So as I mentioned briefly in my last post, I started work at the Participatory Culture Foundation (PCF) this month, mostly to work on Miro. Anyone who knows me probably can guess that a job working on free and open source software, especially related to media, and in Python, is a huge dream come true.

I had mentioned that I was to give a talk at ChiPy about Miro. And talk I did... there's even a recording of my talk available to watch. (The angle's a bit weird to look at, and you're mostly looking at my emacs buffer, but the talk itself is interesting, I think.)

However, that talk is mostly directed at a programming audience, and since this blog is read by some non-programming friends and family, I figured I should write up some explanation of why I'm so hyped about working here.

So first of all, Miro itself is awesome. It's a free and open source internet television player. There's tons of content for it... tons of content... all available on the Miro Guide. (The Miro Guide is itself a really cool project. And yes, it's programmed in Django.)

Part of why Miro matters so much is that it's built on open standards. There are some other internet video players out there, but they often rely on proprietary schemas. I like to think that Miro is kind of like the Firefox of internet TV.

It's also really enjoyable to use. You know, there's that thing.

So for about three months primary to joining the PCF fulltime, I was a volunteer to Miro's codebase. It's been great, partly because I've been able to hit the ground running, but also because during that time I came to really enjoy working on Miro's codebase. Which is part of what makes being hired on to work at the Participatory Culture Foundation so cool... I already knew I enjoyed working on Miro. And now I get to work on it fulltime. Not to mention that all of the people at the PCF are super nice, super fun to work with, super smart, and super productive (giving me a good challenge to try and keep up...).

There's also the fact that the Participatory Culture Foundation has a very clear and noble mission. Aside from just working on technology to consume media, the PCF is interested in helping to inform people on how to make internet television, as well as educating people about issues related to Open Video (something the PCF takes seriously). So overall, this is a very morally fulfilling organization to work for, and they've got other cool things in the works. So, what can I say? I'm super happy to be where I am now.

By the way, I'm now syndicated on Planet Miro. Hello, Planet Miro! I guess Will already beat me to introducing me to the Planet Miro scene, or whatever :).

I'm giving a talk tonight on Miro, which I've mentioned before, at ChiPy. Details are on teh wiki, but might as well duplicate them here:

Chipy's October Meeting will be our best ever.

Location: Skinny Corp, 4043 N. Ravenswood Ave. Suite 106

Date: Thursday, Oct 9th, ~7pm

Topics:

: - Chris Webber: Miro, a free, open source internet tv & video player - KumarMcMillan - freebase, a free collaborative database with a rich API

I know of a few other people who will probably be giving talks but didn't update the wiki or announce on list. Also, you should come since I've been hired by the Participatory Culture Foundation to work on Miro full-time, so you can find out all about it!

What's that? I forgot to mention that I've changed jobs? Well, if you come to the talk you'll find out more! I'll update my blog to talk about it this weekend, but if you come tonight you won't have to wait as long! :)

Edit: Also, SkinnyCorp people are the cool people who run Threadless, which I know a couple people who read this blog are fans of. So visiting their office is one more reason to come!

I gave a talk at ChiPy, the Chicago Python usergroup, tonight; my work, Imaginary Landscape, hosted. It was a really good meeting. Ian Bicking gave a talk on how to write a web app without a framework (which was also kind of a peek into how to build your own python web framework), Peter Fein gave a talk on his Factory module (a really neat approach to simplifying redundant code), and I gave a talk on the OpenMoko phone. We always advertise the current ChiPy meeting as going to be "our best meeting ever", but I really do think this was one of the best ones. That's not just me being conceited (although I might be a little bit, it's hard for me to tell), I thought all the talks went well.

I did think this was the best talk I've given though. Honestly, I wasn't going to do it this month because I didn't think I was prepared enough... I was going to wait until I had thrown together some example applications. But instead it was a bit more ad-hoc... I just described the phone and the history of it, and then gave a brief tutorial of how to make a phone call by entering lines into the interpreter one by one. I promised that it would be less than ten lines of code, and it was (it was seven), but I noted that if we hadn't already turned on the antenna (and it was turned on before I opened the interpreter), I would have needed an additional four, bringing the number to eleven. Well anyway, when I finally entered the last line (with the phone number of an audience member as an argument) there was silence for a moment... and then when his phone rang, well that's when the room started buzzing with excitement.

I've been meaning to write about the OpenMoko phone on here for a while, and why I think it matters, but maybe I can just post the talk... Carl Karsten was kind enough to record it, and we might try to upload it to blip.tv next week. I've been talking about how it would be kind of nice to start a ChiPy videocast, a channel one could watch with Miro and stuff. Sounds like that might happen.

