Front
Size comparison between my humongous old Alienware laptop, the new one, and my Eee PC. As you can see, it’s slightly wider than the Eee PC.
Open laptop on its first boot.

Useful Windows Shortcuts

2009-09-20T00:00:00-06:00

I don’t usually use Windows these days, but when I do, I generally find it a lot easier (and faster) to navigate with the keyboard instead of the mouse. So I thought I’d post a few keyboard shortcuts that can make peoples’ lives a bit easier. These aren’t the usual Ctrl+C (Copy) and Ctrl+V (Paste) shortcuts, but rather some slightly more obscure ones that you may not know about.

Also, just to make sure people know… hitting just the Windows key will show the Start menu, while hitting the Menu key (which usually looks like this) acts like a right-click,

Windows Key + L: If you have the normal login screen enabled in XP, this will lock your computers screen in the exact same way that hitting Ctrl+Alt+Del and then clicking “Lock Screen” will. If you have the Welcome screen, it’ll take you back to that. This works in Vista as well.

Windows Key + D: Instantly minimizes all windows to show your desktop. Hitting it again immediately afterward will return all the windows to their previous state. In all honesty, I find this useless, not useful.

Windows Key + Pause: Opens up the System window, which is the same one as when you go to Control Panel->System.

Windows Key + Tab: It’s sort of like Alt+Tab, the window switcher key combination, but this one only switches the active item on the task bar without bringing the window to the foreground.

Windows Key + E: Opens up Windows Explorer.

Windows Key + B: Put the focus on the first item in the notification area (that place right next to your clock with the icons). You can then use the arrow keys to highlight different icons, the enter key as a left click, and the Menu key as a right click.

Windows Key + F: Open up the search window, which I also find useless because I actually take time to organize my files…

Windows Key + U: Open up the Utility Manager, where you can enable/disable things like the on-screen narrator and stuff.

Windows Key + R: Open up the Run dialog (the same as going Start->Run).

There are also ways to assign custom keyboard shortcuts to programs, so that typing them can instantly bring them up. For example, I have Ctrl+Alt+T bring up the Command Prompt, while Ctrl+Alt+E brings up Emacs. However, setting these things in XP seems a bit buggy at times, and I can’t always get them to work.

Midsized Clients

2009-09-14T00:00:00-06:00

Computers are sometimes divided up into two groups, “fat” clients and “thin” clients. Strictly speaking, a fat client is something like your average desktop or laptop computer, which has everything it needs to run and operate right there inside of it. Meanwhile, thin clients have practically nothing on them, and rely on other computers, usually servers, to provide just about everything except the most basic local hardware and processing power. If you think back to the old days of computing, where you would see text terminals hooked up to a mainframe, and all the computing was being done on that mainframe, those are essentially early thin clients. Today thin clients are mostly graphical, but the concept is the same. The desktop and programs are handled by the server, and the local machine just acts as a display unit.

I think we’re starting to see, or at least have the potential to see, something in-between these two ideas. Compared to normal laptops, nearly all netbooks are very underpowered, and the subset of them that use solid-state drives instead of the normal hard drive even lack a lot of disk space. But I believe that if we change the way we interact with the file system slightly, we can change them into what I call “midsized clients”.

The idea for a midsized client came to me after I noticed that much of the data I use with my Eee PC doesn’t reside on the local machine, but rather on the file server I have set up here at home. Now, there are a lot of ways to access a remote file system locally these days, such as Samba and NFS, but I personally like to use SSHFS. But what if the data wasn’t the only thing that resided on the server? What if some, but not all, of the programs also resided on there as well? Specifically some of the larger software packages, like OpenOffice.

In fact, that’s how I use OpenOffice on my Eee PC.

Naturally, this also seems a lot like accessing applications through a web browser. Google Docs is a good example here since we’re already talking about office programs. But, in my opinion at least, I don’t believe that accessing a web app through an underpowered, low-space machine qualifies it as a midsized client since it doesn’t really fit into the classical examples of thin vs fat clients. To be a true midsized client, the program should not have to use any sort of special interface like a browser to be started. Instead, the user should be able to start it through normal means, such as a menu or on the command line.

Thinking like this opens up some interesting possibilities. Some locations may provide only certain programs. If I only need program X at work, I can leave it on a server at work, mount the server’s directory locally on my netbook, and then access it transparently. Data can then be stored in a folder on the same server that I use for work-related documents. If I really need to access it later, I can always VPN back into work. Meanwhile, my other small applications and my desktop remain local, so I can still use my netbook without a network connection if I need to.

It’s that last point that I think is the main difference between the traditional fat/thin client split, and this midsized client idea.

Of course, none of this is formally implemented yet, and this midsized client concept is working entirely off of very traditional, strict definitions of fat and thin clients. What’s more is that, as with many technologies today, there are some blurred lines of what actually makes something a thin, fat, or midsized client. In fact, this whole idea of a midsized client may not be new at all; I didn’t exactly Google any of this before I typed this up.

In the end, though, I find it a pretty interesting idea, and I’d be excited to see where it can go. I think I’m going to continue playing with it here at home since it seems pretty easy to get things working. I had OpenOffice working within 10 minutes of grabbing a binary package, mostly because of the simplicity of Slackware’s packages. All that was needed was some slight hacking of install script to keep from doing things locally. I also have a fast enough connection here at home that accessing it off-site through SSH is mostly painless.

Let’s see how far I can take this…

Thoughts On The Recent EP

2009-07-21T00:00:00-06:00

It suddenly occurred to me that even though I released my first EP, Opcode-1, earlier this month, I still hadn’t given my thoughts on it! Well, overall I would say that I’m mostly pleased with it. My goal with the EPs is to release music more often by compiling the “b-sides”, stuff that may not end up on normal albums, and remastered or rearranged tracks from long ago.

Opcode-1 accomplished this quite well. The First and last tracks are by far my favorites, while the remasters are probably my least. I think this is just because I’ve heard the remastered tracks so many times (I originally wrote those in high school or right after it). As far as the sound quality goes, I’m actually very happy with it, but for some very specific reasons. At first I was worried that I had left off too much bass on the songs, but the more I listened to them the more I realized that I had gotten them quite balanced for what they are. What’s more is that, although you won’t hear it on the released tracks, working with these actually taught me quite a bit more about dynamic compression. The new stuff I’ve learned will make its appearance on the next Partition 36 release.

My only real gripe with Opcode-1 is on a higher level. The style of songs on it aren’t what I originally imagined Partition 36 releasing. That doesn’t necessarily mean it’s a bad release, not by any means. It’s just not what I had planned to write. So it’s more of a personal letdown than anything :^)

Partition 36 is more than just my “one-man band”. It’s my electronic music project. The whole goal of it was to release music that was a little bit darker. I was tired of trying to write generic epic trance or video game music, and wanted to work with music that was influenced by the EBM and Futurepop scenes. Making this switch is requiring me to learn a lot of new things, and even relearn some old techniques.

Thoughts On A New Remaster Release

2009-06-16T00:00:00-06:00

Many, many years ago (well ok, the summer of 2002), I wrote an album called New, which was before I called myself Partition 36. There were a couple of memorable songs on it, but it was the last track that recently caught my attention. It was called Boku no Kanojo ni Matteru, roughly translated as “Waiting for my girlfriend”, which is exactly what I was doing at the time when I wrote it. After listening to it a few times, I decided that I wanted to remaster it and present it in the sound quality that it deserves.

However, rather than doing a straight remaster of the original, I thought it might be cool to instead change out some of the instruments. The original song was written entirely using the General MIDI patch set on my JV-1010 synthesizer, and the more I listened to the slap bass/fingered bass combination, the less I liked it. So, I started a new project, opened up the old file in a second window, and started copying tracks over.

The first track I copied was the bass track. The sound of the slap bass in the original track was a bit too twingy sounding, while the fingered bass just sounded about right, so I wanted to get a good compromise between them. To do this I replaced them with a customized Prophet V patch that had a more rubbery/synthy pick sound on it but still provided solid bass frequencies. The downside was that this patch tends to phase in and out of being overpowering and just right in the sub-bass frequencies. Some quick processing and adjustment of the patch luckily fixed this.

The drum track was next, which at first proved to be a bit difficult to tackle. In the original, the kick drum was fairly solid in the lower frequencies, but was also very snappy. I wanted to find a sound that would instead provide a decent amount of punch, not as much snap, and would also play nicely with the bass track. As usual, I went through many of the preset drum kits in Battery 2 trying to find something that I could use as a starting point, but nothing really fit. Then I got the wild idea of reusing the drum kit from another song I wrote recently, Space Train. As luck would have it, the kit fit great inside the mix, giving it more of a synth rock feel that the song was supposed to have in the first place. The only thing was that this kit was made from scratch rather than basing it on a preset one, and Space Train never used an open high-hat sound. Once this was replaced, the main work for the drum track was pretty much complete and sounded great.

There’s a certain type of sound used quite a bit in electronic music called a “pad”. Just like the name says, it provides “padding” for the rest of the mix by filling it out harmonically. The original track used the Warm Pad sound off my JV-1010, which has always been a favorite of mine. However, it sounded just a bit too dense for this new mix. This was easily solved by finding a similar sound in my Jupiter-8V software.

Next was the Koto sound used for the bulk of the melody. At first I tried replacing this with custom Prophet V and Moog Modular V patchs, but this proved disastrous. Instead I stuck with the original track, re-recorded it, and just passed it through a better reverb. I then layered a soft pingy sound from my Pentagon software synth with it in certain places to give it something extra.

The melody is split between the Koto sound and another bell-like sound called “Fantasia JV”. This has been one of my favorite synth patches ever, and so I knew in an instant that I would not be replacing it. So I transferred it over and just re-recorded it.

By this time the only thing left to move over was the guitar track. I have never, ever had good luck in finding a decent guitar sound on a synthesizer. Maybe I’m just not sequencing the track right, but I’ve never been able to get something that sounds close to a real guitar. For this song, I first played with the idea of changing it to a piano or electric piano solo, but it didn’t sound right. So I copied the old track over, switched patches on my JV-1010 to the “Big Hair Lead” sound, re-recorded it, and then passed it through a ton of plugins. In the end it sounds better than the original, but is still not very pretty. But I wanted the solo in there and it’s good enough for me.

In the end, I think the remaster/reworking turned out pretty good. The guitar solo is still the low-point of it and I admit it sounds like crap, but I like the new sound a lot better. I decided to change the track name slightly to Ore no Kanojo ni Matteru, which still means the same thing but has a slightly different nuance.

You can download the new version here, and you can also listen to the original version here. Enjoy :^)

Linux Systems

2009-05-29T00:00:00-06:00

Truth be told, I am a die hard Slackware fanatic. Whenever I get a new PC, it’s the first OS I reach for, and it’s the one that I compare others to. Slackware has been what I’ve used on my server here at home since I first bought it. When I first learned Linux, I learned on a derivitive of Slackware, TopologiLinux (a way old version of it, anyways), and quickly switched over to real Slackware (which was at 9.x or right at 10.0 at time time, I think). This combined with my love for minimalist things is probably why I stick with it.

However, for all the praise I can give the distro, there are still two sticky points for me: software selection and lack of Gnome libraries. When it comes to software, there’s a huge amount of stuff available to Slackware in the form of “Slackbuilds”. These are special bundles of scripts and source tarballs that produce binary packages for you (you did know Slackware has a package manager, right?). I keep a collection of them just in case I ever need to reinstall something. But unfortunately, a few packages I’ve needed, either for work or personal use, are either very out-of-date, or not available. A few of these I’ve compiled by hand - a skill I learned like day three of my Linux adventure; it’s not that hard, people. Whenever possible I keep these hand-compiled ones inside my home directory in ~/bin. Unfortunately, there have been a few apps like Ekiga that have forced me to look elsewhere. While I could have gone the hand-compiled route with them, the need for immediate availability was too great.

Gnome support is related to this. Overall I rather like the Gnome desktop. It’s more comfortable than KDE 3.x, leaps and bounds better than KDE 4.x, and is pretty fast and stable. Still, it’s not my prefered desktop, Fluxbox or Xfce are. However, many apps out there rely on Gnome-centric libraries. When I decided to try installing Gwibber from source, I was blocked because I couldn’t get GConf2 to install properly, and this was a required package. Other packages have given me the same problem.

So what do I run on my main laptop, which is where I need immediate package availability and a sightly wider range of apps? Ubuntu Linux, which is my 2nd choice in a distro. In actuality, I switch between Ubuntu and Slackware on the laptop a few times a year, usually when a new version comes out. But until GTK integrates some of the Gnome-centric libraries (which I think I read was a plan somewhere), I’m likely going to leave Ubuntu on it for a while and simply use Slack inside a virtual machine.

For a Slack addict such as myself, Ubuntu actually feels restrictive. Heck, Debian felt restrictive at times. But I don’t think this semi-permanent switch to Ubuntu on my main laptop is going to be that hard.

Of the four systems I use regularly, three are dedicated Linux systems (the fourth runs XP, and is the computer I write my music on). One, my main laptop (Emiko), has Ubuntu installed on it. My server, Nanako, and my Eee PC, Rei, both run Slackware. The server never leaves home and is mostly a file server, plus some extra things, and so it’s already set. Rei is really my play computer, and the one I use mostly at home or on the road. Of the sticky points I have with Slackware, they really only affect me at work. Since Emiko is the only real heavy-lifter for work, it’s OK if that one’s Ubuntu. I actually do very little on it here at home save for some web surfing, IM chatting, and light programming.

The point I’m getting at is that, for the time being, I’ve reached a very good balance with my systems. I have XP here at home to do my music, Ubuntu on Emiko for work, XP on a VM for those few times I need it, and Slackware on my others for home stuff and play.

Pictures Are Worth A Thousand Words

2009-05-16T00:00:00-06:00

In the front of almost any traditional Japanese dojo is the kamiza, or “top seat”, which is where a kamidama is usually found. In Aikido, this is also where the picture of O Sensei is hung (though in some schools, the kanji for Ki is found instead).

From what I’ve experienced so far, the most common picture of O Sensei has been this one. It’s a very noble picture of him, I think, and I like it very well. During the beginning of class, while we’re all lined up in seiza, I’ll sometimes glance up at his picture. Usually I simply tell myself, ”I’m glad he was able to teach his art to so many people,” though sometimes my imagination start to wander.

Know the cliche in horror fiction where the ominous picture on the wall has eyes that follow people? Sometimes I would get that same feeling when looking at this picture, like he was looking right back at me. His stern look felt, well, very stern. It wasn’t scary or anything, but it did feel like I was obligated to perform to my very best. This is all in my head, yes, but bear with me.

When I switched to my current dojo, the first night of class I noticed we had a different of O Sensei hanging up on the wall. I still had that feeling that he was looking at me, but it was a lot more easy going and friendly. During the middle of class, if the previous technique didn’t go well for me, the other picture would have given me the feeling of “I failed.” But this picture almost feels like both he and I are laughing it off internally, like “learn from the mistakes this time and just try again next time.”

Again, this is all in my head and personal experience. I don’t think either picture is more correct, or that one should be hanging and not the other. But it has proved to me that pictures really are worth a thousand words.

New Blog Layout

2009-05-08T00:00:00-06:00

If you haven’t figured it out already (and chances are good you have), I dramatically changed the layout of my blog. The change came about for a few reasons, most seated in my innate fascination with minimalist things, and more generally in technology.

WordPress was working fine for me, but I had a few issues with it. Most of the features I didn’t use, and some of the features I wanted to turn off but couldn’t figure out how or couldn’t to the degree I wanted to (trackbacks were one such issue, as well as people having logins, which made no sense). Then there was the fact that the theme I had been using was actually a bandage I hacked up one day after ruining my previous theme. Managing it was difficult, and I lost a few features I actually wanted (pages and widgets).

Giving it deeper thought, I started to wonder what exactly I was after with a blog. It’s not so much a record of my life (“Today I went bike riding at…”) as much as a set of thoughts I have on pretty much anything. Records are better left to the microblogging and tumblelog services, which I think can capture “The Moment” better, or at least in a more interesting way. So if it’s a collection of thoughts, wouldn’t simple pages work just as well?

I was then introduced to a cool piece of free software called Jekyll. The software is similar to my “WST” program that I started writing a few years ago in that it performs text transclusion, but it does a much better and thorough job. Basically I can now keep my blog stored as a set of mostly plain text files inside a directory tree, which I then manage with Git. Writing a post means I write a new file using whatever editor I want to (I wrote this one in good ol’ Emacs), “recompile” the blog using Jekyll, and then use rsync to upload it to my webspace.

I don’t think it can get much more minimalist than that, unless I abandon all design and instead only post plain text files.

I’m sure there are going to be a few kinks to work out still, and I’d like to expand things a bit, but I think this will work out better in the long run. Now, I wonder if cron could be integrated into this mix…

Some thoughts on negative space, ma, maai, and mushin

2009-04-08T00:00:00-06:00

Growing up, I remember hearing a suggestion for how to listen to music (and also heard it repeated on an early episode of The Simpsons). Rather than listen to the notes, a person should listen to all the notes that aren’t being played. This is very similar to the idea of negative space, where the best example is probably “Do you see a vase or two faces?” While a soloist creates a thread of feeling with his instrument, there are “faces” to be seen outside of his notes which can be just as inspiring.

A lot of Japanese music, especially the traditional music, incorporates a similar concept. In a hypothetical composition, you might hear a few plucks of a shamisen, followed by a bit of silence, followed by more shamisen. But is that silence really silent? Surely there will be a bit of “tail” sound from the vibrating strings. Beyond that, there is the sound of audience (if one is present), the environment, and maybe even the performer breathing. This is called ma in Japanese music, and though there is no direct translation into English, it’s still best described as “negative space” and is considered just as important (perhaps more so) than the notes themselves. In fact, John Cage’s infamous composition 4'33" is entirely composed of ma.

In Aikido, and pretty much all Japanese martial arts, the concept of maai (“mah-aye”) is very important and represents the negative space between a person and their partner. Understanding how far you are from each other, how long it will take to reach a certain position, how long it takes to meet each other there, and the rhythm required are all extended aspects of a technique governed by maai. Again, this negative space is just as important as how to move your arms and legs, and is directly tied to the physical parts of the techniques.

At the beginning and end of class, we all line up, sitting in seiza, and have some silence. But I realized tonight that it’s not necessarily silence. As we sit there resting and reflecting, we are in fact surrounded by ma! But rather than connecting to the smallest parts of the rooms (us and our voices), we are instead mentally connecting to everything else around us. It sort of gives us a context.

Ma, and by extension maai, are pretty important. 4’33” would not have context without it. Without ma, notes would be constant noises without break, and so music would not really exist. Aikido techniques would not work. There would be no way to examine yourself since you wouldn’t have any separation from the room you’re sitting in or the person next to you.

But maybe learning how to manipulate ma is an important skill. Another term, mushin, roughly means “no-mindedness”. In the context of Aikido, it means that a practitioner is able to react to their partner’s movements and perform techniques automatically, without any hesitation or thinking about what they should do. To do so, the person needs to have no ego, no anger, no fear, no desire, and no thoughts, simply intuitive feeling as they move. It takes a really long time to develop, repeating their movements over and over until they feel as natural as walking. But as far as I can figure, on the mental side of things, doing no-mindedness means letting your mind become part of the ma. You don’t think (and therefore separate yourself), you just be.

StumpWM on my Eee PC

2009-03-31T00:00:00-06:00

I have something announce that may be shocking to some people: I can be very geeky when it comes to my computer. Ok, so obviously that’s a poor attempt at humor, but it is true.

