I’ve recorded ~7 minutes of ambient noise out on Polk Place. Note that the bell tower rings half past at 4:12 in the recording. You can find multiple file formats here. Please note that this audio is free to use under a Creative Commons license:

Polk Place in Summer by Cristóbal Palmer is licensed under a Creative Commons Attribution 3.0 United States License.
Based on a work at garp.metalab.unc.edu.

At the Daily Grind just now the young lady ahead of me had a stack of brightly-colored flash cards. On the flash cards were the amino acids. Seeing that brought back a lot of NCSSM memories again. This past weekend I did a fair bit of cleaning/purging and ended up parting with physics, cell biology, and other class notes from my NCSSM days. I’m a bit embarrassed that I still had all those notes, but “molecell” with Dr. Naiman was SO AWESOME. I owe so much to that school.

The Union has a shiny web-based system for room reservations. Ouch:

I have to read and respond to this article for my 08:00 class:

Comeau and Wilbur (2004), “Non-Word Identification or Spell Checking without a Dictionary.”  JASIST, 55(2), 169-177.

The hilarious part is that “Comeau” is misspelled as “Commeau” on the class website. The article is actually fairly interesting. The authors use a context score, 𝛂, as a baseline of how likely a word is to be spelled correctly and develop a method to predict which low-𝛂 tokens will be misspellings. They say their method has a recall of 79% (finds 79% of the misspellings) and a precision of 86% (86% of the things it calls misspelled actually are) for the MEDLINE© corpus they use. Their use of MEDLINE© is important, since a huge percentage of the correct words in the corpus don’t appear in any dictionary. Oh, and their method can distinguish abbreviations/acronyms from words, too. Pretty hot.

I just finished giving a demo for a database class, and I was very happy to do this as soon as I was done:

root@kant:~# dpkg -P oracle-xe-universal
(Reading database … 121041 files and directories currently installed.)
Removing oracle-xe-universal …
Purging configuration files for oracle-xe-universal …
root@kant:~#

When it works, it seems to work fairly well, but there were FAR too many quirks that I had to spend far too long on with oracle. I never quite appreciated how awesome software like SQLite and MySQL really is until I put oracle on here. FSF- and OSI-approved licenses really do make the world a happier place.

Some quirks I can think of off the top of my head:

• having to run ‘/etc/init.d/oracle-xe configure’ after the package installed to get oracle set up.
• funky $PATH
• having to add myself to the ‘dba’ group
• having to fix a broken symlink left by the installer
• an /etc/init.d script that would be magically unresponsive sometimes (would fail /etc/init.d/oracle-xe stop and disable silently)
• a management webapp that sometimes failed to come up when starting with the /etc/init.d script.

Don’t get me wrong: I fully appreciate that oracle provides some powerful features and is happily and successfully deployed in all sorts high-wattage places, but as far as my personal machine is concerned… good riddance. I love you, dear MySQL.

I may be abusing the term BIRG here, but I just saw two people BIRGing by wearing Harvard shirts while going to get coffee. If I had more time I’d try to articulate why this annoys the crap out of me, but fortunately I’m rather busy getting ready for a graduate level database systems exam. Now that I think about it, maybe I should be thankful that I saw those shirts and got annoyed. Suddenly I’m feeling a lot better about having this exam.

Maybe I only find this funny because it’s late…

Subject: Re: project almost done

On Dec 3, 2007, at 11:56 PM, Maarten wrote:
>Hi!
>
>Cristóbal and Maarten here. I (Cristóbal) have stolen Maarten’s
>keyboard because he’s exhausted. We’re almost done!
>
>We discovered (long story) that we needed mysql to be thoroughly
>utf8-ified for our project to work. This turned into a bit of a
>project-within-a-project, since we had to build mysql-server from
>source to get this done. The fact that the debian source package
>lacks the Doc/ directory (something to do with the DFSG) meant that
>THIS became a project-within-a-project-within-a-project. Ack! Hours
>gone!
>

Ouch. Database character encoding issues are one of my least favorite
problems. You have my deepest sympathy.

>
>
>Finally we have an XML file that was produced using a .php page with
>the help of some pear modules. Next step is to use some XSLT and
>AJAX-y fun to make our search suggestion box.
>
>Can we say Thursday for getting this to you? We think it will be
>relatively quick for us, but we both have classes tomorrow and
>Wednesday, so Thursday would be helpful.
>

Thursday is cool.

I’ve been quoted in the paper.

Sadly, the article focuses heavily on the downsides of using p2p networks to get music from RIAA-signed artists. Worse, it equates “famous” artists with RIAA-signed artists. Perhaps I didn’t communicate clearly. I come out like this:

Palmer said the idea of downloading MP3 files for free and legally from famous artists, whether from Ruckus or any other software, is an illusion.

“It’s a unicorn,” he said. “It doesn’t exist. If you really have to have their music, then bite the bullet and get it through iTunes or Amazon.com or any number of other online retailers.”

I spoke for nearly 45 minutes with the reporter and was a bit disappointed that she didn’t mention Creative Commons and only mentioned Magnatune and emusic.com briefly at the end. I showed her the magnatune integration in amarok. I talked about getting Spoon’s new album from emusic.com after seeing a youtube video. Where’d all that go? The graphic and the headline are painful, too. Looks like COSI and Free Culture have a lot of work to do.

Somebody wandered by the ibiblio office to take our picture for a press release earlier today, and I look ridiculous in it:

Sheesh! I wish she’d taken a few more shots. I definitely wasn’t ready for that one.

Seriously though: it’s an honor to be a part of the team that keeps ibiblio running, and we have far too much fun in this office. I look forward to doing more for ibiblio, UNC, and the wider community through my work here.

Looks like I got clearance. The song is in the public domain and the two musicians are happy to share this version of “Sail Away Ladies” under a Creative Commons License. The two musicians are Jason “Jaso” Phillips (fiddle) and David Cranford (banjo), seen below holding their instruments.

Thanks guys! I look forward to coming out on Wednesdays in the future to catch more jamming.