Archive for the ‘musings’ Category

Announcement: New perlwikipedia maintainer

December 8, 2007

Well, I finally bit the bullet today and stepped down as maintainer of Perlwikipedia, my MediaWiki bot framework. My successor is ST47, a fellow admin on enwiki who serves on the Bot Approvals Group and has more bots than I have fingers.

I can’t say that it hasn’t been a long time coming, but I think that ST47 will do a much better job as maintainer than I did. He’s enthusiastic about Wikipedia, is a great Perl hacker, and has written more bolt-on enhancements to Perlwikipedia than there are original lines of code.

In any case, I believe we’ll see a brand-spankin’-new Perlwikipedia release in the near future, one that’s more shiny and can do your dishes.



Distro wars

December 1, 2007

I have a problem. I’m probably the most indecisive person that I have ever known. I have severe difficulties deciding what dressing to put on my salad. Guitar strings? I need an hour of research. So it’s fairly obvious that I’m horrible at choosing a Linux distribution.

The problem is, there’s so many! Since I first saw Linux about 3-4 years ago, I’ve tried Fedora, Gentoo, Debian, Slackware, Sourcemage, SuSE, Lunar Linux, OpenBSD (it’s close enough), and Ubuntu. I’m sure there’s more in that list.

I believe the first distro I ever used was Fedora. That was Fedora Core 4, on a slow-as-dirt i386 machine clocked at around 133 MHz. I then used most of the distributions in the above list on that machine. Guess how well Gentoo performed at package installation on that machine. Needless to say, I soon upgraded my laptop to a blazing fast 1.6 GHz machine. Compared to what I used before, it was raw power.

Of course, having a faster machine also meant that I would need a correspondingly good distribution to handle the awesomeness (Seriously, 1.6 GHz was a huge step up from 133 MHz, it was a power trip). So then I cycled through every single distribution I knew. Finally, I settled on Ubuntu, because it was the only distribution that actually worked with my annoyingly-unsupported i810 on-board graphics chipset. That lasted for a good 6 months. Then, for some odd reason, I decided to switch to Gentoo. I backed up my machine, wiped the drive, and… Gentoo didn’t work with the i810. Great, I thought. I’ll just go back to Ubuntu. And so I did, for another month or so. Then I tried going back to Gentoo. And it failed again. This continued for a good 4 months or so (I didn’t know at the time that there was an i810 driver in the kernel, which I neglected to enable). Finally, Gentoo worked! It was a miracle! And so, I became one of the Linux elite who compiled everything from source and could manipulate the command line blindfolded while drunk. That lasted about another 6 months. Then, I realized that I was spending more time compiling and configuring than I was doing work. So, I switched to Fedora, which had gone through two cores already and landed at Fedora 7. That’s where I sit today.

However, I’m starting to think that Gentoo isn’t so bad after all. I mean, I’m a geek. I like to do geeky things. I like seeing my system exposed, like it was in Gentoo, not like with Fedora, where all I do is type `yum install evolution’ and sit back and watch, blissfullyunaware of what is going on. Fedora’s great, don’t get me wrong. If I had used Gentoo for my MythTV setup, it would have taken at least two weeks. But Gentoo seemed so… fun, I guess (As I write this, I notice that jhbuild is building evolution-data-server, which is the farthest it’s ever gotten. Hooray!).

The real question is, if I switch back to Gentoo, what happens? Will I finally resign from Wikipedia? Become a kernel hacker? Oh, that’s the other thing. On Gentoo, kernel modules were dirt simple to build. Just cd to /usr/src/linux/drivers/misc, write a test driver, insert the corresponding fragment into the Makefile, and type make. Done.

But anyway, I don’t know if I’m even sure that I want to start thinking about switching. I need to do more research, my intuition is notoriously unreliable for this sort of thing.


Jhbuild headaches

November 30, 2007

I’ve spent the last day or so wrestling with Jhbuild, Gnome’s build-from-SVN program. I thought “All I want is to build Evolution, it’s all I ask.” Nah, that would be too easy!