I usually run Gnome on my main laptop for various reasons. Besides appreciating the look, feel, and behavior of the GTK+ kit over Qt (a personal preference), I’ve grown fond of Gnome’s usability. Beyond that, it’s often times helpful to have the notification bar available since I usually have around 15 open over six virtual desktops (about half to two-thirds of which are XTerm windows). Besides, I sometimes have to show various apps and information to people who are used to seeing an environment that looks at least somewhat familiar, and so Gnome tends to work out nicely.

But, being a huge fan and constant user of GNU Emacs, my fingers have built up quite a bit of muscle memory for things like “C-x 5 2”, “M-q”, or “C-space C-q C-y” (“Ctrl-x, 5, 2”, “Alt+q”, and “a way to cut and paste” for the Windows people out there). So much that I often find myself typing C-x C-s (save file) or C-_ (undo) inside of just about every other editor and app without even thinking. So having a desktop environment that shares these keybindings would be very helpful.

A few years ago I came across such an environment called ratpoison, which provided an extremely minimal GUI environment. How minimal? Windows did not overlap, had no boarders, had no title bars, and the whole thing was set up to work entirely from keyboard commands. A cool concept, but was it usable? To find out I took it for a test drive for an entire day, using nothing but ratpoison. It turned out that it was quite usable, but at the time I was not an Emacs user and so I had a lot of problems getting used to the keyboard shortcuts (I think… all that was a while ago).

Another window manager, StumpWM, is basically the same thing as ratpoison (written by the same author, too). However, it’s written using my favorite programming language, Common Lisp, has more features, and is a bit more customizable. I actually did the same sort of experiment with it as I did with ratpoison, eventually arriving at the conclusion that StumpWM was absolutely awesome and worthy of being installed with all my other regular apps.

So as I said, I’m a computer geek, a fact that purchasing an Eee PC did not help. Since buying my second one, I’ve successfully removed the modified Xandros Linux that came with it, installed Slackware 12.1, upgraded to Slackware 12.2, installed SBCL 1.0.24 from source¹, and as of tonight, installed StumpWM from source.

StumpWM runs incredibly smooth on it, and my other apps seem to benefit from the extra resources that usually aren’t available to them. The biggest drawback so far is that my Eee’s screen size is 1024x600, (1022x564 in StumpWM once you account for the infobar I have up top). This means that tiling is somewhat limited unless I connect to a larger monitor, preferably one that is not widescreen. Splitting the screen horizontally helps some, but not with browsing.

I’m going to keep this as my default GUI for a while on my Eee PC to see how it works out. Since my Eee is mostly used for surfing the web, talking to people, playing music, and occasionally an SSH session, I should have no problems whatsoever.

1 - I didn't actually compile SBCL on my Eee. I did this a few versions ago, but to upgrade to 1.0.24 I just copied binaries over from a different laptop since I was short on time.

Uber Pork Chops

2009-03-28T00:00:00-06:00

I’ve been on a bit of a “let’s throw this stuff on some food and see if it tastes good” streak lately. Luckily, the things I’ve tried have actually tasted pretty good, but the pork chops I made earlier tonight took the cake.

Lacking any creative ability to name things, I call them my Uber Pork Chops. Flavor-wise they taste like non-greasy, slightly spicy sage breakfast sausages from Jimmy Dean. I browned mine a bit too much tonight, but I think that was because I had to keep them on the fire while I waited for some other things to finish. As usual, I didn’t measure anything, so you’ll just have to play with the amounts.

Ingredients:

Boneless pork chops (the thin kind)
Powdered sage
Ground cumin
Paprika
Garlic powder (not garlic salt!)
Unsalted butter (you can use salted if you'd like, but I don't like excess amounts of salt)
A sprig of rosemary
Ground black pepper
Peppercorns
Salt

Steps:

Start by trimming all the fat off the pork chops, then laying them out on a cutting board.
Sprinkle the sage, salt, ground black pepper, and cumin on one side of a chop, then pat it all in. I used sage and pepper the most, and small amounts of salt and cumin. Then take the back of a large knife and hammer the heck out of the pork chops to tenderize them.
Flip the pork chops and do the exact same things on the other side.
Put a chunk of unsalted butter in a frying pan and warm it up over medium heat (our stove goes from 1 to 10; I used 5). Once the butter is melted and starting to foam, sprinkle a good amount of garlic powder in it and stir it around. Also sprinkle a small amount of paprika.
Place the pork chops in the pan. If desired, sprinkle a very small amount of paprika and garlic on top of the chops.
Take come peppercorns and crush them coarsely. Sprinkle them on top of the pork chops, then cover the pan.
After about 5 to 8 minutes, flip the pork chops. They should have started to brown. Flip some of the butter onto the other side, and add some more paprika (if desired) and crushed peppercorns. Re-cover the pan.
Cook another 5 to 8 minutes, then flip the pork chops again. Re-cover the pan.
Continue to cook another 10 or so minutes, or until they start to become well-done.
About three minutes before you take them off the fire, sprinkle the rosemary twigs on top of them (tear them apart first). The butter is likely all cooked away now, so add just a small amount (about 1.5 tsp.) of water to the pan. Cover. The water will soon boil away.
Take off the fire and serve.

I had mine with a side of broccoli and cheese, a baked potato, and a honey biscuit, all of which went pretty well.

Likes and Dislikes

2009-02-22T00:00:00-07:00

I’ve always found it amusing just how varied my likes and dislikes are. Like with music, I really enjoy listening to pieces by Franz Schubert, J.S. Bach, and Frédéric Chopin. In fact, one of my favorite pieces ever is Nocturne in E-Flat Major for Piano. Music like that speaks to me much in the same way stuff from Boards of Canada or Kraftwerk does. Yet at the same time, I find great interest in bands such as Front 242, Psy’Aviah, VNV Nation, and Nitzer Ebb:

What’s more is I love hard rock and heavy metal. Iron Maiden has been tied for my favorite band since I first heard them, and I actually like listening to the occasional Slayer as well. Yet on the other side of the spectrum, I like bluegrass music, country, and especially C.W. McCall. So go figure.

It’s not just music, though. I could be up in the mountains fishing or snowboarding one day, then the next I’ll be really into playing video games or programming on my computer. My top three favorite movies are RoboCop, Forbidden Planet, and then the odd one out, ). I love goth fashion and culture, yet I also like the “outdoorsman” look and culture.

Yeah, I amuse myself quite easily when I think about things like this.

Road trips. I swear I'll take more this year.

2009-02-19T00:00:00-07:00

I know I’ve been saying this for a while, but I really want to take more road trips. Weekend getaways where I go off to a nearby state for some sightseeing with a friend. Or maybe not even to a different state, but rather just up camping or fishing for the weekend. Anything as long as it’s longer than a single day.

Trips like this give me a chance to get away from home, work, and school, which all start feeling very cramped after a while. They don’t have to be big and extravagant, since I think that actually starts to take the fun away from them. Instead, I like the trips to be on the cheaper side. Just me, a friend (I get lonely easily, I hate to admit), my camera, a car, and a cheap hotel. After all, that allows me to take more of them.

Last year I definitely increased the number of trips I took. There was my trip to Santa Barbara (which was actually for school, but it was still a chance to get away from things), my trip to San Francisco, and the mini road trip that was encompassed in that to Napa Valley. But this year I’d like to see if I can fit in a few more. There’s already the snowmobiling trip coming up in about a week. Beyond that, I’d like to go back up to Cabela’s in Nebraska before fishing season really kicks off. Then there’s a few places in New Mexico I’ve been wanting to see, like Albuquerque, Santa Fe, Petroglyph National Monument, and the Capulin Volcano National Monument. Near home, I’d like to go to Durango and the Four Corners, Spinney Reservoir, and Blue Mesa Reservoir.

I guess we’ll see how it all turns out. I have a generally good feeling, though.

No More DDR

2009-02-15T00:00:00-07:00

Last year when I started to work out in order to lose weight, I chose a few different workouts to do. Since it was the summer, I of course chose bike riding (a favorite activity of mine), but also chose to do swimming, and to start practicing at DDR again. When I started I was very out of shape, and so completing 30 minutes of DDR was actually fairly tough. But by November I was back to my old self and completing difficult songs with ease. Then I got kind of lax in December during the break period, followed by various stresses and schedule oddities in January, and so everything sort of went to hell in a hand basket. Basically, I gained a ton of weight back very quickly.

Naturally I’d like to try to fix that error and lose it again, and so this past week I started doing DDR again. But, it’s starting to look like I can’t do it anymore because of the way it’s effecting me. After playing my left foot gets funny where each step sends a tingling sensation up between my big toe and the one next to it. This keeps going up over the the top my foot. I also occasionally (luckily not too often) get pains in my heel, a possible sign of plantar fasciitis. I’ve tried wearing shoes during gameplay to help, but this actually makes my feet hurt worse, and so I have to take them off while playing.

I don’t want to ruin my feet, so I need to find something else to do during the winter months. Snowboarding is fun, but I don’t always make it up there. Instead I’m thinking of doing winter bike riding (I’ve been wanting to do this for a while, actually), and going swimming at the rec center. I’ll also probably start doing push-ups and sit-ups before bed like I was doing.

But yeah, I think my DDR days are over. I don’t want to ruin my legs or feet this early in life, even if it does give me incredibly strong calves.

A Goal: Aiming for Better Sound Quality

2009-02-11T00:00:00-07:00

With my desktop computer still dead, I haven’t been able to work on music now for about a month and a half. On the one hand it’s driving me nuts, to the point where I’m starting to think that I’m writing music in my sleep or during a daydream. But on the other hand, I’m using the time to do some thinking and light research regarding sound quality. Specifically, handling dynamics (volume).

The Loudness War. I’m pretty sure I’ve mentioned it before here on my blog. The basic idea is that recording’s dynamic range (maximum and minimum amplitude levels, aka volume) is compressed so that everything is about the same volume. In an uncompressed recording, the difference in volume for individual sounds is what makes some things sound punchy, other things sound delicate, and allows build-ups to actually feel emotional and big. However, people who master many popular records these days tend to compress this dynamic range down to make the song seem louder. At least at first it certainly does sound like that, but in reality, you the listener end up losing many of the nuances the sound contains. Drums that would have been punchy and prominent are now just another instrument lost in the sea of sound. The Wikipedia article in the link above gives a much more detailed explanation, along with pictures of sound waves to show you exactly what’s going on. The pictures help, trust me :) If you still don’t get it, watch this YouTube video:

One of the goals I had for (incf partition-36) was to improve the dynamic range in the recordings. This was a reaction to my previous album, SEPL, which was literally mastered in ignorance as far as dynamics were concerned, resulting in horrible sound quality and an album that was actually fatiguing to listen to. The need for the dynamic range in (incf partition-36) came directly from the ideas I had for the music when I first started writing the tracks. Many of them called for drums that were synthetic sounding, yet natural feeling. In my head I could hear synthetic drum sounds that had slight plastic and metallic feels, but were crystal clear. But in order to make them sound clear instead of muddy and wet, I had to allow them to punch and strike with great contrast to other sounds. So how well did I achieve this goal? Let’s look at a few comparisons.

That’s the waveform for the song Sky High from SEPL. Immediately you can see that the sound is essentially divided up into huge blocks; it’s either loud, or not. This is the problem with over-compression of the dynamic range. The drums in this recording are rather flat sounding and do not add any sort of characteristic other than rhythm to the song. At times they even sound like they’ve been slurred together because they’ve lost so much individual definition. This is most apparent around the 3:00 mark. Now, here’s a different file:

This is Alien Planet from (incf partition-36). It was this song specifically where I started with the synthetic/natural drums idea. Although still a block of sound, you can start to see more spikes in the waveform, and you can hear this too. The snare drum in Alien Planet tends to sound a little bit punchier, the drums are less smeared together, and the high hat in particular sounds much more distinct. While it’s true the song is simply boring in other ways (the dynamics are like “driving through Kansas”, as my high school band teacher put it), and could still stand to be less compressed, it’s still better than anything on SEPL.

This is Yeah, also from (incf partition-36). Although only a single instrument is played in the song, this is a lot closer to what I had in mine. Indeed, this song sounds warmer and fuller than any other song on the album. The “musical dynamics”, so to speak, are still like driving through Kansas, but the parts of the sound that should sound quieter really are, while the more intense parts are, well, more intense. Of course, Yeah leads into the song ROT13:

Although still a bit on the compressed side of things, it does tend to look a little bit better than even Alien Planet did. There are many more spikes and you can clearly make out areas that are supposed to be quieter than others. The drums on ROT13 are also much more clear than anything else on the album. ROT13 could still stand to be compressed less, but it’s probably the next best song after Yeah.

So where do I go from here? (incf partition-36) was generally better than its predecessor, though it still succumbed to the Loudness War in the end. My usual approach when recording is to apply compressors sparingly to individual channels in the recording (sometimes they actually are helpful), then apply a single “master” compressor (usually with this bad boy) to the entire mix. Once I export it out of SONAR, I do a few very light compression passes in GoldWave to even stuff out, but I think that’s where I’m still over compressing things. Why not just let the “master” compressor in SONAR do its job in the first place and only trim up blank space in GoldWave? I should be able to achieve a good mixdown directly within SONAR where I don’t need to mess with volume any more elsewhere. So that’s definitely the first thing I’ll do.

The other thing I’ll do is actually at the very beginning. By default, adding a new track to a song in SONAR puts the volume fader and input gain both at 0db. This should give me an additional 6db of headroom to work with, but sometimes I run out. Instead, I’m going to decrease the input gain on new tracks by about 25% and then go from there. That should give me more headroom to work with, and ultimately more use of the entire 16bit dynamic range a CD gives me.

In the end, part of me also just needs to be retrained, specifically my ears. Rather than immediately thinking “wow, that recording is way too low”, I need to instead look at the entire picture. Is it too low, or does it just use a lot of dynamic range? If it’s using a lot, does it need to in order to sound good?

So there we have it. My next CD will hopefully be even better as far as sound quality goes. I’ll keep researching things in the mean time and see if I can find some good literature on this subject.

Movies Galore

2009-02-09T00:00:00-07:00

Sitting next to my computer desk here in my room is a large stack of DVDs, all ones I haven’t yet watched. Most are movies I’ve never seen, though a few of them I have but just never owned before. About a month ago I made the decision not to buy any new movies until I watch this entire stack. It’s taking a while, but I’m starting to make some headway. So far I’ve watched about eight or so movies from it, while I still have to watch these:

Jet Li's Fearless
Space Truckers
Jet Li's Fist of Legend
Godzilla vs. Gigan
Exit the Dragon, Enter the Tiger
Enemy Mine
Guess Who (PSP)
The Chinese Connection
Tears of The Sun (Blu-Ray)
The Karate Kid
Robotech (Boxed set)
Robotech Shadow Chronicles (Blu-Ray)
95 Worlds and Counting
Krull
Terror of Mechagodzilla
Fahrenheit 451
Ping Pong
A documentary on US Navy carriers
The Beast from 20,000 Fathoms
Gorgo
The Hunted

That’s a lot of movies to watch…

Music and Drumming

2009-01-28T00:00:00-07:00

I haven’t been able to work on any new music in quite a while now, what with my desktop still being dead. It’s sort of bothersome to me since music was a hobby that gave me clear goals and was also my main creative outlet. Still, the downtime is giving me time to think about my future music goals, especially in terms of that live show I still plan to eventually do. I’m to the point where the next set of songs I write I want to construct specifically so that I can perform them live more easily. As before, my idea is for me to play my synthesizers up on stage while I have a live video show take place behind me on a screen. It’s sort of Kraftwerk-like, but I think it could still turn out unique due to the differences in our music, not to mention the fact that I’m a single person who would have to make a stronger stage presence.

So that’s sort of how I’ve been thinking so far. Then the other day I suddenly got an idea while talking with my girlfriend. Anymore the only video games I play are Rock Band and Wipeout HD, which has been the trend for a while. I’ve tried getting back into them, but they just don’t remain interesting to me anymore. As a result, I have a bunch of games and game systems just lying around not getting any use whatsoever. This begs the question, why even keep them? Why not sell them and put the money towards something more useful, like say… an electronic drum kit?

Rock Band’s drums are not like real drums. Being able to play them on expert does not mean I can just sit down in front of a real drum kit and bash out the drum part to "...And Justice For All". But it is giving me an overall feel for drumming and better independent control of my limbs. I’m to the point that I’m actually serious about learning drums for real, and wanting to use them in my live show.

The problem is, electronic drum kits (which I would both want and need) are a bit pricey. A mediocre entry-level kit starts at about $650, with a much more decent one starting at $1k. This is where the video games come in. Why not move on with my life, clean some of the old crap out, and get some money out of it all? Put more bluntly, why not sell most of my games and game consoles?

Here’s what I’m looking to sell:

Sega Genesis
Sega Saturn
Sega Dreamcast
Turbo-Grafx 16
Game Boy (the original "brick" kind... maybe)
Game Gear
PS2
Atari Jaguar

Don’t bother posting a comment here, as I’m not sure about pricing or when I’ll start selling them. My Wii will also be going, but I’m just going to give that one straight to my mom. The ones I’ll keep are these:

PS3 (best DVD player I've ever owned, it plays Blu-Rays, and I keep photos on it)
PSP (it's awesome for long trips)
Nintendo DS (a few fun games, plus it has my virtual cookbook software)
Neo-Geo (I spent waaaay too much money on this to sell it)
Neo-Geo Pocket Color (do you own one? It also has my only computer mahjong game, and not a fake tile-matching one)

Even if the money doesn’t end up going towards a drum kit, I’m still wanting to get rid of these. Music is more important to me than games these days, and I feel like I get more out of it.

Now if I can only fix my computer…

Tiling Window Manager Experiments

2009-01-13T00:00:00-07:00

About two years ago (maybe less), I ran a little experiment where I used ratpoison, a minimalist tiling window manager. It was something of an enlightening experience, where even though I didn’t continue to use it, I found that my computer habits were forever changed from that point forward. Specifically, I became a lot more comfortable with having windows that did not take up the entire screen.

Today, I’m doing the experiment again, although this time I’m doing ratpoison’s successor, StumpWM. It’s a window manager that tends to hold a special place in my heart since it’s written in Common Lisp. It pretty much behaves exactly like ratpoison, but has a few extra bells and whistles.

Right off the bat, I’m expecting the experiment to be a bit more difficult today than it was last time. Previously I had used ratpoison on my Alienware laptop, which has a generous screen resolution of 1920x1200. My normal laptop these days runs at 1280x800. With that reduction in screen real estate, I can’t tile as many windows at once. In particular, my browser window will be more difficult to use, and I’m pretty sure I’ll end up full-screening it quite often.

Why am I doing this? I dunno, to tell you the truth. I’ve always had this odd fascination with computing minimalism, as well as with interface designs, so I’m sure that figures into it somewhat. Or perhaps I just enjoy being rather masochistic to my pinky finger (StumpWM and ratpoison all use Emacs-like key commands).

SIDE NOTE: I am aware that the commenting system on my blog is a bit screwed up at the moment. I’m trying to figure out how to fix it.

Computer won't POST

2009-01-13T00:00:00-07:00

So, my desktop computer (which is the only one I write music on) has decided that it doesn’t want to POST anymore. The machine is an old Alienware tower from around 2003 or so, but it’s had its motherboard, processor, sound card, and video card replaced since I first bought it. Currently it’s running an Intel Core 2 Quad Q6600, nVidia GeForce 8800 GT, and some old DDR 400 RAM (1.5gb). Earlier tonight I went through and did some basic troubleshooting by re-seating all the components, and swapping RAM in and out to make sure the sticks were good. So far nothing, so I’m still not sure if something is dead, or what it could be.

