I actually turned 30 a few weeks ago, but I've been super busy and haven't had time to get around to posting about it. Now I have.

Turning 30 doesn't seem really different or exciting or ground-breaking. That pretty much sums it up.

I used to maintain a links section of my blog which I called my "link-blog" (though I may not have called it that out loud). Here is the :doc`blog entry launching my link blog <links>`.

Anyhow, I went through all my link blog items and moved them to delicious. I'll toss a link to my menu and in the future maintain links there.

Bunch of bug-fixes for things I didn't catch the first time around.

Changes:

• fixed num_entries behavior

• fixed RSS 0.9.1, RSS 2.0, and Atom 1.0 feed templates

• fixed $body_escaped variable

• fixed problem with static rendering where it'd render /index.html and //index.html if you had entries in your root category

• merged ReadMeForPlugins documentation into the manual and added/fixed some more content

It's a good release. Many thanks to Norbert (current Debian packager), Joey, Joerg, and the many people who upgraded to PyBlosxom 1.3 despite the fact we have no regression or unit testing system [1] --they are brave people.

I admit that "solid bug report" is vague and probably different depending on the nature of the bug being reported. However, the amount of time it takes to research and fix a bug that has a well-written bug report vs. the time it takes to research and fix a bug that has a craptastic bug report is orders of magnitude in difference. In some cases, the latter bug report doesn't shed any light on the issue at all and thus won't result in a fix in any amount of time.

Anyhow, many many many thanks to Joey Hess to submitted a bug report in the Debian bug system (passed on to me by Norbert) regarding a problem with static rendering. He provided a wealth of information including his whole setup via subversion. He gets ten gold stars.

On a side note while I'm kvetching, I dislike bug reports that include a patch but don't specify what the problem is and why the patch is a good fix for the problem. It seems silly... you'd think I could look at the patch and work backwards to figure out what the issue is, but that is rarely the case. Often people who are providing the patches don't really understand what's going on at that point in the code and they've created a patch that "makes the problem go away for them" which is really different than "fixes the problem".

I finally beat it. I've sort of been working on it for a few years now and I'd get stuck at various points. I never got stuck puzzle-wise. I always got stuck trying to kill one guy without having my party eviscerated.

Anyhow, I'm glad it's done and off my hard drive. Beyond that the story was interesting, but the UI was kind of irritating (especially on my laptop). I have Baldur's Gate II, but I think I'm going to wait to play that for a month or so.

Norbert took over Debian packaging and has sent me a plethora of fixes, questions, and errata most of which I've applied, replied to, or at least thought about. Martin has also sent in several issues that will be fixed for PyBlosxom 1.3.1. Additionally, I went through all the bug reports on the SourceForge bugtracker and dealt with them.

In the process of doing all that, there were a bunch of bugs with 1.3 that got fixed. I also fixed Joey's problem regarding num_entries behavior. I'm re-thinking how some of PyBlosxom works (or doesn't work) and I've gotten all excited about making some changes.

I've done another huge push on the manual fixing a bunch of minor issues and folding documentation that existed in other places into the manual in the proper places. The manual is in docbook format and when I compile it (or whatever the verb should be) into a PDF, it clocks in at 76 pages. It's definitely the longest manual I've ever written. As time goes on, it gets more and more comprehensive. I hope people use it if only because I've spent days writing, honing, editing, and cursing it.

Next week some time (time-willing), I'll do another pass at overhauling the web-site and fixing the massive problem with plugins that we have. The issue is that I don't want to be doing contributed plugin packs--it's a pain in the ass and none of them are tested except the ones I use. On top of that, over the years people stop hosting their plugins which has caused some of the entries in the registry to become obsolete. I'm going to switch how I do things, ditch the registry plugin, and change the system so that the site hosts all the plugins that are offered. I'm also going to enforce some rules regarding things plugins need to have in order to be added to the registry.

S and her parents work at the Me and Thee Coffeehouse in Marblehead, MA (it's 30-45 minutes from Boston depending on how you go). I go every now and then and catch the shows I find interesting. There are a few venues in the Boston area that have a solid concert list and Me and Thee is one of them. For example, in the next few months they have Ellis Paul, Antje Duvekot, and Lucy Kaplansky.

Definitely worth checking the concert schedule to see if there are upcoming shows you might be interested in. It's a great venue and when J is there, they have amazingly decadent cookies.

Me and Thee coffeehouse

2005-2006 concert schedule

I went to the Sipser colloqium today and he discussed P versus NP. I've taken two Algorithms classes and a Theory of Computation class, too, but I still found it really interesting especially given that Sipser wrote the Theory of Computation book we used.

So he does this whole presentation thing and it's really interesting and such and then he takes some questions. The last question was from a woman who had just taken a Theory of Computation course (using his book) and wanted to know why it should be required. I always find questions like that curious. In this case, I'm glad I know about the things I learned in that course. While they're more mathematical in nature than what I'm typically used to, I think they provide an awesome grounding in what kinds of problems are doable and what kinds of problems are computationally difficult. While I'm not planning to continue investigating whether P = NP or P != NP, it's really interesting to know that they exist, what they are, and why they're really interesting.

Anyhow, long story short, it was a really interesting colloqium, I'm glad I'm in grad school, and people who need validation for the things they're learning beyond the knowledge itself puzzle me.

Many apologies for the length of time that this has taken me to complete.

Changes:

• Overhauled the flavour template finding system allowing people to more easily build, trade, and install "flavour packs"

• Added a built-in Atom 1.0 flavour and an RSS 2.0 flavour

• New logging system making it easier to track down PyBlosxom issues and plugin issues

• Documentation comes in the .tar.gz file, though it's in docbook format

• Some other bug fixes here and there

Plugins that work for PyBlosxom 1.2 should continue to work with 1.3.

Overall, I'm pretty happy with it. It's not everything I originally set out to do, but it's a good amount of work for a release. Many many thanks to everyone who helped out.

You can download it from the PyBlosxom website.

It turns out my internet connection was sucking for a week because evil ninja squirrels ate the cable that goes from the telephone pole to the building my apartment is in and then the cable filled with water. Who knew evil ninja squirrels were such a threat to my bottom line?!

The technician replaced the line using an 18' ladder precariously parked in the middle of the semi-busy street in 20 minutes. The frosting on the cake was when he came inside and said, "Yeah, so squirrels chewed on the line and then it filled with water. There's nothing we can do about that because we can't kill all the squirrels."