There’s a reason that they say trunk is unstable. I’ve had more build errors trying to get jhbuild to do a clean run on Evolution than I did trying to compile everything from source on Fedora Core 4.

I’ll post problems I encounter and their solutions, assuming I manage to get this thing built finally. It might be easier just to grab Fedora’s srpms. Meh, I’m a developer, let’s take the hard way.


Linux rocks!

November 25, 2007

I own two laptops. One is connected to a widescreen TV via the VGA-out connection on the back, and the other is my personal machine that I do all of my work on. Both have wireless cards. I use wireless for the TV laptop because I’m too lazy to pull cable, and I use wireless for my personal laptop because, well, it’s better than being tethered to an Ethernet cable at my desk.

Both cards, however, have Broadcom chipsets. For those of you cringing at the sound of “broadcom,” I feel your pain. Up until recently, the bcm43xx-series of chipsets were impossible to use under Linux. Then the fabulous bcm43xx drivers were released, and now Fedora incorporates the b43 driver into their standard kernel (All of my machines run Fedora, too).

The end result? Linux just works. I know that many have seen the “Mac vs. PC” adverts on TV, but they just have nothing on Linux. I went out and bought a Linksys WPC54G notebook card today, so I wouldn’t have to pull the card out of my TV laptop whenever I felt like going wireless. I pulled the card out of the box, unwrapped the anti-static cover, and plugged it into my laptop. Ten seconds later, I received GNOME’s wonderful “You are now connected to wireless network..” message. Ah, the sweet sound of something working properly.

On Gentoo, though, I think it was more fun to use Linux. I was like an adventurer going into the jungle, not sure of what obstacles lay ahead. I had to compile my own kernels, build software from source, and (gasp) select my packages’ features. It was sweet control-freak bliss.

On Fedora, I think that my Linux experience is closer to the “Macs just work” theory. I do some simple shell commands to extract the firmware from the Broadcom drivers, and then I can use any Broadcom-based wireless card. Simple.

Linux just works.


I want to be a developer.

November 24, 2007

It’s hard to get into open source. I mean, for me, anyway. I’m not sure what exactly I want to develop, or even how to start, but I know I want to help out the community.

I’ve been looking at kernel module programming as of late, but the trouble is that I don’t know what to write. I don’t have any devices I use that don’t work on Linux, and there’s not really a central place where people say “Hey, it’d be nice if this device worked.” Not to mention, I doubt that I have enough C / kernel experience to write a decent driver anyway.

Wikipedia is great, but it will only get so far. I’ve realized that I’m not on Wikipedia anymore as an editor, which bothers me. AntiSpamBot used to be great, but it just seems to be annoying more and more people lately, and I’m seeing little to no net benefit from it. It seems to have lost the Useful Purpose effect, unlike the Tawkerbots, which people seem to adore more than their own children. Perlwikipedia is good as well, except that I’ve been told on a number of occasions that it’s written poorly. I realize that. You can’t expect me to write a flawless work of art that rivals OpenBSD the first time.

I think I’m eventually going to burn myself out of Wikipedia. I’ve been coding in C# lately to write DiffShovel, but I don’t think that DiffShovel is going to be of much benefit to the community, in a manner similar to most of the code I write. And it, too, is written poorly. Even I can tell that, although I don’t think I should be writing good code when I’ve only known C# for a month. At some point, I’m going to realize that I don’t add anything to the project any more, I’ll find others to maintain the bots and Perlwikipedia, and then I’ll vanish.

I’ve looked at contributing patches to apps that I use, like Banshee (my music organizer/player) and the Linux kernel, but I think they’re too big for me to wrap my new-developer-mind around. I need to start small and work my way up, but I can’t find something to contribute to!


(In other news, I’m at 38,841 words, and this post was written solely to stall for time. In about 30 seconds I’ll be forced to open vi and get to 40,000.)

(Yep, right about now.)

(In a minute.)