For a while now I’ve been thinking about doing a major upgrade on the box, so I’m thinking that I’ll just go ahead and rebuild most of it. Since my main goal is to simply get it working again, my first purchase will be some sticks of RAM and maybe a new motherboard. The board I have right now supports both the older DDR 400 or newer DDR2 667, so if its just the RAM that’s dead, I can get back to work on it in no time. If not, I’ll get a newer motherboard.

After that, I’ll invest in a new case. The Alienware case I have is actually running out of room inside (I have three hard drives in it plus a DVD writer), has a couple of noisy fans, and they don’t cost all that much. Depending on the DDR2 specs, I may need a new power supply. If this is the case, I’ll purchase one with the case.

If the computer is still not working after this, I’ll have a decision to make: try replacing the graphics card, or go ahead and replace the processor. The graphics card is still AGP, so I might just be able to get away with testing it inside another computer. If it’s bad, I’ll replace that followed by the processor. Otherwise, I’ll do it in reverse. Either way, I might have to replace the graphics card when I replace the motherboard, depending on if I can find a decent board that still supports AGP.

In the mean time, I’ve gone ahead and purchased a small Edirol UA-1ex USB audio interface so that I can temporarily move my studio onto my Alienware laptop. The machine isn’t too bad spec wise (Pentium 4 3.2ghz, 1gb ram, GeForce Go 7800 GTX), so I should be able to get basic music work done on it. Plus the USB audio interface supports ASIO so it can handle all the heavy sound lifting. The added benefit of moving my virtual studio onto this is that, afterwords, I’m left with a portable virtual studio :)

Review: Super Street Fighter II Turbo HD Remix

2009-01-02T00:00:00-07:00

Growing up, I used to play Street Fighter II’: Special Champion Edition on my Genesis quite a bit. In truth, it was the first fighting game I played where I actually knew I was playing a fighting game. I still remember being confused by the special movies, where upon reading D,DF,F+Punch, I would attempt to individually press each direction on the D-pad. It took me quite a while (and a friend explaining to me) before I learned that D,DF,F+Punch meant “A smooth quarter circle forward plus punch”. Anyways, the point is, I have lots of fond memories with SFII. The later games never appealed to me quite as much, especially the Alpha series, but SFII was my starting point with 2d fighters.

Super Street Fighter II Turbo HD Remix (SSF2THDR) is essentially a remake of Super Street Fighter II Turbo, with new high-definition graphics and a remixed soundtrack courtesy of OC Remix. The gameplay itself is mostly identical to that of the original game, though the “remix” mode offers some additional balance. However, what I found was that the game is a lot harder than I remember, probably because I’m severely out of practice when it comes to 2d fighters. This hard difficulty is probably more of a good thing than a bad thing in my opinion. On the downside it’s going to frighten off new players, but at the same time it will give fighting fans an incredibly addicting challenge. I’m especially interested to find out what people who grew up playing 3d fighting games will think of it.

Musically, the new remixes sort of… well, suck. The title music is good and really fits, but much of the remaining stuff seems aimless and forgettable. A small part of me is hoping that they used "Club Del Toro" by Razya for Vega’s stage, though I’m not sure exactly what their selection process was so it’s doubtful.

Graphically it’s a mixed bag. The new HD graphics look amazing…as long as they’re in screenshots. Once they’re in motion, however, things fall apart very quickly and become the most disappointing part of the game. Rather than actually improving the entire look of the game, they instead used the exact same number of frames as the original animation and simply improved their resolution. Don’t get me wrong, each frame looks amazing, but seeing it in motion is more akin to seeing a slide show. This is not an SNK fighting game.

In the end, I feel more disappointed than happy with SSF2HDR. I was expecting an amazing game that played like the original, but looked and sounded incredible. Sort of how I felt when I played SVC Chaos on my Neo-Geo for the first time. What I’m left with, however, is an overpriced game that feels like a bandaged up childhood memory. Its challenge will keep me going back until I beat the game, but I likely won’t touch it again after that.

Free Software Philosophy vs. The World

2008-12-31T00:00:00-07:00

Many of my friends know (or have recently found out) that I’m a pretty staunch supporter of Free Software. With the exception of my music software and a couple of games (though not necessarily that many), nearly all of the software I use these days is free software as far as I can control. In the areas I can’t control, I’ve at least tried or have found work-arounds. For example, my car’s stereo deck supports playing mp3, WMA, and AAC files, but not Ogg Vorbis. It’s firmware is also closed-source. I don’t have the technical expertise to give it a new, free software firmware, and I don’t know of any other decks which have open firmware. So I do the next best thing and load up my iPod, which runs Rockbox, and use an 1/8” mini jack (AI-Net) adapter to hook it up to my deck. Since not all of my music is in Ogg Vorbis format yet (I’m working on it slowly), I also use a USB hard drive and another AI-Net adapter to listen to the other music. Believe it or not, this isn’t as big of a pain as you’d think.

Anyways, earlier tonight there was a post on a mailing list I follow that linked to blog post with a really good point in it. Basically, the Debian folks had another disagreement because of a literal, absolute interpretation of their Social Contract. This was also centered around firmware, and if one uses an absolute, literal interpretation of the document, then it cannot be allowed within the main package archives unless it’s Free Software. The blog post then goes on to mention Ubuntu's Code of Conduct in a more positive light because it avoids such absolute language as “100% free” and “never”.

I recently had to make a tough decision, where my own beliefs in Free Software and where I stand with it were called into play. At the time I felt as though I made the right decision, and after reading the posts in the blog and mailing list, I still feel the same way. However, it has led me to actually sit down and think about just how hardcore of a Free Software zealot I’m willing to be.

In terms of writing code, I’m pretty firm in my decision to only write open-source software because I believe, rather strongly, that Free Software is the ideal towards which we should all strive for as developers. I know that a utopia will never be achieved, but as in all areas of human society, it’s always a good idea to at least try to get close.

In terms of usage, however, I tend to make a few exceptions. If there is no viable alternative, I will consider non-Free Software. For example, I want to watch all the episodes of Monty Python’s Flying Circus. I can either go the illegal route and attempt to find them all online in an open format, illegally find them all in a closed format, legally find pieces of them, or just buy the boxed set. I don’t condone illegal behavior, and I’m wanting to see them all. Since I’m not developing them or the technology behind them, I’m willing to buy the boxed set on DVD. It’s not desirable, but it seems like a better idea than the alternatives. Does that stop me from wanting them in an open format? No, and I would happily purchase them again in such a format. But it works for now.

The same thing goes for my music software. Linux-based software is years behind in terms of usability, and I simply do not have the technical expertise (or the time) to help the existing projects out in this sense. Therefore, I use proprietary software, albeit only where I need to.

That last bit is key: where I need to. Whenever there’s a viable open-source alternative (especially if it’s Free Software), I will almost always turn to it unless it is severely crippled compared to the alternative. Overall, I see myself only developing open-source (and preferably Free) software. But as far as the user in me is concerned, I try my best to protect my freedoms as possible, and will always opt for the Free alternative first if it’s practical enough, but sometimes you’ve just gotta give a little.

Reigniting a Spark

2008-12-27T00:00:00-07:00

Out of nowhere, I pulled out my old copy of Heretic: Shadow of the Serpent Riders on Christmas and started playing though it again. For me, Heretic was always much more fun than Hexen was, probably because I didn’t really like the hub system in Hexen.

Regardless, I miss the 3d FPS games from that era. Gameplay was much faster back then, a trend that I think started to die out once Halo hit the scenes, and the levels were quite creative. These days it’s always the same thing over and over: a (bombed) urban city, an alien planet, Europe during WWII, or the middle east. Can’t I at least get an FPS game taking place inside a medieval castle, a sci-fi underwater biosphere, or on the moon?

I still have some Doom levels sitting around on my hard drive that need to be released. It’s just a matter of getting off my lazy butt, getting Doom Builder working again, and getting my editing environment set up. That last piece is actually the whole reason I’ve been putting it off :-P But a combination of things lately, ranging from Heretic to needing to do some MIDI editing using guitar patches, seems to be reigniting a flame that’s buried deep in my heart.

Filesystem go asplode

2008-12-04T00:00:00-07:00

During a trip I took to San Francisco last weekend, the filesystem on my Eee PC, for lack of a better word, broke itself. My screen went black (I was in an X session), and after both a reboot and a quick fsck of the filesystem, I found that the /usr/lib and /usr/bin directories were missing. I could still log in and navigate around since /bin, /sbin, and /usr/sbin were intact, but I couldn’t do much else. Everything else was in /lost+found and had been renamed with generic file names. The first instinct of a user is probably to just reinstall their OS, but as I sat there thinking about what I could do, I came up with an idea.

My Eee PC is running Slackware 12.1. I keep a local repository of the package tree here at home since I have quite a few PCs here running it, and have an NFS share open to access the tree from any connection. My idea was to simply reinstall the packages for the system so that the missing directories could be repopulated.

Booting into the OS, the first thing I did was delete everything inside of /lost+found. After this, I did a quick check and found that wpa_supplicant was still available, but the OpenSSL libraries it depended on were gone. Furthermore, the kernel was unable to detect my ethernet port all of the sudden, and so networking was out. Since I wanted to install the packages over NFS, I needed to get something working, so on my main laptop I copied over the openssl, dbus, and openssh (just to be safe) packages to an external hard drive. Once copied, I mounted the drive on my Eee PC, and installed these packages by hand using the /sbin/installpkg utility. Once this was done, I was able to launch wpa_supplicant, connect to my wifi, and mount the NFS share with the Slackware packages.

The next step was to reinstall only the packages that I had on my system before the filesystem broke. Slackware’s package management system keeps a log of currently installed packages in the /var/log/packages directory, where each file has the same name as the package it represents. Given this, all I needed was to run a simple one line command within the root package directory (well, one line when I typed it in; I split it up to make it easier to read):

for in in /var/log/packages/*; do 
    find . -type f -name `basename $i`*.tgz -exec /sbin/installpkg \{} \; ; 
done

This went through each previously installed package, found the corresponding tarball in the package tree, and installed it. Once this was done, I did a quick check with the df -h command to make sure the disk space was about where it was before (it was), ran /sbin/ldconfig by hand once as a sanity check, and then started X as normal under my usual account.

And…everything’s back to normal now.

An Early Review of Resistance 2

2008-11-18T00:00:00-07:00

This review is “early” because I’m not going to punish myself by playing a sub-standard game in a pathetic attempt to say “at least I finished it.” Instead I’m going to take this game back and turn it in for something much better.

The first Resistance ended up being one of the best games I had played all last year. It combined a very strong story with good gameplay, great firefights, fun weapons, and great level design. Unfortunately, the sequel seems to have traded all of this for nothing more than very pretty graphics. To start with, I can’t make heads or tails of exactly what’s going on. Sure, I understand that the Chimera have finally reached American shores, and that I’m now a part of some special forces team with other Human/Chimeran hybrids, but the story just doesn’t explain much past that. How did I go from walking in Britain in the middle of a snow storm to being inside a secret bunker with a bunch of other special forces guys? How did I go from Iceland to California? I mean come on, these aren’t even on the same continent, yet with the way the story works, you’d swear you walked out of the Icelandic bunker one day with your PJs on and suddenly arrived in Santa Barbara. They’ve removed the wonderful narration from the first game, which I think is part of the problem. Rather than feeling like you’re playing a story, you feel like you’re going through motions. Let’s not even talk about the characters, who are so bland and generic that I found myself wanting to rid myself of them by shooting them.

Sadly, the game mechanics don’t help any. While it attempts to build on the firefight-oriented gameplay of the first game, it ends up falling on its face by making them way too easy and way too short. Powerful guns like the Auger are given to you early on, and ammo is never an issue. Beyond this, everything feels like it’s been done before. Not just once, but hundreds of times. In fact, there was one level where I literally paused the game to make sure I wasn’t playing the first Halo again because it felt almost identical.

On the technical side, Insomniac has totally messed up the controls. The crouch and run actions are bound to a single button by default, which caused me much grief before I learned that there was an option to turn it off. The melee attacks are now mapped to R3 by default, which was also the cause of grief because I seem keep accidentally melee attacking things like propane tanks when I mean to simply look around them. They also feel a bit more sloppy this time around, even after I adjusted my looking speed. Physics are more wonky this time around as well. Ragdolls that get forced into odd positions upon dying have come to be expected, but I know for sure that refrigerators should not bounce silently like rubber balls while I can’t move an obviously detached park bench a single inch. Refrigerators should also not launch into the air when pushed over a curb that would not be more than two inches high in reality.

Just about the only thing it has going for it is the graphics, which are actually really good minus a few Z-order, rendering artifact, and unusually dark texture glitches here and there. But, just like a coin has two side, the quality isn’t always consistent. Usually I give developers a break here since making it consistent is pretty difficult. But why do the special forces dudes who wear gas masks have really, really good textures on their models while a just-as-generic scientist looks like he came out of the first Half-Life? Why are some wall textures well done, but a simple door frame look like it came out of Quake 2?

Multiplayer might make up for it, but I highly doubt this. Online play has never been a favorite of mine, especially on a console game. Besides that, I am thoroughly convinced that the three greatest online FPS games have already be made: Doom, Quake 3, and Rise of The Triad.

In the end, this is the game that finally did it. I am no longer confident that an FPS game can be properly done on a console unless it is very different (a la Metroid Prime). I give this a 3/10.

EDIT: I forgot to mention something else I saw. Near the beginning of the game, as you’re escaping an underground bunker (that’s a surprise), the entire level ends up shifting by about 20 or so degrees to show you just how unstable the area is. Now that’s pretty cool. Except that when that happened, it also changed basic physics we all take for granted. Fire no longer had to move up, but could burn sideways at an angle. Water also did not auto-level, and instead sloped naturally with the rest of the level. In a time when games are focused on realism, I fou

I'll have to post a funny movie I took of Resistance 2 where a car suddenly spawned 2 passenger doors after I accidentally melee attacked it. It subsequently got stuck and the physics engine couldn't decide if the 2nd door should be open or closed.

Review of Dead Space

2008-11-17T00:00:00-07:00

It’s taken me a while, but I’m finally almost finished with Dead Space on my PS3. Overall I’d have to say it’s been an interesting ride, one that will likely stay on my mind for quite some time.

The story, though cliche, does a pretty good job at getting you interested and keeps you wondering how it will resolve. You play as Issac Clarke, an engineer who happens to have a love interest on the ship your sent in to repair. Of course, as you approach you soon learn that something has happened to their ship and nothing is how it should be. From there on out, the survival aspect is emphasized and used to a pretty decent degree, with minor twists being introduced as time goes on. Sadly, some twists (like the religious one) don’t twist enough, and feel tacked on at the last minute. If there would have been introduced earlier on in the game and then emphasized more as time went on, they likely would have worked better.

One thing I enjoyed was that audio/video transmissions from your crew mates were generally kept short and to the point, yet still advanced the story. To me, this is much more effective than taking 20 minutes in-between each level to show what’s going on. There are also audio logs to be found from the crew members that do a real good job at keeping the atmosphere going.

Gameplay feels like a cross between Doom 3, Resident Evil 4, and The Thing, though the “strategic dismemberment” ended up being nothing more than a gimmick. What I found was that you can make it through most of the game using the plasma cutter, which is the equivalent of a pistol essentially. I think this is due partially to the equipment upgrade system, which allows you to turn the plasma cutter into a BFG pretty quickly. Once you realize that you now possess something stronger than the BB gun it once was, limbs become pretty easy to sever.

The game really shines in the audio and graphics departments. The sounds in the game are the most dead on I’ve heard in quite a while and really give you the feeling of being there on the space ship Ishimura. Likewise, the music, again though cliche, does an excellent job. It would be nice to hear something than overly emotional orchestra pieces, however (this does not mean I’d like to hear angst filled teenage “metal”, either). The graphics are also really good and also give a huge boost to the atmosphere. The textures are very sharp, and there aren’t any odd graphical glitches, save for the odd white dots that seem to come with games these days on edges of polygons. Brown and greys make up most of the palette, though there are enough useful splashes of color to keep it interesting. Lighting is real-time and per-pixel, though still not to the same quality as Doom 3, but like that game it is effectively used to give you a sense of creepiness.

Let’s stop there for a second and talk about that very issue. The game was billed as being a survival horror game, and hence was meant to be scary. Aside from just a few “oh crap!” moments (I can count them on one hand), I have yet to be genuinely scared by Dead Space. The game is just too well lit at times, and there is never the feeling that things may not turn out OK. Sure, there are things that jump out at you, but not that many, and sometimes you can see them coming. For a game that reminds me so much of Doom 3, this is the one area where it falls flat on its face. Doom 3 is still, by multiple orders of magnitude, the scariest game I have ever played.

That being said, the game is still very cool. It’s very brutal and violent, with player death animations sometimes taking upwards of 20 seconds to complete. The blood doesn’t always stand out against the textures, but it’s certainly there in buckets. Level wise they’ve done an excellent job. Areas are very well detailed, lighting is used effectively, and there is ample balance between cramped and open spaces.

One level in particular that I loved was the greenhouse level. The balance between well-lit and dark areas was spot on, the combat was well thought out, and the whole thing just worked really well. But the real reason I loved it was because it reminded me of an old game I used to play called Zero Tolerance. It was one of the few first-person shooters to have been released for the Sega Genesis, and the first third of it also took place on a space ship with a greenhouse area. That small bit of nostalgia touched my heart.

Overall, I really like Dead Space. It’s not nearly as scary as I thought it was going to be, but it’s a good game, if somewhat generic. I give it an 8/10.

Todo-Lister Experiment Finished

2008-11-11T00:00:00-07:00

As I described in my previous post, I conducted a bit of an experiment this past weekend where I tried simplifying Todo-Lister. I’m happy to say that it was quite a success.

My approach was pretty simple. I started by going back to the original C++ versions (the last being v1.21) of the program, looking at how the user interacted with it, what it produced, and what wasn’t needed. The very first thing I did was ditch XML. This wasn’t done in frustration with the language (I love XML), but rather because it made it harder for a user to look at the data files directly, and it just complicated things overall. Instead, Todo-Lister now uses plain text files that have a very simple syntax. Each item is now contained within its own file, and all of the files are stored in the user’s home directory under the .todo-lister folder. Using the file system as the database simplified pretty much everything from there on out.

C++ was also ditched in favor of C (along with GLib), though this time it was due to a preference in language.

So yeah, this new version is quite a bit more simple. If there aren’t any times, running the program displays nothing. If there are items, it will tell you so. Any overdue items will also be reported. Additionally, items have an “Exec” field where they can be configured to execute a command if they are overdue. Deleting an item means deleting the text file in your home directory. Adding an item can be done through the command line, though it’s likely easier to do just with a text editor (unless you want to use Todo-Lister in a script).

Source code can be found here.

One Last Try at TodoLister

2008-11-06T00:00:00-07:00

Earlier today, I found this article on Slashdot, which basically asked for some of the more obscure commands one could use on a Unix/Linux command line. Some of the things people mentioned I knew, some I took for granted (history), and some I never knew about before but found very intriguing (using a combination of at and cat /dev/urandom > /dev/dsp as a makeshift alarm clock). Either way, I’ve been thinking about the Unix philosophy yet again this past few days, and this article has helped push me along a bit.

I think I can safely say that any version of TodoLister (remember that name?) after 2.0, and Honeypot in general, were total disasters. Code bloat, poor design, and feature creep were all present and bearing their horrible teeth. Luckily, after abruptly deciding to take time off from working on them, I’ve come to my senses. TodoLister needs to do one thing and do it well. It also needs to learn to interact better. But most importantly, what I will term “software bling” just isn’t that important.

I’m going to try an experiment out, starting tonight and ending by the end of coming weekend. I’m going to go back to the old C++ code for TodoLister, pretty it up a bit, and try writing it to be more simple. Not “complete” or “correct”…simple.