I was just talking about the work I've been doing on Miro, and now there's an article featuring Miro on LifeHacker. Nice! It's a pretty good overview as well.

You might consider digging the article, if you're into that kind of stuff.

If you read Planet Miro, you may have noticed that I was recognized in Will's blog for the volunteering I've been doing on Miro. I've been planning on mentioning my participation for some time on here, so this seems like a good time for me to do so.

It's been fun. But most of all, it's something I feel is really important. Television is the most consumed medium in modern western culture, and with the internet, there's a chance to shift it away from its original place of control by just a small number of megacorporations toward something that's as democratic as the web. There are a lot of IP-TV systems emerging, but Miro's the only one that really takes open and decentralized video playing seriously.

That's still a pretty vague explanation for why I think this is so important. Hopefully I'll find some time to really flesh out this reasoning soon, because it really does matter to me. In the meanwhile I can tell you that there's a lot of really exciting development happening in SVN trunk, including an entire user interface overhaul. We're switching the codebase from a lot of embedded HTML to actual widgetry. I guarantee that Linux/GTK has never looked better, and the code is getting a lot cleaner too with significantly less fragmentation across platforms. It's also faster and more featureful. Right now though, we're still in the process of reimplementing a lot of the old code. I wouldn't run from trunk right now, especially because it changes the database in a way that makes it incompatible with the last stable release. So if you upgrade to the development code, you're stuck with it. Don't worry, SVN is evolving at an astonishingly fast rate. It's already more enjoyable for me to run SVN than the last stable release, but I really should re-emphasize the fact that there are still a lot of important features missing.

The Miro people also sent me a t-shirt for the volunteering I've done. I picked the pretty one with the bird, and it's already one of my favorites in my collection. I might buy the exploding TV one soon too.

This has been really fun. It's nice to work on an application that's not web related and which is used by a lot of people. And everyone from the core Miro team has been really great to work with.

2.0's gonna be awesome. You'll see.

Whee! Siggraph, the world's leading graphics conference, is rolling around the corner, and this year is looking really good. Probably won't surprise anyone that my interests are largely Blender oriented, but I'm really enthusiastic about the festival in general. It looks like it's running from the tenth to the fifteenth of this month, but I really want to go on the thirteenth, because that's when The Making of "Big Buck Bunny": An Open-Source Evolution is going to present.

It's here in Chicago, and it looks like a one day pass is a mere $45. That doesn't sound so bad.

Hm. The thirteenth is just next Wednesday? Maybe I can't attend that. If not, at least it's good to see Blender and Big Buck Bunny gettin' represented.

From eighth grade all the way through high school, there was a single, late-dos era two-player game that all my friends and I knew and loved. Upon many occasions that we got together, we would play it. I'm talking about Liero, that fantastic game that's somewhere between a two dimensional game of Quake or a real-time Worms. It was great, though a little buggy, but damn did the physics felt nice (in a wonderfully unrealistic way), and the ninja rope was a freaking blast to use.

But it was closed source, and apparently the author lost the source code due a disk crash, and for years we were stuck with playing the same tired old binary. Then Windows XP came out, and sound stopped working entirely. And eventually it moved from a game we played to a game we reminisced.

Over the years I've checked on the number of clones which have come about, but none of them really had the same feel as the original Liero. But now that's changed, for a new free and open source Liero clone is out, called OpenLieroX, based off the Liero Extreme codebase. And it's awesome. It supports the original Liero weaponpack, as far as I can tell, flawleslly, and has a vast wealth of mods and levelpacks available for download. On top of that, it has network play... which isn't perfect, and I often find myself running into my own bullets in ways that would never happen offline, but it's still pretty good.

Installing it on Linux actually isn't that hard. There's a .deb file available on the sourceforge download page, if you happen to be running Debian or Ubuntu, as well as windows builds. But I'd recommend running from subversion, which currently has auto level and mod download support (the latest stable releases don't). To compile, run these commands:

svn co https://openlierox.svn.sourceforge.net/svnroot/openlierox openlierox
sudo apt-get install libsdl1.2-dev libsdl-mixer1.2-dev libsdl-image1.2-dev libgd2-noxpm-dev zlib1g-dev libxml2-dev
./compile.sh

...then you can run either 'sudo ./install.sh' './start.sh', depending on whether you want to install it or just run it locally without installing.

I'm actually interested in playing a few games online with people I know... maybe even having a small local openlierox lan party. Anyone interested?

The title of this post is actually less me trying to make a bad SQL joke as it is me ripping off my favorite Dinosaur Comics comic. But yeah. Lots of deadness of non-funness. Anyway, this is one of those boring personal life posts that people probably don't care about, but it at least has a fairly simple moral to the story.