If it fails, TodoLister is officially dead. If it

Let's try this again...

2008-11-06T00:00:00-07:00

If it fails, TodoLister is officially dead. If it succeeds, then it should be a fairly useful tool.

Blog of interest

2008-11-03T00:00:00-07:00

Aside from reading a few interesting posts on the LKML archives and a few quotes on Wikiquote , I actually don’t know that much about Linus Torvalds, the guy who created my operating system of choice. So when I found out he had a blog, I thought it would be interesting to follow it for a while on my RSS feed reader.

I’m glad I did this, because he’s pretty much sumed up my view of Obama, and the GPL and FSF, in a few paragraphs. That’s something I’ve had trouble doing before

http://torvalds-family.blogspot.com/2008/11/black-and-white.html

New Album Finally Released

2008-11-01T00:00:00-06:00

After a little over two years, my new album is finally being released: (incf partition-36). Many (not all) of the songs on this album have been featured on my Latest Music podcast. But, even if you’ve heard them on there, I still highly recommend downloading them again, as pretty much all of them have been updated in some way, be it audio quality or instrumentation. Not to mention the fact that tracks 5 and 6 are meant to blend together.

Individual mp3 files, as well as zip files containing the entire album and the album artwork, can be found on my Discography page.

This time around I decided to go with two separate zip files. The first contains the mp3 files, as usual. But the second one contains higher quality FLAC files. It’s much bigger, but using these will give you true CD-quality sound, where as the mp3 files are all encoded at 192kbit/s.

So yeah, enjoy! And keep watching the news page on my site for the announcement of my next album(s)…

New Album Has Gone Gold

2008-10-29T00:00:00-06:00

The term “going gold” in the world of software and video games generally means that the product has been completed and the master reference copy from which all other copies are made has been burned to CD. Well tonight I’d like to say that (incf partition-36) has gone gold.

I think I finally worked out all the last minute kinks from the album, so the release date of November 1st is still on schedule. For now, here is the album artwork I’ve now completed for it (fairly large JPEG image).

New Song Release - "ROT13"

2008-10-27T00:00:00-06:00

That’s right, I finished off a new song this evening! This new one is called ROT13, which is also the name of a simple "encryption" algorithm (though I use the term very lightly). I almost named the song “Bruce” after security expert Bruce Schneier, who is someone whose work I greatly admire. But in the end I decided against this.

So yes, this release means that my next album, (incf partition-36) is now completed! During the coming week I’m going to be putting the final touches onto the album as a whole. Once I decide that it’s 100% ready I’ll make an announcement both here and on the official Facebook page for Partition 36 (you have said you’re a fan, right?).

In the meantime, an mp3 of ROT13 by itself can be found here. On the album, the previous track ends up fading into this song.

Overall, I really like this song, and I think it just became my favorite off of my new album. I decided to try a bunch of new things with it, such as the harsher sawtooth bassline (created using Arturia’s Prophet V software), using my PolyEvolver Keyboard for the melody, and even some primitive vocals. In truth, vocals are something that I’ve wanted to get into but just haven’t. Perhaps this is a sign of changing times?

If you have a good stereo system that can put out a lot of bass, I suggest checking it out on that. You may just be surprised.

The Role of Anime in My Life

2008-10-26T00:00:00-06:00

I think I could probably divide my life up into four separate segments when it comes to anime and the influence it’s had on me. Growing up, there were certain shows I liked that I didn’t even know were anime at the time, such as Voltron and Robotech. During this period, anime was simply the “dirty videos under the animation section at Blockbuster I (assumed) couldn’t watch.” This assumption was mostly based on the “Restricted Viewing Under 18” stickers that most of them had. Still, Voltron was one of my favorite series growing up, and I also remember renting Robotech quite regularly. It was also during this time that I had an incredible love for Godzilla movies, which is how I was first introduced to Japanese culture.

It really wasn’t until high school (maybe even as early as the last year of middle school) that I really got into anime and understood what it was. The science fiction ones, especially with hot female characters, were my favorites. In particular, this was the period when I was introduced to Project A-Ko, Bubblegum Crisis, Ghost in The Shell, and Akira. Anime ended up affecting my musical tastes (it’s how I started listening to J-Pop and J-Rock, as well as hard rock/heavy metal), movie tastes (more interest in science fiction movies dealing with space and robots), and my overall character quite a bit. At the same time, while I still loved Godzilla, I didn’t watch his movies nearly as often.

By the time I got into college, I considered myself quite the otaku, though in retrospect I had only begun to scratch the surface. I had found others who held similar interests, learned about more anime, and started reading manga. It’s also the time when I started to really build up my anime collection, though unlike many, I actually kept to the legal methods and bought my episodes on DVD. Looking back, I can see that this trait was quite important in leading up where I’m at now.

Of course it was also in college when I started to formally learn how to read, write, and speak Japanese. These studies led to me being exposed to many non-anime Japanese movies, as well as J-Pop/J-Rock artists who were not connected to anime. Godzilla was still a favorite, but it was pretty rare that I ever watched one of his movies. Anyways, my studies ultimately led up to the period of time when I studied abroad in Japan for four months. I was extremely excited to go over there, not only because it was the one place I wanted to visit since I was a child, but also because of the anime.

Something happened while I was over there. I had found friends who also liked anime (this, unsurprisingly, was not that hard to do), and I remember having a great together. Aside from doing normal things together, we would hang out and watch anime, as well as visit Mandarake and various manga stores. My time in Japan also exposed me to alot of live-action Japanese movies and TV shows, which I had also started to take an interest in. But it was sometimes…awkward being an anime fan. One of my goals while living there had been to blend in as much as possible so that I could more deeply learn about the culture, in effect adopting Japan as my home away from home, at least in my heart. But the two didn’t seem to go together.

When I returned back to the US, my love of anime continued, albeit differently. I was now much more picky about things like subtitles, which manga to buy, and what made a certain anime “good”. The other study abroad friends I had met in Japan were ones who seriously collected anime, and by the time I returned I wanted to have collections like theirs. But remember how I said that I had always bought my anime on DVD legally? Well, I think it was a combination of that, my picky behavior, and the rift I felt between the Japanese culture that I had internalized and my otaku side (as well as a few relationship issues I had at the time). Pretty soon after coming back, I just sort of stopped watching anime. I really wanted to watch it, but every time I did I felt dirty. Not immature, which is what I tried to pass the feeling off as, but dirty. Dirty and depressed (mostly because it reminded me of Japan, which then made me nostalgic, which then made me sad that I’m not still over there).

This is a problem I’ve dealt with for the past four years, though I think I’m finally starting to see the light at the end of the tunnel. Much of it has had to deal with internal reconciliations between my respect for the everyday Japanese culture I had experienced and deeply respect, my love for anime, and where I’m at professionally.

Yes, I think my job ultimately played into the whole situation. In the old department I had really started to embrace a sort of elitist professional image. Rockport shoes, kakhis, polo shirts, parted and combed hair, nice outerwear… all of this was to seem mature. It was also because of the crowd I was around at work (whom I don’t hold any of this against, mind you). Anime had no place in the world of this new Andrew.

Luckily things changed. Departments got switched up, sense got knocked back into me (thank you Dhammapada, Scrubs, and Josh), and I started to live my life as myself again. I’ve had a chance to look back on my experience with anime, listen to myself and learn what I really want out of it, and decide where I want to go with it. It’s not about collecting as much as I can anymore, the hot anime babes, watching as many series or reading as many manga book as I can, or using it to remember my time in Japan. It’s about finding a show I like and watching it, and connecting with people with similar interests, plain and simple.

That being said, I’m going to go watch the first anime I’ve bought in over a year: a boxed set of Robotech.

Just for fun - Common Lisp Attribute Support

2008-10-25T00:00:00-06:00

One of the reasons I like C# is that it gives me the ability to include metadata within my code by tagging things with attributes. I’ve found this to be helpful when writing any sort of plugin or loadable module framework, where I can specifically declare certain classes or functions to have certain properties.

While I haven’t had a great need for it in Common Lisp, I have, from time to time, thought it would be nice if I could say “give me all the classes from this package that have this property” or “give me the slot on this class that was designed to perform this function.” You can already do this without metadata in Common Lisp, but it would involve extra packages, global variables holding lists of symbols, and other less-elegant things. So, being a bit bored the other night, I decided it would be fun to add this to my favorite language.

The result is a library that I call CL-Attributes. It works off of Common Lisp’s MOP, which is basically how the object system (CLOS) is implemented. When you create a class, you use a new metaclass defined in CL-Attributes, from which additional class and slot options get inherited. Property lists can then be tagged onto the class/slot definition through these options. Then the library includes a couple of “query” functions to find classes or slots with certain attributes. It also allows you to add and remove properties while the program runs by giving you a couple of convenience functions. Documentation on how to do all of this is available here.

The source (released under GPLv2) for the library can be downloaded here: http://partition36.com/~personal/cl-attributes.html. Unfortunately, it only works with SBCL right now, but if I get around to it (or if someone sends me patches), more implementations will be supported.

Wii Fit Review

2008-10-20T00:00:00-06:00

Today I was lucky enough to get my hands on a copy of Wii Fit. I had originally gone over to Circuit City at 10am this morning, which is when they open, but they never received any in their overnight shipment. Feeling lucky, I ran across the street to the Best Buy, which I had already checked yesterday. When I walked into the store I saw that there was a single copy left standing next to a table. Success!

The game feels like most Wii-centric Nintendo software that is released these days. Emphasis is on Mii-like cartoon graphics, cell shading, bright colors, and peppy music. It started off by asking me how tall I was (5’8”), followed by a quick test of my weight using the Wii balance board. Not surprisingly, I was overweight (which I confirmed from a doctors visit a few months ago, though I was happy to see I’ve at least lost 13 pounds since then). However, the game never really made me feel bad about this. Instead, it made let me set a goal (which maxes out at losing 22 pounds; I need to lose 40), and then it helped me feel as though I could really achieve this. Obviously just playing the game won’t help me lose weight, as I also need to change my diet to a more healthy one. This is one area I felt was a bit lacking, as I would have liked to see it promote eating a balanced and healthy diet.

Gameplay/training centers around one of four categories: yoga, “strength” training (quotes added by me), aerobic games, and balance games. When you first start the game has you chose a trainer, male or female (I of course chose female), for the yoga and strength exercises, which can be changed later if you’d like. The balance games are probably the most fun use of the balance board, as you actually feel like you’re playing a mini game as you keep your Mii from falling off a tight rope, skiing through flags, or doing a ski jump.

The aerobic games are generally good at getting my heart going, though the nearly DDR-esque rhythm game where you step on and off the board wasn’t so much of a work out as it was simply good fun. There are two running exercises (I’ve only tried one), which you don’t actually use the balance board for. Instead, you hold the Wiimote and then run in place. The game recommends putting the Wiimote in your pocket if you can, and I’ve found that it works best when it’s in your back ones. Though it hasn’t explicitly said, I’m thinking that the goal of the running ones is to keep more of a constant speed than running really fast.

“Strength” training exercises haven’t been that hard for me yet, except for the push-ups. This is probably because I have really strong legs and abs. So, I would definitely have liked to see at least one more arm training exercise right off the bat. However, I had never heard of jackknifes before. I’m still a holdout who still does sit-ups even though I know they can be bad for your back, so learning this exercise is actually pretty cool.

The yoga exercises were interesting, as I had never tried yoga before. They were quite calming, and at first I thought I wasn’t actually exercising. However, you never really realize just how tiring holding your balance can be, especially on that dang tree pose. I can definitely see myself doing more yoga than just about anything else in the game, with some good amounts of the jogging game thrown in for good measure.

So does it actually give you a work out? I think I can safely say that yes, it certainly does. I was sweating as though I just came back from a 30 minute bike ride when I finished my work out this evening. This is pretty encouraging, as I’ve been looking for a way to continue exercising now that it’s getting a bit cool to go for bike rides. The only thing is that I think you need to use Wii Fit just a tiny bit longer than the amount of time you usually spend on your exercise routine.

Overall, I really like Wii Fit.

Wii Fit

2008-10-20T00:00:00-06:00

Overall, I relly

Evangelizing Parenthesis

2008-10-19T00:00:00-06:00

For me, my journey of learning how to program in Common Lisp has been such that progress seems slow overall, but has occasional jumps. For example, it took me almost a full year before the language first “clicked” in my mind. After that progress was fairly flat until ASDF and packages suddenly made sense and were no longer a burden. More recently, my decision to focus almost exclusively on SBCL has lead to yet another leap in progress, namely that it’s OK to make such a decision.

This weekend I checked out books from a library for the first time in nearly six years (I’m not much of a reader). The ones I settled on were a book on intelligent systems, and one about CLOS, the Common Lisp Object System.

"Lisp is worth learning for the profound enlightenment experience you will have when you finally get it; that experience will make you a better programmer for the rest of your days, even if you never actually use Lisp itself a lot."

That was something that Eric S. Raymond once said, and I have to say from experience that it holds true even today. Opening the book on CLOS and reading the first few pages of chapter one reminded me of this, as already I am seeing examples and explanations of CLOS that are impacting the way I think about object-oriented programming. Like mixins, interfaces and abstract classes in other languages aren’t just incomplete classes that can be used as skeletons or bases, but are things that ”…augment the structure or behavior of other classes.” What’s more is that the Meta-Object Protocol (MOP) in CLOS gets me thinking more about C#’s reflective capabilities, and how they now seem incomplete compared to what’s in CLOS (even the dang slots, which are analogous to fields, are instances of classes!)

Maybe it’s just how I learned to program in school. Maybe (likely?) enough formal software engineering and language design material was absent from my, and others’, college education that Common Lisp only appears to offer this much insight. Whatever the case, I’m glad I’ve taken it upon myself to learn this crazy thing.

What Would Be Nice

2008-10-15T00:00:00-06:00

Know what would be nice? If the Alpine deck in my car that can read USB sticks would go the extra step and actually make life easier for their customers. Like playing files in alphabetical and numerical order rather than the order it appears in the FAT table. That would make it so that I would not have to worry about the order I copy the files in, or write crazy Python scripts to fix them after the fact. Or supporting Ogg Vorbis and FLAC files.

Know what would be nice? If ext2/3/4 were the dominant file systems for pretty much everything. It’d sure make it easy to carry my data around. And it’d also be nice to symlink some files on the USB drive I use with my Alpine deck…

Know what would be nice? If Sony put support for Flash 9 on their PSP as well, not just on the PS3. As well as support for Ogg Vorbis, FLAC files, and XviD on the PSP.

Know what would be nice? If Mono shipped with a set of working Gtk# 2.12 libraries rather than 2.10. I know there are some issue with it, but you’d think this would be a priority so that the Windows platform doesn’t lag behind.

Know what would be nice? If you could use Reason as a VST or DXi plugin rather than a ReWire device. I don’t like using two sequencers at once.

Know what would be nice? If Palm made it easier to program for their devices. They used to have Java for it, but that sort of disappeared. If only Mono/.NET, SBCL, or even an easy way to run GTK+ apps existed for it.

Know what would be easy? If companies start realizing that programmers are users, too. Just because I’m a programmer does not mean I love jumping through hoops to figure out how to get something working on a single platform. It also does not mean I love reading through tons of documentation to learn a non-standard API. Open source - generally - has it right, people.

Know what would be nice? If DRM finally died. Not everyone likes the iPod (or uses the factory firmware, for that matter).

Know what would be nice? If companies took customers more seriously in their hearts than in their corporate image. Anyone can say that the customer is important.

Yeah… that would be nice.

Friend's Song Release, My Own Upcoming Album Plans

2008-09-22T00:00:00-06:00

A friend of mine, Mike (aka Esselfortium), has released a new song called Invis. It’s a really cool song that I highly suggest checking out. To me it sounds like a mixture of Boards of Canada and some MOD files I used to listen to. Mike has been a big inspiration to me for the past few years, especially when it comes to spending more time on sound quality and complexity of my compositions. This is what he had to say about it

I think it's time for me to finally call this song finished, after having initially started on it almost a year and a half ago. The original idea came from me hearing a few seconds of some Autechre song (no, I don't know which one) on a streaming radio station, with some awesomely glitchy drums, being inspired by the sound, and immediately setting off to write a song based on that inspiration. The resulting song is very much my own, though, and it's definitely not a remake or imitation of anything that I know of. :P

You can download it directly from here.

On my side of things, I’ve decided to expand my previous song Yeah and make it the lead-in for another song. This new one starts out a lot like stuff found on TaQ’s album Ongaq:Stromatolite, and uses similar drum patterns and sounds. Right now I have it leading into some strings, but it sounds overly pompous and I don’t like it very well. If I can work out the wrinkles, I’m going to try replacing them with another idea I had for a rather wicked hoover/supersaw sound I made on my Poly Evolver. Together with an ostinato part, I should have a melody going. Once this song is completed, (incf partition-36) will be officially finished. In the mean time, ogle at the cool cover I made for it ;^)

Some Thoughts on Live Shows

2008-09-16T00:00:00-06:00

What is my music?

This was a question I came up with tonight, where I’m not talking about genre so much as I am about intention and conveyance. On my bio page I state that for me ”…music is something that an artist can use to express images, thoughts, and sounds found within their heads.” Awesome! What the hell does this mean?!

It’s occurred to me that my ideas for live shows fall through because of an inherent conflict of goals. On the one hand I wish to present my music as being distinctly electronic, influenced by the underground scene, and non-commercial (free as in beer and speech). But on the other hand, I want to portray electronic music in a more noble light, where it can be music that people sit down and listen to. To be analogous - and rather exaggerated - I’m looking for a rave in an opera house.

Like a friend of mine has been mentioning to me, it’s all boiling down to my target audience. To that I would probably add “and my target goal.” What do I want my music to be?

Conceptually, my music has always been rooted in video game environments, and the act of recreating the feelings, experiences, visions, and colors I think of when playing a game as sound. BountyH, for example, is designed to sound like it’s on a somewhat lush alien world, and has been directly influenced by the Metroid Prime games. But BountyH is also about sound in terms of bass vibrations reacting against higher pitches, as well as pulsing rhythms. This would be the exact goal I’m looking for: a blending of my own personal thoughts (What was it like for me to land on Bryyo in Metroid Prime 3? What did the visuals sound like to me?) together with musical aesthetics.

So I’ve found my target goal, but what about a target audience? I’ve tried very hard not to think about this in any sort of logical or business-like manner up until now, usually resulting in me saying “anyone who listens to it.” Unfortunately, there’s a social side to business I cannot ignore if I want to find people who will want to watch a live show. The musical aesthetics side is somewhat limiting, appealing more to musicians who appreciate electronic music in the first place. But the fact that I pull influence from video games (and to a lesser degree - at least these days - anime) means that I should be able to strike a chord with gamers.

Getting back to doing a live show, I now have a target goal and a target audience. Now I only have to figure out where I could perform so that my target audience can more easily find me, and come up with a format that supports my goals.

New Song Released!

2008-09-12T00:00:00-06:00

I’ve finally finished off a new song! This one is called Iridescence and will likely be the last one I release on my Latest Music podcast from my upcoming album (incf partition-36) (I want to save two of them as surprises).

This new song is a big departure from my usual style. It came about when I found that a phrygian scale is often used in Goa Trance. Since I really like this form of music, but have never been able to compose anything like it before, I decided to give it a try. In the end, I think it turned out really well, especially how the song transitions from the voice samples to the alternate melody around 2:04.

The bass line is mostly composed of my Poly Evolver Keyboard layered with an instance of Prophet V. All of the melody lines are done with Prophet V in various modes. I also used my older JV-1010 quite a bit for various background parts, more so than usual.

Worse really is better

2008-09-10T00:00:00-06:00

I tried, I really did. I tried going back to Thunderbird as my main email client. In truth, I love using Thunderbird and I would recommend it in an instant to anyone. But the truth of the matter is, when it comes to how I use email, and what I need out of an MUA, Mutt just fulfills my needs better.

Open-Face Cube Steak and Fried Tofu Sandwich

2008-09-01T00:00:00-06:00

This was an impromptu recipe I came up with tonight when I found myself with cube steak, tofu, and a hungry sensation in my stomach. I never made any measurements when I fixed it, so these might be off a little bit…so adjust to taste. This serves two.

Tofu Mixture

1/5th block firm tofu, chopped into small pieces, almost like thick matchsticks
About 1 tbsp olive oil
2 1/2 tbsp red onion
1 1/2 tbsp garlic powder
1 tsp paprika
2 tsp soy sauce
black pepper

Put about 1 tbsp olive oil into a small frying pan. It should be just enough to cover the bottom. Add garlic powder.
Heat oil on medium-high heat, combine other ingredients. Add black pepper to taste
Let fry, stirring occasionally
Once the tofu is a golden brown, set aside and keep warm.

Cube Steak

Two cube steak patties, each about the size of a slice of bread
2 tbsp red onion
2 1/2 to 3 tbsp mirin ^\*
Salt and black pepper

Heat a frying pan using medium heat.
Place cube steak into the pan and lightly brown each side.
Once each side is browned, pour mirin over each side. Add red onion on top of meat.
Let steaks simmer in mirin
Once mirin is mostly evaporated, turn heat up to medium-high or high. Add salt and pepper to taste.
Continue letting the meat grill on each side, flipping frequently. Keep piling the red onion which falls off back into the meat and press down on it with your spatula. Continue until the meat is thoroughly cooked and lightly grilled on each side.

Serving

Place bread (I recommend multi-grain sandwich bread) onto a plate. Spoon tofu mixture onto the bread first, and then lay the meat on top of that. Additional soy sauce, pepper, or paprika might go good with it. My grandma ate it with A1 sauce and said it was delicious, so you may also want to try that. I also found that it goes really well with a side of whole kernel corn.

*Mirin is Japanese cooking wine, and shouldn't be too hard to find; I found some at my local grocery store (Kikkoman's is actually really good). If you can't find any, you can substitute it by combining 1 tbsp white sake with 1 tsp sugar. When I say "white sake", I mean the clear stuff.

Good Article, silly Comments

2008-08-27T00:00:00-06:00

I found an interesting security-related article this morning about how the “Forgot your password?” links are a security risk. But what’s even more amusing are some of the comments I’m seeing on the page and seeing just how bad security theater has influenced the general population for the worse.

One person mentioned that some easy ways to protect yourself are to spell your answer backwards, drop the vowels out of the answer, or to switch answers. But what this person is not considering is that these techniques have to follow the same principles as passwords. Removing letters makes your answers less secure (automated cracking programs have to do a lot less work, and a good cracker would skip trying things by hand), while reversing them does nothing. Switching answers is also pointless if things are being generated automatically. Saying my grandmother’s paternal eldest first aunt’s middle name was “Denver” can still get cracked.

Another person mentioned that their bank uses secret images to authenticate users. These and the “secret messages” (the “do you recognize this message?” things) are some of the most annoying things of all since they do absolutely nothing. If a cracker can set up a fake website that looks identical to the one you think you’re logging into, they can create a small interface to send your responses back to the real website. In other words, you start by typing in your user name, and then the fake website sends that to the real one. It then gets a response back from the real one, parses out the secret message or image, and then displays that same message/image back. At that point you recognize your secret message/image, figure it’s the real site, and type your password away.

I especially like the ones who suggest that users “lie” in their answers. Again, with an automatic cracking program, this does absolutely nothing.

Further down I spotted a person giving hints out about how to make a strong password. They got it right for the most part (“alpha and numberic combinations with at least one capital and a special character”), and I’m not criticizing them for this suggestion, but I’d like to add something to this. As Bruce Schneier mentioned in his blog, if you want a password that will be hard to guess by automatic programs, you should follow this:

...something not on any of the root or appendage lists. You should mix upper and lowercase in the middle of your root. You should add numbers and symbols in the middle of your root, not as common substitutions. Or drop your appendage in the middle of your root. Or use two roots with an appendage in the middle.

What I’m usually more concerned about is if the password recovery page is going through a secure connection.

First Impressions of Lost Planet: Extreme Condition

2008-08-24T00:00:00-06:00

Take this as proof that I need to stay more on top of games.

While up at my local Gamestop today looking for a new DS game, I happened to come across a used copy of Lost Planet for PS3. I had heard mixed things about the game, but since I was in the mood for an action/adventure game, I figured I’d give it a try.

So far I’ve played through just the first two missions of the game’s eleven missions, but I’m already feeling hooked by the gameplay and story. In many ways, it feels like a game with more old-fashioned run-n-gun gameplay, with vehicle combat thrown in for good measure. However, what’s interesting is the use of “thermal energy”, which is required to keep your character alive and can only be obtained by killing baddies or blowing up scenery.

The basic idea is that since the planet you’re on is so cold, you need this energy to keep yourself warm, otherwise you freeze to death. If it reaches a level of zero, your health starts draining down. At the same time, the game features the popular regenerating health bar, where as long as you don’t get hit for a short amount of time you regain health. But this time, letting it regenerate actually uses more thermal energy. So not only does it keep you alive as you play, but it’s required to heal you as well. For me, this is a nice touch. The one downside so far is that keeping on top of your thermal energy needs isn’t that difficult, and I haven’t had many times when I’ve run low yet.

Graphics wise it’s not bad, but I don’t see anything spectacular about it. The textures are fairly bland and at times low-res, and there isn’t much variation in lighting. But the environments are pretty detailed, while the character and mech models are really well done. Overall I’d say the theme looks like a cross between Killzone and Coded Arms, but frozen and full of snow.

We’ll see how the game turns out. I’m pretty sure I’ll finish this one, and so far I’m having a blast playing it. But that’s what I’d expect from the man who made both Mega Man and Onimusha, Keiji Inafune.

DDR and Strategy

2008-08-17T00:00:00-06:00

A couple of years ago a friend of mine mentioned that DDR essentially had no strategy to it, especially when compared to other games such as StarCraft or fighting games. I certainly agree with this in part, but not entirely.

Strategy in DDR doesn’t lie in any sort of goal involving conquering bad guys, but rather in conquering one’s own goals. Such goals might be either long-term, such as beating a very hard song (i.e., Max 300 on heavy), or getting a AA or AAA rating on one. In this scenario, the strategy would be understanding how to gauge exactly what you need to work on to attain that goal, and which songs to play to prepare yourself. For example, my main problem used to be endurance, and so I’ve been doing things such as bike riding and hiking to build that up. These days timing is the main thing that is keeping me from beating Max 300 on heavy with a B or better (my current long term goal), and so part of my current strategy is to work on getting A and AA ratings on songs that have odd or irregular beats.

Short term goals would be things such as deciding to simply beat a more difficult song before you’re finished playing for the day. This actually requires you to actively think about what you’ll be doing while you’re playing the game, forcing you to think about multiple things at once. For example, if I’ve been playing for 20 minutes and want to beat Exotic Ethnic on heavy with a decent score, my strategy would be to play a slightly faster song first, play two slower songs (possibly one with lots of slide steps*), and then attack my goal. This would give me a boost to get my energy up again, allow me to rest just slightly and build up a little more, allow me to get a couple of practice slide steps in (Exotic Ethnic has a bunch with the way I dance it; yes, there are multiple ways to dance a song), and then I should be ready to try the song.

That’s just the thing that people who never play DDR or never push themselves on it never recognize. You have to make plans, think about them, and be ready to modify them as you dance. Aside from what I just mentioned, there is also a sort of “immediate” strategy related simply to the way you dance a song. If you see the pattern “left-down-up-right-down-left-down-right”, then you need to plan out ahead of time - even in the middle of a song seconds before you arrive at that pattern - to start that pattern on the left foot. The rest of it just falls into place after that, but knowing how to work the current pattern you’re on to arrive at that left foot starting point is where the strategy comes in. Do you keep alternating feet, or throw in some slide steps? Are you in a position where you can even make it to this goal? Is your body going to be able to get into the proper position to start on the left foot? All of these things need to be taken into account.

\* For those who don't know, a slide step is one where, because of your current position on the pad, you're required to use the same foot to hit two consecutive notes. An example would be using only my right foot to hit the sequence "up-right-up" in quick succession.

Photo Plan

2008-08-10T00:00:00-06:00

I bought a new camera today, a Nikon Coolpix (my personal favorite line of cameras), model S550, 10 megapixel. My previous camera, also a Nikon Coolpix, still works and is still a good camera, but I wanted a smaller one that had a rechargeable battery.

For the past few days I’ve been thinking about trying something new. My collection of pictures is unfortunately very small (especially of friends, family, and myself), and I’m wanting to expand it. To do this, I’m going to try taking 15 pictures every month. I’m sure most of them will be of scenery, but during this small experiment I actually hope to start getting pictures of my friends so that I can start remembering these days later down the line. Then, at the end of each month, I’ll pick a few of the best pictures, likely scenery ones, and post them either to Facebook or Flickr. This should also help me learn a bit more about photography, which I’ve wanted to do ever since a friend took a class in it.

"Extreme" Culture and Modern DVDs

2008-08-07T00:00:00-06:00

Ever notice the number of DVDs sold these days that have special “unrated” editions available these days? It almost feels like every action, horror, and sci-fi movie is almost required to have one. But from my experience, the unrated version doesn’t always appear that much different, if at all, from the normal version. I remember getting the unrated version of Doom and struggling to see any difference (which I think was just a few extra shots of violence). At least the unrated version of RoboCop actually shows enough extra gore and violence to justify it’s label.

I think part of the problem is that the whole idea of an unrated edition tends to plant one of two ideas (both of which are likely false) in the buyer’s head: it has a sequence where the female character you developed a crush on when you saw it in the movie theater shows her naked breasts, or three times more gore than the first two RoboCop movies and Starship Troopers combined. This makes it easy to sell unrated versions to buyers who already bought a normal one.

SLE Screenshots

2008-08-02T00:00:00-06:00

I put together a small page this evening with screenshots from a Doom project I once worked on called SLE. The levels I had made for it were some of the best I’ve ever created. Which is sad, since I’ve never released them. I should get on that…

Small bit of 1337-ish Code

2008-07-31T00:00:00-06:00

I’m proud of this. I’m sure there is better code out there…but I’m still proud of this since I didn’t think I could do it this tersely.

 
(defun split-string (string delimiter) 
  (loop for i = 0 then (1+ j) 
     as j = (position delimiter string :start i) 
     collect (subseq string i j) 
     while j)) 
 
(defmacro csv-string->string (str) 
  `(string-trim '(#\") ,str)) 
 
(defun get-num-if-num (str) 
  (multiple-value-bind (num len) 
      (parse-integer str :junk-allowed t) 
    (if num 
	(if (equal (length str) len) 
	    num 
	    str) 
	str))) 
 
(defun read-csv-file (filename &key (delimiter #\,)) 
  (declare (pathname filename)) 
  (with-open-file (in filename) 
    (let* ((raw-header (read-line in)) 
	   (raw-ish-header (split-string raw-header delimiter)) 
	   (header (loop for i in raw-ish-header 
			collect (csv-string->string i)))) 
      (values (loop for line = (read-line in nil) 
		 while line collect 
		   (loop for i in (split-string line delimiter) 
		      collect (get-num-if-num (csv-string->string i)))) 
	      header))))

CSV file parser for Common Lisp. Takes a CSV file and returns multiple values, where they are the parsed data (with integers converted to real integer) and the headings, respectively.

Self Care

2008-07-16T00:00:00-06:00

With how things have been the past two weeks, I’ve been trying my best to keep the stress levels down and just sort of decompress, but it’s been difficult since I haven’t had a lot of time. So today I decided to start out by changing a few of the things I do for self-care. First and foremost, I plan to ride my bike more frequently, probably on the order of a few times a week; this should also help keep me in shape. I’m not planning any of those “omg I’m going to ride from here to the other side of Northglenn!” things that I used to do though, since I think they actually kept me from riding my bike more often than I wanted. My guess is that riding my bike with that mindset burns me out more than it helps me.

To be productive, I’m also taking advantage of a sudden spurt of interest in our pool here at home by taking extra good care of it. Vacuuming a pool is actually very relaxing, believe it or not. Plus it gets me outside and active, like bike riding does.

I also have to be honest with myself and admit that I’m not as into video games as I once was. Over the weekend I sold six or seven games I just never played (and got back $114 to boot) and I honestly don’t miss them. I pretty much (still) only play Rock Band/Guitar Hero 3 anymore, with a couple of Wii Virtual Console games thrown in for a good mix. I think it’s just that I’ve regressed into being more of a casual gamer who likes playing for shorter time spans.

I’d like to list fishing here as well, but it’s become somewhat of a burden recently. See, I’m the sort of guy who doesn’t like to go off and do many of things by himself. Since I don’t know anyone who likes to fish as often as I do (or at all, for that matter), I just don’t go very often. Besides, fishing is also like bike riding in that if I go on a “omg I’m going to a mountain reservoir!” style trip too often, I get burnt out too easily.

Moderation seems to be the key. Short bike rides, fishing at a nearby pond, small rounds of Rock Band/Guitar Hero 3 here and there, that’s most of my new self-care plan.

Beat 'em up!

2008-07-16T00:00:00-06:00

For example, there was a time when I was interested in learning how plugins for software were written, and also how the software they were written for was written. That’s why TodoLister 2.x got plugin support (which, thankfully, was later removed). My new interest is embedding scripting languages into programs. Guess what I’m using to toy around with this idea? These days I’m a lot more careful about what I release with TodoLister, but an internal version may have experimental support for this sort of thing.

I think it’s important to have beat ‘em up programs so that we, as programmers, are free to explore ideas. It’s sort of along the same lines as having a language for work (C#) and a language for play (Lisp, Python). They aren’t going to be black and white like that - things never are - but I truly think that mental split gives programmers that fuel that drives our inner creativity.

I also think it’s important to let people know that a piece of software, no matter how serious we are about it, is considered beat ‘em up quality. That doesn’t mean that it’s poor quality, and in fact beat ‘em up software probably shouldn't be of poor quality. Instead, I think it should say something about how stable a program is in terms of features and its API. The programs I write for work, and some of my other programs I release, don’t always see a lot of updates because I intend for them to be more stable. They need to be in case someone else needs to build off of them. This is the sort of things that serious users need to understand about them, while anyone who uses TodoLister should understand that, with the next release, they might have something radically different. And that it’s OK.

Eee PC, Meet Slackware

2008-07-05T00:00:00-06:00

It was going to happen eventually. I’ve finally ditched the default Linux distro that my 4G Eee PC (701) that it came with and replaced it with Slackware. It runs extremely well, actually. Installation was only slightly convoluted, with me needing to first make a bootable USB stick and then copy the package tree onto a USB hard drive. The latter gave me only one additional problem, which was that the system on the bootable USB stick didn’t have udev or something, and so the /dev/sdc device wasn’t being made properly. No big deal, since I could just create it manually by doing:

mknod /dev/sdc b
8 32 mknod /dev/sdc1 b 8 33

After everything was installed, I installed a new kernel and various other packages from http://slackeee.strangled.net/, which I had downloaded to a USB stick beforehand. This fixed any issues I would have had (for example, no wired or wireless network access), but required me to manually fix /etc/lilo.conf and /etc/fstab to point to the correct root disk (the new kernel shows the internal SDD as a /dev/sdX device, not as a /dev/hdX device). Rather than hard coding the device into the /etc/fstab file, I instead used the drive’s UUID (which can be obtained using the blkid program). After modifying /etc/lilo.conf (which still needed the device hard coded in), I ran Lilo and everything was working.

Wireless works. I wasn’t able to get the Wicd program on the SlackEee site working, so I’ve decided to configure and run the wpa_supplicant program by hand, as well as iwconfig. For the X server, I used the X11 config file over at http://armcandy.fr/eee/. I also used his shutdown fix. To save some wear and tear on the SDD, as well as boost performance slightly, I added the “noatime” option to the /etc/fstab file for main partition.

This is actually the first time I’ve used Slackware 12.1, and so far I like what I see. The system is extremely responsive, even on my Eee, and has a good number of packages. Mono is, of course, still not included, but I was happy to see that Python 2.5 was. wxPython wasn’t included, but was easily installed with a SlackBuilds package.

What’s really cool is that I was able to get more free space out of the SDD with Slackware installed (including development tools, but minus TeX, TCL, Emacs, and the kernel source). I haven’t started trimming out packages I don’t use yet though, so I’m still waiting to see just how much more space I’ll end up with.

Overall…I’m definitely sticking with Slack! If only I didn’t have Gtk# apps to develop, I’d use it on my main laptop too…

Sheet Music and a Music Stand

2008-06-26T00:00:00-06:00

So earlier tonight, I bought a music stand. Given that I’ve started making plans to put on a live show later this year, and I plan to use sheet music as a backup should I suddenly forget how to play something, I figured I needed one.

This, of course, brought about an interesting realization: I haven’t read sheet music in about four or five years. I have ways of generating it from my music files (SONAR can print sheet music, though NoteEdit + GNU Lilypond), but it’s obvious that I’ve forgotten a few pieces when it comes to reading it. Most of it I still remember, and I was able to read through my own song last night. But there are a few things that I’ve had to relearn already: I forgot that the symbol for sforzando existed (and what it meant), how to write multi-measure rests, what a segno looks like, and many of the accent marks.

Reading and writing sheet music is definitely a skill I want to keep up with. The piano roll view in most modern software is definitely easier to use when it comes to electronic music and laying down drum beats, but I’ve always felt that it detracts from the music in a different way. Maybe in the future, I’ll do more with my music in terms of engraving it.

PDA Showdown: Palm T|X vs HP iPaq rx1950

2008-06-16T00:00:00-06:00

Ok, I’ve been using my Palm T|X for a while after switching from an iPaq rx1950, and I have to say “wow”. What a huge difference!

My old iPaq was always a decent PDA that had all the things I needed (calendar, contacts, todo list, notes), but it also came with a lot of quirks. The biggest one was synchronization, which is heavily based on Outlook, and so I never got working under either Linux or Windows. The second was the very quiet alarm sounds on it, which I never was able to hear when it was in my pocket, or if I was asleep.

Then came the odd behaviors it had. The WiFi would turn on the system randomly, even if I didn’t set it to automatically download e-mail for me, causing the battery to drain quickly. This forced me to run it with WiFi turned off almost constantly. But the really odd one was how it would constantly emit event alerts and exhibit incredible slowdown unless I went into the diagnostics test and tested the event notifications (which would clear out some sort of internal cache, I’m guessing). Incredibly annoying since I needed to do this every time I reset the device. But in the end, the lack of WPA with AES for WiFi is what got to me, along with the desperate need for synchronization with a calendar.

My Palm T|X has given me a much better experience overall. It’s interface is much more responsive, and it generally gives me more options with its programs. What’s really surprising is that its handwriting recognition is very easy to use and has given me almost no problems whatsoever. But the best part is that it’s capable of synchronizing with Linux without a hitch (currently I have all my data inside of Kontact, which can import iCal data from Google Calendar just fine). If there’s one thing I don’t like it the lack of a file browser, but that’s not a big deal given what a PDA is to be used for.