So I noticed from work on Friday that my ssh connection kept resetting to my home machine. Every once in a while, RCN is flaky, so I figured that's what it was until I realized that emacsclient wasn't connecting to my existing session and that my uptime was only five minutes. It was raining, so I figured that maybe the power went out several times.

Well, I got home and quickly realized that my computer wasn't turning on. At all. I took off the case and noticed that, while the power supply fan was running, the processor fan wasn't. Cycling the power supply and then hitting the power button would lead to the processor fan spinning for a couple of seconds and then stopping. So I tried reseating the ram. No luck. I tried with individual sticks of ram. Again, no luck. So it didn't seem to be the ram. My couple of years of diagnosing and repairing machines at my previous job have made me pretty decent at fixing machines (though I mostly got by at that job with using my software skills to overcome my lack of hardware skills), but I'm pretty rusty. But anyway, at this point I figured there was a good chance that it was the powersupply, despite the fact that the fan was running, as it might not be supplying enough power to power up the machine entirely (which I've seen happen before). So I tried replacing the powersupply. Nope, no luck. Luckily I had made a backup of my data a couple of weeks ago, so I wasn't really worried there. Even so, most likely I figured the drive wasn't damaged, so I probably hadn't even lost anything within that group of time.

At this point, I figure it's probably the motherboard, but it's not within my budget to replace that right now. So I asked Morgan, my wonderful fiance, if she'd mind if I "moved in" to her computer until I was able to fix mine. She was fine with it. I figured this would also be good motivation for me to actually start regular backups on her computer, which I've been promising to do for a while. About six months ago, I installed a second drive on Morgan's computer so she could easily run Ubuntu. She hasn't touched the windows drive much since then, except to pull data over. So it shouldn't be too hard for me to move over.

So I pull out my drive, and with one of those nice little PATA-to-USB adapters I started rsync'ing my data over to my home directory.

Except, bad news. That drive I installed Ubuntu on for Morgan was a rather old one, and apparently moving several gigs of data over was just too much for it to handle. Halfway through the transfer, it died. Completely died. Like, I couldn't even 'dd' the drive to an image so I could try to inspect and recover its contents. I deliver the bad news to Morgan. She takes it pretty well, since most of her data was still mirrored on the windows machine, but she probably did lose a paper or two. At any rate, I was pretty clearly an ass for not backing up her drives sooner. So I promised to back up her Windows drive right then, to be sure that no more freak accidents could happen where she would lose all of her data.

So I took the Western Digital Passport I had bought expressly for this purpose several months ago. Fired up rsync, and... well, it didn't die, but it kept clicking and then disconnecting during the middle of transfer. Thankfully, rsync is inherently well designed for this scenario, but... dammit.

So I went out and bought two new drives, one to replace the dead Linux drive, and one to put in an enclosure for regular backups. Those drives are fine, but I'm pretty disturbed by the multiple system failures I experienced. The moral of the story is pretty obvious. Backups are critical. Losing your computer is a frustrating experience. Losing your data can be a traumatic experience. Thankfully, not much data was lost, though that small amount of data that Morgan did lose was totally preventable.

I'm also planning on writing a tutorial on how to do easy backups with rsync and either an external drive or another remote computer. I'm also thinking about starting to back up to an entirely remote location, like either a family member's house or maybe even something like Amazon S5. But if I did that I'd want to encrypt all my files with something like GPG. I'm not sure about that yet, but I think it's generally a good idea in case some sort of disaster were ever to strike us. Hopefully such a thing won't happen, but with the kind of luck I've had this weekend, I'm all about preparing for the worst.

A lot of people have been complaining about not being able to install, or know where to install, Firebug for firefox

1. Thought I'd spread the news that, yes, there is a release available. It is a beta version, but I haven't seen any problems so far, and it's a bit nicer than the previous version too.

Ok, I'll stop slacking and get back to work now. ;p

To celebrate the release of Firefox 3, the folks at Mozilla are trying to set a world record for most downloads ever in a 24 hour time period... aka, download day!

Well, download day just started (though their main page, at the time of writing, still says firefox 2). But you can get it now, while its hot, for Linux! OSX! Windows!

You want Firefox 3. All those memory leaks? Haven't noticed a one since I've switched. Plus better native deaktop integration, and.. have I mentioned the awesome bar?

Yes. You definitely want this.

Update: Firefox 3 is now properly appearing on the Mozilla homepage... when it actually loads, that is. At the time of writing, I'm hearing on IRC that downloads are between 3000-5000 downloads/minute.

Update 2: Oh hai, download stats counter! (Currently > 12000 downloads a minute... holy crap!)

Update 3: The world record page page is showing a lot more downloads than that other page is (by like, double). I wonder why?