So yeah, I think I’d highly recommend a Palm PDA over any Pocket PC based one.

libtodolister 3 - FAIL

2008-06-05T00:00:00-06:00

Live and learn, that’s what I’m taking away from this. Last September, I had the idea on how to add a huge performance boost to libtodolister. Previously, the TodoList class acted exactly like a list: it held items which had certain properties, and queries could be performed to extract only certain items. The performance boost changed this to where the TodoList class contained four separate, hidden containers (TodoItemContainers) which acted as sorting bins for the actual items. This meant that running a query such as “return only the future items” resulted in a single operation, not a whole bunch spread over a loop; constant time versus linear time. The problem was this change was not transparent.

Adding or removing items means figuring out which container to work with before actually performing the operation. This has been minor so far, but adding this whole “active versus inactive” items concept these past few days has shown that the decision not to make this transparent was a major design flaw. Not so much because of the minor nuisance of adding an extra function call here or there, but rather because many operations heavily depend on item indicies.

Here’s a basic algorithm for editing an item in the GUI:

Obtain the original item and copy its current contents to the GUI for editing
If the user selects "cancel", ignore any changes made and simply return.
If the user hits save, copy the edits made from the GUI back into the item.
Remove the item from its current container
Put the item into its new container
Save the TodoList and return

Above, the whole “remove item from container, put it in a new container” thing is to ensure that items with changed start dates, due dates, and flags are internally stored correctly. However, you do not necessarily know if you have a reference to a temporary TodoItemContainer object, a real one inside a real TodoList object, or a real one inside a temporary TodoList object, and the nature of the performance boost makes copies a necessary evil. When I added the whole “active vs. inactive” concept, it only exacerbated this necessary evil. To the point that TodoLister and TodoLister-GUI are being held together with duck tape and glue internally.

So, where to go from here. The boost in performance is something I definitely want to keep, but I need to modify it so that the API becomes much cleaner than it is currently. The TodoList class should have its own Add(), Remove(), Replace(), and Delete() methods, and the programmer using the API should never have to touch the TodoItemContainer class directly (if it’s even still needed). Accessing items should behave just like a normal list: sequentially and logically.

I might look into having the TodoList.At() method simply do some number fumbling, treating the internal container counts as list offsets. Query functions would make a return since they’re actually quite useful. Either way, this is required work since, without it, TodoLister-GUI is currently broken. Any fix I apply to make it would would be akin to duct tape and super glue.

Wow, An Update on Foxy!

2008-05-31T00:00:00-06:00

Remember Foxy? Probably not, so let explain. Foxy is my first attempt at creating my own programming language and interpreter, with practically no previous training or experience in doing so. The language is a Lisp-like language (so it uses S-Expressions) with an interpreter implemented in C#. It’s designed to be easily embeddable in other .NET programs as a lightweight scripting language, while providing an extreme minimum number of internal functions. For simplicity (remember, this is my first attempt at doing this), writing new functions for Foxy is done by implementing them in C#; there is no ”defun” function yet.

A while back I had a very basic test working, but then I added the TYPE-OF function and realized that the method I had come up with for defining functions was broken. After modifying function definitions a couple of times, I finally settled on a method where methods would be tagged with a specific attribute, which in turn would take internal “Foxy” parameter information. The actual method would then be expected to have a matching signature in C#. For quite a while I had hunted for some bugs in the parameter checking (do the internal parameters match the signature?), parameter generation (given a list of objects, make a list of parameters for the C# function, taking into account ignored or optional parameters in the Foxy function’s signature), and function invocation (given these generated parameters, execute this function and return the result). But I’m happy to say that as of tonight, I’ve finally squashed them all (as far as I can tell).

The current code can be downloaded from my Nightly Builds website. I’ve gone ahead and run these builds early tonight to coincide with this post. Additionally, the sample program I tested Foxy with (which is also in the Nightly Builds tarball) is below after the jump.

If you do download the code, a word of warning: I took no attempts to make this code pretty, optimal, or elegant. Most of it was written on the fly with little planning, so it’s basically a mess. I’ll work on cleaning it up for it’s initial release, but for now you’ll have to deal with it ;^)

 
;; Test file 
 
;; Declare some variables 
(decl *global-var* 42) 
(decl *global-var2* 42) 
 
(if t (print "You should see this right off the bat")) 
 
(progn 
  (decl test-var t) 
  (print "Checking IF statement: " :no-newline t) 
  (if t 
      (print "Pass") 
      (print "Fail")) 
  (print "Checking IF statement with false condition: " :no-newline t) 
  (if nil 
      (print "Fail") 
      (print "Pass")) 
  (print "Checking IF statement without second expression: " :no-newline t) 
  (if t 
      (print "Pass")) 
  (print "Checking IF statement with variable: " :no-newline t) 
  (if test-var 
      (print "Pass") 
      (print "Fail"))) 
 
(progn 
  (decl *global-var3* 42) 
  (progn 
    (decl *global-var4* 42) ; Should not exist after this progn block... 
    (decl hello "This is the value of the variable, 
                complete with line!") 
    (decl num-reps 4) 
    (print "Variable \"hello\" declared inside progn (should be true)? " :no-newline t) 
    (print (decl-p "hello")) 
    (print "Variable \"num-reps\" declared inside progn (should be true)? " :no-newline t) 
    (print (decl-p "num-reps")) 
    (print "\"hello\" equals: " :no-newline t) 
    (print hello) 
    (print "\"num-reps\" equals (should be 4): " :no-newline t) 
    (print num-reps)) 
  (print "hello declared outside progn (should be false)? " :no-newline t) 
  (print (decl-p "hello")) 
  (print-variable-tables)  ;Let's see the variable tables. 
  (progn 
    (print "Hello, world!") 
    (print "The type of \":THIS-IS-A-KEYWORD\" is: " :no-newline t) 
    (print (type-of :THIS-IS-A-KEYWORD)) 
    (print "Does \":THIS-IS-A-KEYWORD\" have a value? " :no-newline t) 
    (print (keyword-has-value-p :THIS-IS-A-KEYWORD)) 
    (print "Does \":THIS-IS-A-KEYWORD t\" have a value? " :no-newline t) 
    (print (keyword-has-value-p :THIS-IS-A-KEYWORD t)) 
    (print "Value of \":THIS-IS-A-KEYWORD 27\": " :no-newline t) 
    (print (get-keyword-value :THIS-IS-A-KEYWORD 27)) 
    (type-of "After type-of test") 
    :test 
    (print "gonna exit now") 
    (exit)) 
  (print "You shouldn't see this.")) ; Just to be safe. 
 
;; Just to be safe.  Again. 
(print "Or this.")

Staying On Top of J-Pop

2008-05-30T00:00:00-06:00

Wow, I really have to find a way to stay on top of the Japanese music scene better. Somehow the fact that Utada Hikaru, my favorite J-Pop artist, had a new album released in March slipped by me.

Any good links to sites that can help me keep current with what’s happening over there?

New TodoLister Vocabulary and Concepts

2008-05-30T00:00:00-06:00

Not quite satisfied with the evolutionary jump that the next version of TodoLister will make, I’ve been thinking of areas where it could be enhanced as of late. Tonight, I had a new idea hit me that I’m really excited to have since it’ll help give more credibility to it’s different way of handing something resembling a calendar.

Previous versions were built around the concept that TodoItems were either Current (they’re going on right now), Overdue (they’re going on right now but should have been completed already), Future (they aren’t going on yet), or Deleted (sent to the trash can). With start dates and due dates, TodoItems had a way to be Current only for a certain amount of time. In this sense, TodoItems are very much like the events on a calendar. However, the idea of being Current without start or due dates, and without having the concept of an end date, also helped out by giving TodoLister a way to remind users of day-to-day or “fuzzy” things. Effectively, it’s a sort of smart sticky note when it’s used like this.

One thing TodoLister has lacked has been the equivalent of more complex reoccurring items, such as an item that occurs every week but only on Mondays. To fix this, I have created the concept of being Active. It works the same way, where an item could be active only on Tuesdays and Thursdays. If it has no start or due date, it’s now the equivalent of a reoccurring item. However, that isn’t to say such TodoItems can’t have start or due dates; they can. This can be useful if you don’t want to be nagged about an overdue item that’s only active on the weekend (maybe it’s a weekend project that doesn’t apply to work).

This is all implemented in the next version of the library, but shouldn’t break backwards compatibility (except for the file format). I still need to add support for these in the old list importers, as well as in the TodoLister and TodoLister-GUI programs themselves, but it’s at least there and working. By default, new items are active on all days (which effectively replicates the behavior of older TodoLister versions).

My Newish Alpine Deck

2008-05-26T00:00:00-06:00

I think it was about two months ago when I bought my new stereo deck, an Alpine CDA-9886. It’s a nice unit that allows me to hook both an iPod into it (useless for me since I use Rockbox, but useful when others are in the car) and any external USB storage device, such as a hard drive or USB stick. So far I’ve used three USB devices with it: my 30gb iPod (acting like a disk), my 40gb external hard drive, and my 8gb USB stick. While the first two give me a huge amount of storage, I’ve found that my smaller 8gb stick is actually the best thing to use with it.

I think it’s because my 8gb stick is faster, both in the speed it can access files and the fact that I can browse to what I want to listen to more quickly. Think about it, if I use my 40gb drive and it’s mostly full, and if I want to listen to some Kraftwerk but I’m 100+ folders away, then it’s going to take a long amount of time to scroll over to them. My iPod made it worse because, as a disk, it has incredibly slow access times. On top of all of this, the deck scans through all the folders on the device prior to starting any music, probably so that it can locate all the playable files. On the 8gb stick, this only takes about five to eight seconds. On my 40gb drive it takes about 15 to 20 seconds. But on my iPod, it takes about a full minute, if not longer.

Another strange thing has happened ever since I got this radio. At first, when I was trying to decide what to put on my stick, I copied over individual folders. This got to be unorganized, so I instead copied over parent folders instead and then removed things I didn’t want. In other words, rather than copying “Minimum-Maximum CD 2”, I just copied my entire Kraftwerk folder and then removed stuff I rarely listen to. This has given me two things: a portable library of music that I can almost always find something to listen to from, and the rediscovery of music I forgot I even had. A good example of the latter is DJ Baby Anne. I had listened to her quite a bit back in 2002 and 2003, but somehow forgot about the CDs I had.

The real kicker is my subwoofers, which I just got reinstalled into my newer car (my ‘04 Outback). These are two Pioneer 10” 350w subs in a bass reflex enclosure powered by an Alpine amp. Nothing special, but still pretty good sounding. In my old car (my ‘92 Legacy) they mostly accentuated the punchy bass frequencies, around 110-150 Hz, making them good for techno but not as good for the deeper bass found in other music (like some of Kraftwerk’s music). But somehow– and I’m thinking it’s mostly because of the deck and higher quality speakers –the subs actually give me more bass in my new car than in my old one. Not only that, but they give me some really good, deep bass as well.

I think this deck has been one of the best buys I’ve had in quite a while. It just keeps surprising me with things.

Woohoo! TodoLister compiling on Windows again!

2008-05-09T00:00:00-06:00

Remember how a few days ago I ditched Autotools and went back to straight Makefiles? Well, while I was messing around with Qt 4.4 on my Windows machine the other night, the installer had given me a minimal installation of MinGW, which includes GNU Make. After installing Mono (since I didn’t actually have a C# compiler on my machine), I did a quick test, made some small adjustments to my Makefiles, and wouldn’t you know it…TodoLister was compiling (and working, including TodoLister-GUI) easily under Windows again!

While I used Autotools, this wouldn’t have happened. To get it compiling under Windows while that was being used, I would have had to install many more utilities or even a Unix-like environment, such as Cygwin. Now, at most a user would need to install a minimal amount of MinGW (basically just GNU Make…heck, if you can get Make on its own, use that) and Mono, and at least just the minimal MinGW if they already have a C# compiler (though TodoLister-GUI wouldn’t compile, no Gtk#).

What’s even cooler is that the new Gtk# distributed with Mono under Windows actually works really, and even gives me a working tray icon.

The changes to the Makefiles aren’t in Subversion yet, but will be within the hour while I double check some things.

On a related note…I might ditch the name “Honeypot” and go back to just TodoLister. I found out today that it also has some dirty connotations with it (it’s a slang term as well, turns out).

Qt 4.4 Released

2008-05-07T00:00:00-06:00

The past few times I’ve tried learning Qt I’ve generally been dissatisfied. That doesn’t, however, mean I think that it’s a bad toolkit. It’s wonderfully powerful, and it provides a huge library of useful things. Qt 4.4 just got released, and Ars Technica has a good in-depth look at it if you’d like to check it out. What’s cool about this release is that it now includes Phonon.

When I say I’ve generally been dissatisfied, I mean with the experience of using it. I think that thus far this is because I’ve 1) approached the toolkit wrong, and 2) don’t like C++ that well anymore. It’s sort of ironic, given my love for C#, that C++ has fallen out of favor with me for good ol’ C, but tonight I came across a comment on Slashdot that has given me reason to go back and look at C++/Qt again:

Java / .NET descided like Trolltech that C++ was too complicated. Sun created the java language, MS the C#, Trolltech just decided to limit themselves to a subset of C++ and add some extensions via macros (and a precompiler which generates the boilerplates) but globally the aproach is similar.

Before, I saw Qt as being simply a library to use with C++, not a subset of the language. But taking this into account, Qt actually makes more sense now. Rather than learn how to use Qt with C++, I simply learn Qt and think of it as being a C++-like language.

There are a couple other annoyances I have with Qt that mostly stem from what I’m already used to with GTK/GLib/the CLI. Learning a new framework takes time, and lots of it. But as I need to create more cross-platform programs, both for fun and for work, I’m finding that I need something better than what I use now.

Maybe I’ll take another look at Qt, but use it with a fresh project idea. TodoLister is pretty set with .NET, and my work stuff is also pretty set. Ideas?

Media Players, All Platforms

2008-05-06T00:00:00-06:00

I’ve got to admit, I used the old WinAmp v2.5 for a very long time when it came to listening to music on my computer. That’s actually pretty much all I used it for, too. CD burning was, and still is, always delegated to Nero (or InfraRecorder these days), CD ripping is always done with CDex, and mp3 encoding is always done with LAME. When I added Ogg and FLAC into the mix, I used the command line tools. So WinAmp has always been only to play music for me. This still holds true even as I use WinAmp 5 (though still with the old default 2.5 skin).

What’s funny is that I don’t use media players for the organization of my music at all since I’ve generally found that embedded tags don’t always produce the results I want. Instead, I’ve had a system where I organize music by genre, sub-genre, sub-sub genre if needed, artist or DJ, album and CD number all with folders, then finally the files. In some instances, there is no single artist and no DJ, but instead music is organized by something else. A good example is the music I download off of OverClocked ReMix. Here, my organization generally goes Music/Games/[system]/[game name]/OC Remixes/[files]. This organization is the exact reason why I use Rockbox on my iPod, since I can browse for files and folders rather than use playlists or tags for organization.

Given all this, I thought I’d share my thoughts in general about different media players that I’ve used (though this certainly isn’t all of them). Now that I’ve explained my organizational style, I’ll say right now that any music player that does not give me the ability to easily handle things by file/folder doesn’t get used. So I’m going to attempt to ignore this as much as possible while I get my opinions out so that they aren’t quite as biased. Got that? Cool. Here some of the media players I’ve used:

Winamp 2.5 and 5.x: Ignoring my love for the old style skins and my dislike of the new style skins, Winamp has always been a very good player. It has handled most file formats I've thrown at it (ogg vorbis, FLAC after I installed the codec, mp3, tracker files, midi files, etc.), and has generally handled them very well. The one part I've never really liked was its handling of MIDI files, which for some reason always felt sloppy. Maybe it's because the MIDI settings were hidden inside a plugin compared to other players (see the next one), and so I just had a bad start with it. But overall, Winamp really works well. Its interface is probably the easiest to use, yet doesn't require huge amounts of screen real estate. This is my first choice under Windows since Audacious doesn't exist on that platform as far as I know. I should point out, though, that I never use Winamp for anything but music playback. Other functions, including video playback, is delegated to other programs, mostly out of experience (others just work better, imo).
VanBasco's Karaoke Player v2.53: I'm only listing this one because, for straight-up MIDI listening, it wins all competition, hands down. It's old, but still runs on todays computers, and has some cool features. What I've always really liked about it is that I can watch the MIDI output in realtime, and its pretty easy to manage settings. The one downside is that it only plays MIDI files. Again, my first choice, at least for MIDIs, under Windows.
Apple iTunes: This one plays common file types (including MIDI), but tends to not handle most others, including Ogg Vorbis (except through a special QuickTime codec) and FLAC. Overall, I always feel as though iTunes takes up way too much screen real estate and is incredibly slow when resizing the window or scrolling, even on my quad core processor. It also has a larger install size than other players. The one saving grace used to be the iTunes Store, but now that I've found Amazon.com's store (and its DRM-free mp3 files), iTunes has fallen from whatever grace it once held with me. Generally, this one is my next to last choice.
Windows Media Player: No, just no. Good only for watching movie files. Last choice on any platform.
Audacious: This is a Linux-only player that looks and acts like Winamp. It's a fork of the Beep Media Player codebase, which was in turn a fork of the XMMS2 codebase...which in turn was a rewrite of the legendary XMMS player. Audacious actually works really, really well. Its interface is top-notch and Winamp like, so it takes up little space on the screen. It supports just about any file type through its plugins, and also has support for XMMS plugins (not sure if it supports all of them). Its settings window is actually easier to navigate than Winamp's, and much easier than XMMS's. If it weren't for it being Linux only, this would be the only one I use.
XMMS: Absolutely amazing player for all the same reasons Audacious is. It's just older and has a difficult settings window.
Rhythmbox: This is an iTunes-like player for Linux. Unlike iTunes, this one is very responsive, and supports a bazillion file types using the GStreamer framework. But like iTunes, it needs a lot of space on the screen because of its layout. I was actually using this one for a while, during a time when I tried getting used to using ID3 tags for organization. It worked great...but I never got used to the organization style. This is my second choice under Linux, and my third overall.
Amarok: I tried liking this one, I really did. It a Linux-only player (for now) that has awesome support for file types, great performance, and a really neat interface. So why don't I use it more? Its interface still takes up a lot of space, and as neat as it is, it also tries to guide you towards its own view as to how a media player should function: playlists. They have a file browser, but I've always felt they pushed playlists more. Still, it's a really good player that's at the same level as Rhythmbox, so I'd probably put this one fourth. It should be available for Windows soon as well, though I'm not too sure about the new 2.0 interface yet.
mpg321/mplayer CLI: These two I'm lumping together since they're command line driven. Sure, mplayer has a GUI, but I've never used it. mpg321 (and mpg123) are Linux only, but mplayer is cross platform. Overall mplayer a lot of different file types, while mpg321/123 plays only mp3 files. Don't expect much usability unless you're only planning on listening to a single file or a couple of specific ones. What these are great for, though, is checking out a single file you just downloaded, or listening to music while working at a text-only terminal. They're also great for scripts, and I actually use mplayer to listen to internet radio stations pretty much all the time.

So there ya go, my two cents on media players. Notice that I didn’t really mention video all that much. That’s because I’ve noticed little difference in the handling of video files between players, except for Winamp, where videos didn’t feel right in that setting. For those…just stick to (G)Xine or Totem under Linux, and Windows Media Player under Windows. They get the job done and generally don’t have problems.

Overall, for Windows, I pretty much only recommend Winamp, except for MIDI files. Under Linux, I recommend Audacious first, Rhythmbox second, and probably XMMS third if you can get hold of it, otherwise another XMMS clone or Amarok. Stay away from iTunes if you can, and especially Windows Media Player. If you have the choice to use either platform, go for Audacious first, then Winamp, then Rhythmbox.

Funky Cross-Platform GUI Programming with .NET

2008-05-01T00:00:00-06:00

I wasn’t quite expecting this. I’ve been working on some software for work for a while now which is designed to be cross-platform like Honeypot (.NET, Linux and Windows out of the box, uses a GUI). The one thing that set it apart was that it only had to run on a small number of computers, all of which I’m able to install additional libraries on, so I essentially had no restrictions on the GUI toolkit I could use. Given this, my natural inclination was to use Gtk#, which in the end works fine and gets the job done.

The problem came when I upgraded my laptop (where I do my programming) to Ubuntu 8.04. This new release uses Gtk# 2.12, while the latest version of Mono for Windows comes with 2.10. Long story short, I didn’t want to bother having a second version of Gtk# lying around, didn’t want to compile a second time on a virtual machine, and have been looking for a good cross-platform GUI solution anyways, so I decided to consider alternatives.

Most of the time we think of the GUI being both the front end with the actual heavy lifting being done in the background, as well as the program’s manager. One alternative idea I had was to have both the heavy lifting done in the foreground with the GUI done in the background. This proved to be logically unsound (kinda hard to have a GUI not know about the data and want to do something to it), so I made a slight tweak.

Make the GUI a swappable “middle ground”.

What I have now is an executable that does program invocation, which constitutes the front end, and a library to do heavy lifting, which constitutes a backend. Sharing this library are additional DLL files, each of which defines a GUI. One class inside these GUI DLL files inherits from an interface, IGuiController, and has an attribute, GuiController, to mark it as the GUI entry point. The main program simply figures out which DLL file to load at runtime, finds the correct class, instantiates it, and calls the methods defined inside the IGuiController interface to get it running.

This may seem like it would introduce a lot of duplicated code, and therefore be a bad idea, but the truth is there isn’t much code duplication. Gtk# and Winforms are similar, and are both lightweight enough that not much code goes behind them in this particular program, but are also different enough that they require different methods of programming. In Winforms, for example, you usually make a class that directly inherits System.Windows.Forms.Form to make a window. This is possible under Gtk# as well, but the way I’ve usually seen it done is that you define a wrapper/driver class that holds the Gtk.Window object obtained from Glade, and then use the wrapper class instead. Timers are also different, with Winforms having a separate Timer object and Gtk# needing to use the GLib.Timeout object’s static Add method with a GLib.TimeoutHandler delegate. Don’t even get me started on lists and trees ;^)

In the end, I think I found a good way to achieve the goal of “out of the box cross-platform compatibility over Windows and Linux without crazy additional requirements and an easy install” for .NET programs. This might be what TodoLister-GUI needs to properly run on Linux and Windows without the need for Gtk# on the latter.

Simplifying Honeypot

2008-04-30T00:00:00-06:00

I’m under the belief that this still lies within the plans I’ve made for Honeypot. Today I completely threw out its reliance on Autotools and went back to straight-up Makefiles. I’ve come to see the system as being horribly complex and a bit unwieldy for this size of project. It has its place, but I think that just using Makefiles is a better idea.

On top of this, everything is now a single source package once again. That means that libhoneypot, libtodolister, todolister, and todolister-gui are all considered a single collection of modules. But it’s not quite like it was before. The difference is that it’ll actually be modular in the sense that you can compile just the libraries, just the CLI version, or just the GUI version. Or multiple ones.

New Releases of libtodolister, TodoLister, and TodoLister-GUI

2008-04-15T00:00:00-06:00

I think I had mentioned this a couple of days ago, but I just got around to uploading them.

libtodolister has been updated to v3.0.1, which is mostly considered a bug fix release. Fixes include BOM characters being prepended onto the XML files it makes (which confuses some parsers, like S-XML), some exception fixes, and date handling fixes. The one improvement it has is that the constants TLConstants.DefaultTodoListFile and TLConstants.UsersTodoListerDir now support mobile platforms.

TodoLister is now at v3.2.1, and is also a bug fix release. Fixes for it include command line argument handling, and fixing a crash when no TodoList file is found and the user does not want to create one.

TodoLister-GUI actually has a new overall release, v0.2. Changes and fixes include a slightly improved OSD window, fixes to the tool tips on the tray icon, synchronization fixes, hash codes get computed when adding items, possible crashes when the program is run for the first time, and lock file fixes for multiple platforms.

The actual lists of changes, as well as file downloads, are linked from my homepage: http://www.partition36.com/~personal/index.html

In terms of plans for the immediate future (i.e., the next release), these include:

Installing libtodolister and libhoneypot into the GAC
Easier handling of TodoItem and TodoList objects through a new "AggregatedTodoList" object
First release of PyHoneypot for Eee PCs
Bypassing of hashes in TodoList files (needed for PyHoneypot)
Hash codes being computed with MD5 instead of SHA256 for wider compatibility (we're not doing cryptography here)
Improved command line arguments for TodoLister that will make it easier to handle items
Ability to add an item with TodoLister solely on the command line (no need to use the menus, useful for scripts)
Enhancements to the TodoLister-GUI GUI

Some of these are already completed and in Subversion.

Eee PC Usage and A Small Rant

2008-04-11T00:00:00-06:00

I suddenly noticed tonight that my Eee PC has become rather integrated in my day-to-day work. Used to, when I got home from work, I would generally spend my time equally between my laptop and my desktop. But these days I keep my Eee running in my room, where it has slowly replaced my laptop. Now that I have the Xandros repositories installed, as well as some development tools such as Emacs and Mono, it is more like half of a desktop replacement.

There’s only one thing, which isn’t exclusive to my Eee PC but rather to LCD monitors in general, that still annoys me. Widescreen LCDs have been more of a pain for me than a perk. 4:3 resolutions (or 5:4, which is 1280x1024) are much more conducive to surfing, editing, and programming. In fact, the only reason I’d want a widescreen resolution is for a movie, which I never watch on a computer in the first place.

People who don’t agree with me should try using Emacs with a few terminal windows open to watch screen output sometime.

I gave in

2008-04-10T00:00:00-06:00

My desire to use LaTeX on my Eee PC, and my actual need to use Subversion on it, has finally forced me to install some additional Xandros repositories on my Eee PC. What I’ve decided to do is carefully watch the required dependencies when I install something. As long as a Python component or GTK+ doesn’t get updated, I should be able to work on PyHoneypot just fine without worry of it not working on a stock Eee PC. Besides, I may be able to get my boss to help me test it on her stock unit.

So tonight I installed LaTeX, Beamer, xpdf, AUCTeX for Emacs, Subversion, reinstalled Emacs, and removed the overly bloated Adobe Reader. I also installed more RAM in my system, bringing it up to a full gig. But already I’m wondering if I should have gotten two gigs instead, since I’m using 700mb of that already.

The Incredibly Versatile (and fun) Eee PC

2008-04-04T00:00:00-06:00

It occurred to me tonight that I haven’t actually written very much about my Eee PC yet. Last night I mentioned how I installed vpnc onto it without adding additional repositories, and I did my mandatory “first impressions” when I first got it, but not much else.

So what else have I been doing with it? Well, keeping in mind that I’ve tried to keep it pretty close to a stock Eee PC, I’ve:

Installed GNU Emacs 21
Installed SBCL
Installed SLIME
Enabled the IceWM "start" menu and task bar
Customized my simpluirc file to give me a more customized menu

Nothing too spectacular, but getting SBCL working was pretty fun. Given that GNU Make is not installed by default (and you need a Debian repo to install it), I got away with the install by installing it on another system. Basically, on the second computer, I installed into its own directory (~/sbcl), then copied it using scp to my Eee PC. All I needed to do then was add a line to my .bashrc that set the SBCL_HOME environment variable to the proper folder (which in my case was /home/user/sbcl/lib/sbcl).

Hopefully I can keep this up and get some other cool things running on it. I keep wanting to try ioquake3, but my last attempt failed. Maybe I’ll try again this weekend.

Installing vpnc on an Eee PC - The Easier Way

2008-04-03T00:00:00-06:00

In order to get online through the wireless network at work, I need to connect to a Cisco VPN. The Cisco VPN client that’s provided is hopelessly out of date and half the time won’t compile or install. So instead, I’ve used vpnc for the past two years to connect.

My Eee PC has been really fun to use wireless, but because of the VPN at work, I haven’t had a chance to try it there. Installing vpnc on it usually means adding some Debian repositories, but since I’m wanting to develop apps that work on stock Eee PCs, I’m very hesitant about doing this. What people don’t always know is that for simple installs of software that have almost no dependencies, you can just download the deb files and use dpkg manually. It may be a bit dangerous, and updates aren’t automatic, but it hasn’t harmed me yet either. In fact, this is how I’ve gotten vpnc installed without adding any additional repositories.

The first step is to download the i386 deb files from Debian’s repositories by hand. That means pointing your browser to http://ftp.debian.org/debian/pool/main/v/vpnc/, and downloading vpnc_0.3.3+SVN20051028-3_i386.deb to your Eee PC. Don’t use the newer version since it depends on a newer libc package, and that’s one we do NOT want to touch.

Next, we need to get libgcrypt11. This is done the same way by going to http://ftp.debian.org/debian/pool/main/libg/libgcrypt11/ and grabbing libgcrypt11_1.2.3-2_i386.deb. Again, newer versions require a higher libc version and we don’t want to mess with that.

All that’s left is opening up a terminal (Ctrl+Alt+t), and installing each of these by running ”sudo dpkg -i <file name>” for libgcrypt11 followed by vpnc. Well, actually you then need to configure vpnc, but that’s not hard at all.

I don’t make any warranties that this won’t screw up your system, so do it at your own risk. All I can say is that it worked fine for me and I was able to successfully connect just fine.

Honeypot's Plan Revisited

2008-04-03T00:00:00-06:00

After doing some rethinking, and looking into why exactly I keep thinking something is “wrong” with Honeypot, I’ve arrived at a conclusion: it’s straying away from its plan.

TodoLister is, and always has been, a command line program for UNIX-like environments first, and a fluffy GUI for X11/Windows users second. At its very heart is a lower-level to-do list notification facility that can be used within scripts, with pipes, interactively, and non-interactively. Deeper yet is libtodolister, a set of classes that allow people to construct lists of “things that need to be done” without the sometimes unwanted nature of a queue. What it’s turning into is what I’m going to start calling a “corporate solution.”

That term is meant to be pejorative, and should be taking to mean something that strives to be a “solution” in the buzzword sense, even though it has a different inherent nature. In simpler terms, TodoLister is meant for somewhat lower-level usage but I keep trying to shoehorn it into a business setting by making it a “solution” (I’m beginning to hate that term) for consumer’s problems.

I think I’m going to keep TodoLister-GUI and PyHoneypot about where they are in terms of features since they certainly fill a gap (I find them really useful sometimes). But my focus will continue to be on the command line program and library.

Tomorrow I’m releasing new versions of libtodolister, TodoLister, and TodoLister-GUI. The first two are mostly small bug fixes, while the latter has both fixes and enhancements. I’ll post about my ideas for TodoLister and libtodolister later, though.

Level Editing

2008-03-29T00:00:00-06:00

I’m not sure how long after I first encountered the game it was, but after I got introduced to Doom I quickly got interested in creating levels for it. Back then, the only editing tool I had at my disposal for levels was DEU, which actually remained my editor of choice until Doom Builder finally came out. Even today, I occasionally use Yadex (an X11/*nix port of DEU) to quench whatever thirst for Doom level designing may come up instead of Doom Builder.

The theory behind Doom maps is that a group of vertexes can be tied together to create lines, representing walls. These lines are then grouped into sectors, which must be closed shapes. Textures were applied to the sides of the lines. When two sectors share a line, it is considered a 2-sided line and may or may not have a texture on one or both sides. Sectors are used whenever difference in height, floor/ceiling texture, or z-height is required, or when a door or lift needs to be created. Here’s an example. The first picture shows the top-down view from within Doom Builder, while the second view shows the room from within the game. Objects in the game are represented as “things”, and include players, enemies, weapons, ammo, decorations, keys, and power-ups. Special things such as doors, switches, and lifts worked by assigning “specials” to lines and, when appropriate, assigning them a “tag” number so that they could control sectors with the same “tag” (this is how switches and trip-wires is done). Sector-based effects included flashing lights, damaging floors, and crushers.

That’s pretty much all there is to Doom mapping. After some practice, I eventually got to be pretty good at it (with that level being from sometime around October 2004). Once Quake hit the scene, things changed in terms of the theory of game maps. Lines, sectors, specials, and things were all gone and were instead replaced by brushes, entities, scripts, and “assets”. Levels were no longer specified in terms of valid playing space, but were actually defined as huge amounts of void with the brushes making up world geometry (floors, walls, pillars, bridges, etc.). Entities, which I usually thing as being roughly the equivalent of things, not only put objects into the game world but also affected the world and would sometimes be brushes themselves. Lighting was also done through entities.

I remember being really excited about mapping for Quake, but also remember that it was like hitting a brick wall in terms of difficulty. By then, Doom’s map theory had been so deeply ingrained in me that moving to true 3d was difficult. Still, I eventually understood the basics and could at least make rooms. Sadly, I have never finished a Quake map, though.

Unreal came out in 1998, the year after Quake 2 came out, and used a similar map concept. However, rather than having a bunch of empty space, the world was represented has a solid block of void which you carved out of. To my mind, this was much closer to how Doom worked, and so Unreal mapping came surprisingly easy to me, at least in terms of generating geometry. Once I had a copy of Unreal Tournament, I actually did finish a map, but never released it and eventually lost the file. However, I remember being confused and generally dumbfounded as to how I was supposed to make maps look good in the game. Perhaps it’s just that I hadn’t had enough experience with true 3d maps yet, but the overall feel of mapping was different.

When Unreal Tournament 2003 came out, I quickly purchased it and started to learn mapping for it. But here is where mapping turned from being fun for me to being not fun. Not because of difficulty, but because of how it was performed. Like the Unreal games before it (and to a large extent, the Quake games), UT2k3 had brushes and actors (entities), but unlike its predecessors, there was a huge reliance on what were called Static Meshes. In the first two Unreal games, Static Meshes were mostly used as decorations, weapons, and powerups, much like certain entities were used in the Quake games and things in Doom. But in UT2k3, they were used as geometry.

UT2k3 changed the theory of mapping, where brushes would be used to construct only the simple world geometry, and the rest of the level would be constructed using Static Meshes in much the same way a traditional level designer used textures. Want a wall? Add the mesh and copy it a bunch of times. It sounds awesome in theory, and in practice, but it just feels alien to me. Static Meshes places a need to have many more people working on a single map indirectly, since 3d modelers have to create world content. Levels, in other words, went from being “mine” to being “ours”, and so in my mind went from being “my artwork” to “our game assets”.

Games that use Unreal Engine 3, like Gears of War and Unreal Engine 3, rely almost entirely on static meshes these days. Even Doom 3 used less geometry and more meshes. I’m wondering if, in the future, levels will be designed entirely within 3d modeling software or CAD programs.

In the meantime, I’ve given up trying to create maps for 3d maps in the post-Quake world. Instead, I think I’ll dust off some old Doom maps of mine and prep them for release.

I love sleeping on things...

2008-03-28T00:00:00-06:00

…because I sometimes wake up with a better outlook.

Ignore the previous thoughts on Honeypot. The original plan of a primary version in C# with the Python Eee PC version secondary is still in effect and will continue to be.

Correcting an Error: The Goals of Honeypot

2008-03-26T00:00:00-06:00

I’ve written before (though not in this specific blog) about how Honeypot’s/TodoLister’s overall project organization is messy. So when I first thought up a new scheme tonight for how the pieces fit together, I also thought “am I just beating a dead horse with this?” as well. But the more I think about it, this latest idea isn’t so much a radically new one as much as it’s the further extension and improvement of what I’ve already started.

Time for a short history lesson. 2005, along comes “Todo-Lister”, a single command line program written in C++. It does exactly what I need it to, but only under Linux. I need a Windows version, so I write a GUI version from the ground up in C# and WinForms. It does exactly what I need, but only runs under Windows. Then comes the need for a GUI under Linux and a command line interface (CLI) under Windows. That’s when TodoLister 2.x comes about along with LibTodoLister. The GUI is a heavily modified (almost ground-up) version of the Todo-Lister for Windows, and uses Gtk# for a GUI. The corresponding CLI program is a ground-up rewrite. Then I merge the GUI and CLI programs into one, make some stupid “plugin” library and a standard set of plugins, and release the awful TodoLister 2.1 and 2.2.

Around this time I realize what I’ve done, re-split the GUI and CLI, continue working only on the CLI, and release it and libtodolister as TodoLister 3.0 with a promise of a GUI “soon”. The CLI and library both get an update or two, and in the meantime I decide I want to expand the overall outlook of the program to include other PIM-related features. The collection of programs is renamed Honeypot, TodoLister CLI and libtodolister get new releases, a new centralized library for all the “components”, libhoneypot, gets a release, and the GUI finally does too…under the name TodoLister GUI (it only handles TodoLists). The plan is to rename the GUI to Honeypot once there’s an additional component. I also plan on combining libhoneypot and libtodolister into a single library/API to eliminate over-modularization. This effectively gives me one GUI, a CLI for each component, and a single centralized library.

But that’s the future. Where we sit today is: two libraries (libtodolister, libhoneypot), a CLI program (todolister), and a GUI (todolister-gui), all of which are collectively called “Honeypot”. PyHoneypot, a 100% new GUI written in Python, will join this bunch to provide support for the Asus Eee PC by the end of April. But the question is, what does this mean?

PyHoneypot only shares file specs with the other programs, nothing more. By doing so, it also reimplements part of the libtodolister API, which is a bad thing since that’s not good code reuse. Simply making bindings isn’t an option since A) I know of no way to do so between .NET and Python except for IronPython, and B) the Eee PC has no .NET support to begin with. Python was the best choice since it’s easy and cross-platform.

And there’s that word again, cross-platform. It’s what .NET promised me and only delivered halfway. CLI apps work pretty well under both Linux and Windows, but the GUI (Gtk# or WinForms) has very annoying quirks that requires double-programming and platform detection to make programs work right. I eventually took these quirks out of TodoLister-GUI and made it Linux-only…only to add a few quirks back once I figured out how to get Gtk# 2.10 working under Windows.

But yet, expecting users to install Gtk# 2.10 for Windows isn’t feasible. There is no official release of it from the developers, only from a 3rd party that needed it themselves. It works, but I’d prefer not to use it just to keep it simple end users.

But for all these quirks, none addressed the problem of nonexistance. A stock Eee PC simply does not have .NET/Mono/DotGNU installed and the average person I foresee being a Eee PC user wil not want to go through the hassle of installing alternate repositories.

Joel of Joel on Software makes a very good point when he says that rewriting a program from scratch is something that you should never do. I believe in it entirely. But the way I see it, I’m faced with an unusual predicament. I have a working implementation that does not fulfill my goal of cross-platform compatibility nearly “out of the box”. The reason for this goal not being realized is that I chose a language and platform that I thought would fill the gap but didn’t in the end.

So for a moment, I’d like to lay out my goals as a programmer for Honeypot:

Out-of-the-box Cross-platform compatibility for Ubuntu/Debian Linux, Windows, the Asus Eee PC's Linux, and Mac OS X.
A minimal amount of "quirks" in the code to achieve this cross-platform compatibility, taking into account restraints on each platform
An expectation that the user will have to install, at most, one 3rd party package for platform support.
Standardized file specifications for the users' Honeypot file, preferences file (GUI and CLI), and any other files that may come up
A single GUI that looks basically the same on all platforms, minus the look-and-feel of the widgets.
A single GUI for all components, but a separate CLI for each component
A todo list component (TodoLister), a contacts list component, and a journal component

Given these, .NET fails at the first and third. What I need is a platform/language available on all of my selected platforms, and one that’s not a pain to install. In hindsight, .NET was a lousy choice for these goals (which were always present, just not written out). What I need to do is rewrite it in Python. Joel is definitely correct in that you shouldn’t rewrite code, but I think this is a grey area since the language itself is acting as a barrier.

So, time for the nice juicy bit of how I organize all of this. Step 1: finish PyHoneypot, make new release of it and the current .NET software. TodoLister-GUI will inherit PyHoneypot’s GUI (which is mostly identical, but improved). Step 2: formalize file specs in documents. Step 3: Begin rewriting libtodolister and libhoneypot in Python (using current PyHoneypot code) with the goal of having new components in the future. Step 4: rewrite the TodoLister CLI. Step 5: release and continue on with original plan, except under Python.

Running Linux on my PS3

2008-03-17T00:00:00-06:00

Today I installed Ubuntu Linux onto my PS3. Believe it or not, installing it is really simple. The only two things that remotely resembled a pain were backing up my existing data, and getting the resolution configured correctly. Most of this post is an overview of the process I used, but if you want, you can click here and jump to why I actually did this.

The first issue was actually saving what I had on my PS3 hard drive. Rock Band doesn’t allow the copying of its game data, which includes downloaded song packs, to USB devices. I went ahead and copied over most everything else to my external HD by hand, and then used the integrated backup utility Sony provides in hopes that it copied over my Rock Band data. If not, then I just lost six songs. After backing stuff up, I reformatted the hard drive and gave Linux a 10gb partition. The reason for only giving Linux 10gb is because I don’t plan on installing many apps for it, and most of the data (music, videos, etc.) I’ll be using with it will be stored on my file server here at home, so more space simply wasn’t needed.

After backing stuff up, I followed these directions to get Ubuntu installed and running.

Booting the system is a little slow, but I’m going to guess that’s due to both the hypervisor and the small amount of RAM the PS3 has (256mb, but only 224 is available since the XMB reserves 32 of that for itself). Once I booted it for the first time, I was greeted with a very low-res (640x480) login screen. That’s when I followed the directions here, which mostly worked. The only I did differently was I used the ps3videomode -h command to simply guess the mode that I needed. Then I ran the command ps3videomode -v 0 to autodetect my resolution, then ran ps3videomode by itself to find out which mode it liked, which turned out to be 35 (RGB 720p @ 60Hz) and matched my guess. After I created /etc/event.d/ps3videomode according to the instructions, I rebooted and things worked OK.

The default install works well, but runs sort of slow (again, because of the memory). On the psubuntu page, they recommend using xubuntu, which is based on the XFCE desktop instead of Gnome. This is great, and I love XFCE, but Fluxbox is even less resource intensive, and is probably my second or third favorite desktop enviornment. So I installed that instead and then ran “sudo update-menus” manually in a virtual console to get my right-click menu working for it.

So…why the heck did I do this. Believe it or not, using it as a multimedia box was not my main intention. Earlier today, while messing around with my laptop and GKrellm, I learned how to remotely monitor other computers with it. Since the fileserver I have here at home isn’t exactly the newest box on the block, I sometimes worry about the health of the system, as well as how much strain I put on it. The addition of a 500mb RAID-1 array on it and the integrated tape backup drive helped calm some of my fears, but I still like to keep close tabs on the machine. That’s when the idea hit me: why not set up a dedicated monitoring computer in my room and let it run off my HDTV?

That’s the main reason for doing it, turn my PS3 into a computer that monitors my other computers and notifies me when something goes wrong. The fact that I can also use it as a media player to play the media off my server is simply an added bonus.

Do I really need a MacBook?

2008-03-13T00:00:00-06:00

I’ve been debating about getting a Mac laptop for at least a year now, but it’s been hard for me to actually nail down a good reason why. All of the Linux software I use, which makes up my day-to-day software, I’ve been very happy with. The only software I use under my Windows machine at home (aside from Firefox and Pidgin) is the music software I use, and that works fine. When it comes to programming, some of the tools I’ve based my work around are either the same on Mac or don’t work that well (Emacs and gcc are on Mac, but Mono with Gtk# and PyGTK both run sort of funky).

Last night as I was falling asleep, I thought about the whole thing, and most of what I do can be accomplished under Linux. For programming, I pretty much only use Emacs and makefiles. For surfing, Firefox. For e-mail, Evolution and Thunderbird. For music, Rhythmbox. The recent webcam/VoIP/videoconferencing stuff I’ve started tinkering in for the desktop also all runs under Linux for me. In the end, I think it would be too large of a paradigm shift for me to switch to Mac, and the specialized tools they have for it simply wouldn’t apply to what I do.

Rock Band - The Rest

2008-03-03T00:00:00-07:00

I’ve been playing Rock Band pretty much every day since I bought it last Tuesday, so I think it’s safe to say that I’m hooked. Since then, the guitar controller has grown on me. I still prefer the raised buttons of the Wii controller, but the size and weight of the Rock Band one is more comfortable for me now, and I’m able to perform the same fast runs that I could on the Wii.

The drums are also coming along quite well, but I’ve found that they aren’t quite as sensitive as I first thought they were, though they still work great. Most of my problems, however, stem from me having never actually played drums in school. Medium gives me no problems, but as soon as I bump up to hard difficulty, I start failing songs somewhat regularly. I did figure out how to hook them into my computer and use them as a MIDI controller, however, so maybe I can use that to practice. I’m also thinking I’ll post a video of how I did that on YouTube since I used a slightly different method than the ways that have been presented so far.

Song wise…there are some great ones in the game, and the songs available for download are also great (I grabbed the NIN and Metallica packs), but I still feel as though Guitar Hero 3 has the stronger song selection out of the box. In comparison to the ones in Rock Band, they feel more driving and less repetitive. However, one thing I do like in Rock Band is that the difficulty curve is much more smooth; there are no brick walls to hit during a solo tour.

Rock Band - First Impressions

2008-02-26T00:00:00-07:00

I finally caved in and bought the special edition of Rock Band for my PS3. Ultimately it was some of the songs that it contained that swayed me over (Don't Fear the Reaper, My Sharona, Highway Star, and a couple of Metallica tunes), though the fact that I could finally play drums also helped quite a bit.

I’ve been playing Guitar Hero 3 on my Wii quite a bit since I got it for Christmas, so I was already used to the basic gameplay. What I didn’t expect, however, was just how different the Wii guitar felt compared to the one for the PS3. I think I actually like the raised buttons of the Wii controller better, while I definitely like the much softer strum bar on the other. However, I’m not sure if I’m doing it wrong, but according to the manual I can “tilt the guitar skyward” to use my energy bar. I haven’t been able to do this yet, so I’m wondering if the batteries are just dying (I’m just using the pack-in ones), or if I’m doing it completely wrong. Either way, having to hit Select is horribly uncomfortable. Bottom line: I think I actually like the Wii controller better, even if it isn’t quite as accurate.

The drums are another issue, though. Being able to play drums is something I’ve wanted to do since high school, and Rock Band is definitely satisfying that desire quite nicely. Aside from the loud noises it makes, the controller is extremely responsive and natural feeling. They even ship real drum sticks with it (though they feel a bit rough…I’ll use my own instead). I was able to play through a song on Medium without much trouble, which surprised me.

I’ll have to see how Rock Band shapes up in the end. The song list doesn’t have the same opening energy that GH3 did and instead feels…simple. But I’m pretty sure it’ll pan out in the end and be just as good as GH3 is (even if they don’t have Hier Kommt Alex in it).

Computer Security - It's Hard

2008-02-22T00:00:00-07:00

For some reason, I like learning about computer security. The only problem is that I just don’t have the math skills to actually implement some of the cool things in the field. However, that hasn’t stopped me from doing what I can with the available tools.

When I took a computer security class at work, I found out about Bruce Schneier through his book Applied Cryptography. Since then, he’s become sort of an idol of mine. What follows was a cool interview he did at one time:

Serial IRC Experiments

2008-02-09T00:00:00-07:00

There’s an episode in Serial Experiments Lain (I think it’s either “Kids” or “Society”) where lain is going around searching for information on the Knights and just generally hanging out on the Wired. While doing this, there are small voice overs which are supposedly excerpts from spam and/or mail she’s hacked into and is reading, or chat rooms. It’s always been an interesting moment for me since it really reflects our Internet quite well.

Every once in a while, out of the blue, I get the idea that I want to log into IRC and join a channel. Usually this is just a big waste of time since I either can’t find a channel that looks interesting, or the ones that do don’t have anyone actually talking in them. Perhaps I just don’t get IRC, but these are the main reasons I’ve never gotten into it.

Tonight is one of those nights where I had that same idea, again out of the blue. While looking at the channel list, I couldn’t help but feel a little like Lain tonight. In fact, I could almost hear those same voice overs as I passed over channels such as “#jesus” with the description “Welcome to #jesus! Our purpose: to get us all closer to Jesus.”

Ironically, I’ve been idling inside a Lisp channel (#lisp) this whole time.

XML-RPC Library Woes

2008-02-07T00:00:00-07:00

Socket programming is something I’ve never been all that adept at. Back in my first year of college programming classes I was able to get some simple C++ sockets working (BSD Sockets), but socket programming in .NET has given me fits for a while. The problems always seemed to be with the connection itself, either keeping it alive or starting one. Maybe it’s because I haven’t done asynchronous socket handling, but my initial experiences were enough to keep me from trying it again. These days I could probably get sockets working, but I’m too lazy to come up with a decent protocol to use with them.

So when it comes time for me to write a program that actually needs to communicate over the network, my first choice is XML-RPC. It’s a simple protocol, but that’s exactly what I like about it. I’m not out there to write web services, and so if I just need to make a simple RPC call or transfer a small message, XML-RPC works just fine. There are 3rd party libraries for this protocol in the three languages I use most, C# (XML-RPC.NET), C (Xml-Rpc C), and Common Lisp (S-XML-RPC), plus my own library for C# (libsxmlrpc-sharp, which uses no sockets directly, but some higher-level classes for the HTTP connections), so it’s not like I don’t have choices with it, either.

However, I’m not sure what it is exactly, but getting a client written with one library to work with a server written in another is a total pain. XML-RPC.NET doesn’t like to communicate with Xml-Rpc C all the time, and almost never with S-XML-RPC, and vice versa with all of these XML-RPC.NET also doesn’t like to communicate with itself when the server runs on Linux, but I think that might be a .NET Remoting issue with Mono. My library usually works with S-XML-RPC, but has issues with XML-RPC.NET sometimes, and hasn’t been tested with Xml-Rpc C.

The issue is always the same, too: the client hangs thinking the server should respond, while the server thinks more data should be there or that the connection is already finished because it already sent back the response. Basically, a deadlock. The .NET libraries may be having this issue because of their XML writers, where using System.Text.Encoding.UTF8 appends junk at the beginning of the file, while System.Text.Encoding.Unicode doesn’t. But that doesn’t explain S-XML-RPC or Xml-Rpc C. New line issues? Differences in how sockets are handled? Bad content lengths in the HTTP headers? Small differences in Unicode handling? Not sure, but it’s driving me insane.

See, usually I just stick to one language for the client and server, but that’s not always a good idea or a possibility. Like with C, I could write a GUI client, but it would have to use a cross-platform GUI toolkit like GTK+, and if I need to run it in Windows, that’s a problem since I don’t have a good C compiler for Windows (and compiling GLib programs under Cygwin is a headache in and of itself). Writing a server in Common Lisp is a wonderful experience, but I don’t have the experience to write a GUI app in the language.

These problems all came to light while writing a very simple program. The client, which must work under Linux and Windows, inputs text into four fields (name, subject, another name, and comments), and the server stores this data in an XML file. The situation is such that there is no possibility for a web-based app or for the client to store the data itself. Hairy situation, but totally feasible. The program works when using XML-RPC.NET, except that the server (which is run using detachtty, since Mono has no real way of running daemons yet) stops listening and forgets it had a port open after about 10 minutes. Xml-Rpc C would work if I had the chance to do a Linux-only client. S-XML-RPC worked for a server, but no client worked with it. My library worked, except that it had problems sending messages that were too long (likely an HTTP issue).

I think I may go back and re-propose that the whole thing be done with PHP and be a simple web app instead. So, so much easier. Still, this won’t solve all my problems, ‘cause you see…I use XML-RPC elsewhere at work, too, and would really like to use it in more places. But until I figure out exactly what’s happening with these libraries, or fix my own, I don’t see that happening.

Maybe I’ll get better at socket programming and develop my own message-passing protocol instead. Or look into XMPP, which is getting popular.

Second Eee PC Impressions

2008-02-03T00:00:00-07:00

Spending some more time with the Eee PC, I’ve found out a couple cool things, mostly from accessing a terminal (uxterm) from Ctrl+Alt+T. Xandros is based off of Debian, and so it has all the familiar APT-* and dpkg-* tools. This is good since I’ve been wanting to deveop a couple of apps specifically for the Eee. Running dpkg-query, I found that Qt 3.3 and 4.3 were installed, as well as Gtk+ 2.8 and GLib 2.12, and so I already feel right at home with it. Running apt-get connects to Asus’s own APT repositories, which don’t seem to have Mono on them, so I may be stuck with C/C++ , but that’s not a big deal at all.

The tabbed interface is a program that Asus wrote called AsusLauncher, which lives in /opt/xandros/bin. I haven’t tried yet, but I think that the source code to it is in a package called “Eee PC 701 Source” on the official Eee PC website. The program itself uses an XML file and a bunch of PNGs to generate its menus, both of which are in /opt/xandros/share/AsusLauncher. I’ve already had success adding an item to the XML file (simpleui.rc) to give me direct access to KDE’s Konsole. Last night, on a forum, I found that someone had written a simplistic editor for it. I may try my own hand at this and see what I can come up with.

At first I figured that the tabbed interface was pretty much the entire GUI, but I did find a couple of other things with it. Holding Ctrl and hitting left or right moves to one of the four virtual desktops (not sure why these aren’t disabled…), which is easier to work with if you launch kpager. I found a file in ~/.AsusLauncher called launcherMode, which I’m guessing dictates how AsusLauncher behaves, though I’ll have to do some searching with Google to really understand it. Also, changes don’t take immediate effect when you modify the XML file that defines the menus, but you also don’t have to kill X (Ctrl+Backspace, which still works, though Ctrl+F* doesn’t) to get it to reload. Instead, going to Settings->Personalize and then changing the color scheme once reloads it. Given this, I’m wondering if there are command line options to AsusLauncher to get the program to reload. I’m also wondering if it can run on another distro…

The programs themselves are mostly KDE apps with a couple of others (Firefox, Thunderbird, and Pidgin aren’t KDE apps, for example). All of these simply have different names in most places (usually the About menu shows the real app). The cool thing is that they included Amarok as their “Music Manager” and Gwenview as their “Photo Manager”, both of which are extemely good KDE apps. Their video manager is just the Xandros file manager with some option turned on to ignore non-video files, though. As a video/multimedia player, I think they use Kaffeine.

Out of all the apps they included (which includes OpenOffice, a definite plus), the most surprising were the games Tux Racer and “Frozen Bubble” (not sure the real name of this one). The Eee PC has an intel GMA 900 graphics chip in it, which for an integrated graphic chip isn’t all that bad. Sure, it’s not a dedicated card, but I may be able to get Quake3 running on this. Albeit from a USB stick :^)

Which brings me to storage. Let’s face it, the Eee PC doesn’t have that good of a storage capacity built into it. But, what is it designed for? This is something to keep in mind. It’s obvious that it’s partially geared towards children, but I’m also finding that I could easily write up reports, presentations, notes, and (obviously) blog posts, and so it’s also a decent document machine. On the very first tab is a link to Google Documents. It also has three USB ports and an SD card port. So in my opinion, between the online storage of Google Documents and huge USB sticks (I recently got an 8gb stick for $30, and I know there are huge, cheap SD cards out there), storage simply is not an issue. If I need to write a document, I can do it on my stick, which is probably a good idea anyways in case I need to move it to another machine. Likewise, I can also type up and share things with Google Docs the same way. I think that, in the end, I’ll probably almost never use the solid-state drive built into it except as scratch storage.

Before this weekend is up I hope to try out a few development options for it. I have this idea that, since the important hardware is consistent between the different Eee PC models, I can probably develop Linux programs where they don’t need to be installed into the system the traditional way (.deb/.rpm/.tgz packages or from source), and can instead provide programs specific for the Eee PC that can run from a USB stick. There’s nothing that says Linux can’t run this way…it’s just that it’s not normal.

Eee PC: First Impressions

2008-02-02T00:00:00-07:00

Well, thought I’d give this a try…see how well it works being here…

Anyways, I got my Asus Eee PC today. As soon as I got to the box I knew that I had something special here. So what I figured I’d do is write up a cliche “First Impressions” post. Size: This thing is small. Like, REALLY small. Imagine a DVD box. Now, add about 3/4ths of an inch on each side and you have the Eee PC. Thickness wise, it’s about 1.5 to 2 DVD boxes high. But the weight is what I really like. The thing is light as a feather at just over 2lbs.

Screen: I was originally concerned that the screen would be a highly limiting factor. But to my surprise, the 800x480 resolution is actually not that hindering at all. Asus has obviously taking care to adjust the applications, fonts, and icons to work well on this display. The onle problem I’ve had so far is when I tried setting up an account in KMail, where the settings window was too large for the screen. However, Asus doesn’t seem to really push this as the mail email app.

Mouse Pad: Pretty small, just a little larger than a GBA cartridge. The feel of it is very nice, and it’s cool that it has a slider area on the right to emulate a mouse wheel. However, the physical button right below it takes some getting used to. To click, you have to push almost straight down.

Keyboard: Well, I can safely say that I won’t be programming on this thing any time soon :^) The keys are extremely small, about the size of a dime, but feel good. Typing on it just takes patience while you adjust. In the mean time…I’m typing this on a USB keyboard that’s literally 3x wider than the laptop itself.

Sound: Good, actually. I didn’t know it had built-in stereo speakers to the sides of the screen.

Performance: Even with the processor underclocked to 633MHz (It’s a 900MHz chip) and only 512mb ram (upgradeable, mind you), the thing runs really well. This was my biggest surprise of them all since I was expecting something similar to my iPaq rx1955 PDA. Instead, I get an experience comparable to my usual laptop.

Wireless: No problems here. The configuration programs are easy to use (this is actually my first time setting up WiFi graphically in Linux), and the performance is great. It would be nice to be able to select a “Default” wireless network, though.

OS and Software: This is my first experience with Xandros Linux. So far, it’s pretty cool. Not sure who did this, but it uses IceWM as its default window manager with a custom tabbed interface. It’s actually really easy to use, and I haven’t had any issues.

What’s really cool is that it has support for multi-language input out of the box. Yep, ctrl+space enables the familiar (to me) SCIM input method editor. Chinese is installed by default, so I’ll have to figure out how to get Japanese working.

Asus has hidden a lot of the usual Linux functionality I’m used to from users. However, I found out I can get to a terminal with Ctrl+Alt+T, which gives me access to pretty much the entire system.

The one issue I had was with the Add/Remove Programs application. This is actually the updater app, and it’s a somewhat confusing one. Some items say “Install”, which are apps that you don’t have installed yet. This isn’t the bad part, as these apps just auto-download from Asus’s APT repository. The bad part is updating them, or rather the wording of it. The instructions tell you to click “Update” and then “Remove” once it’s finished updating. Huh? Clicking “Remove” actually uninstalls the app! This is a case of Engrish, I think.

So there you have it. The Asus Eee PC is something that I know I’ll be using very often. It’s easy to use, feature-filled, and actually closer to a laptop than I thought (I was expecting a glorified PDA). Definitely gets my thumbs up.