god owes us one

avoidably detained >>> Sorry about the lack of new stuff this week; I've been in a bit of a work funk since Thanksgiving, and I finally started to come out of it, resulting in a lack of blog time. On the work tip, however, things look like they're starting to click again: I got about seven solid hours of hacking in today, finishing up the refactoring of the Perl libraries for one of my two projects. Reduced the code size by about 25%, but totally FUBARed the interfaces, so the next task is to rewrite all the CGIs that use those libraries.

I also shot my mouth off one too many times about how good a language Perl is, with the end result being that I'm leading a Perl study group. I'm hoping it won't be too much extra work, and I do have an ulterior motive: after I've build up a group that has some good basic skills, I hope to start an advanced Perl study group to read and talk about things I'm interested in. In the meantime, if anybody out there has advice about leading a medium size group (~70 people in ~25 person chunks) through the Llama, I'm all ears.

bigger and badder >>> There's been lots of news recently about Emacs21, which should be released sometime in the new year. It's going to feature a whole slew of new features, like a GTK GUI. You can see some screenshots with accompanying commentary over at userlocal.com, or others over at alpha.gnu.org. I'm quite impressed (in a 'boy that looks pretty but I could never work that way' way) with this one.

If you run Debian, you can pick up some sooper-sekrit Emacs21 .debs, if you know where to look. Be warned, however, that they just core-dump for me, so it might not be worth the bandwidth.

speaking of emacsen >>> Recent nice article about reading mail in Emacs, which also pretty much applies to my editor of choice, XEmacs. It might seem silly at first to read your email in your text editor, but writing email is basically all about manipulating chunks of text, so why bother learning a second app just for mail reading? Odds are, you've already got an editor that you know how to use, and prefer.

what's this white stuff? >>> Since it's supposed to snow any day here, and since I gather that DC is pretty wimpy about shutting things down when that happens, I'll probably need to remember that government closures are announced at the Office of Personnel Management, specifically on this page.

in the unlikely event... >>> ...that you need to move SQL-ish things to and fro on your PalmOS device, SQLpilot might come in useful.

not that i'm mentioning this for selfish reasons >>> If you've got a geek on your shopping list this holiday season, you could do worse than something from copyleft.net. This 'got DeCSS?' shirt, for example, is quite fetching.

a+ idea; b- execution >>> Maynerd Olsen's latest idea is trying to get the biomedical research complex (specifically the drug development part) to stop focusing on finding genes that make people sick, and instead to look for genes that keep people from getting sick. Excellent idea; the issue is how to find them. Basically, you'd need a situation where somebody should have gotten sick, but didn't, and you'd need a way to be reasonably sure that it wasn't just chance. Ideally (speaking utterly without regard to morality or ethics), you'd like to see a region with two mostly (but not completely) distinct, largely inbred populations -- Icelanders and Pacific Islanders, for example. Then you'd like to see the introduction of some sort of nastiness that killed off or sickened people of one background, but not the other. Then you'd take the people with ancestry from both populations and try to correlate some gene(s) with the susceptibility to $BAD_THING.

Of course, if these types of test conditions were plentiful, somebody else would probably have come up with this idea by now...

a bargain at twice the price >>> If you've ever wanted your own personal MCSE (Microsoft Certified Service Engineer) around, to answer your Windows tech support questions and whatnot, wait no more -- your prayers have been answered by Virtual MCSE

odd brit stuff number one >>> Wednesday, a woman in Newcastle was convicted of biting off her friend's husband's testicle. (Got that? Not her friend's testicle, not her husband's testicle, but her friend's husband's testicle.)

I was boggling at the headline, but then I read this:

"The defendant was not aware that she had bitten his testicle until after the incident," prosecutor Stephen Duffield said.

As if that makes it better? And then this:

It was only after police arrived at the house in Gateshead, near Newcastle, that the missing testicle was discovered under a picture frame in the sitting room.

My mind exploded at this point, and I was unable to carry on.

odd brit stuff number two >>> I recently learned, via the miracle of foreign cow-orkers, that the word 'fanny' in British (as opposed to Merkin) is not a mildly archaic word for 'butt', but rather a current and fairly crude word for 'cunt'.

Use of this knowledge to shock or startle the Brits in your life is not only allowed, but encouraged.

we're a corporation; those rules don't apply >>> As they crow about in this press release, Celera has submitted a draft paper describing the sequence of the human genome to Science, one of the top journals in the bioscience field. Normally, this would mean that they had deposited the sequence(s) being discussed in the paper (i.e., the human genome) to GenBank, which is the public repository for all sequence data, and the standard place to send your sequences prior to publication. In fact, that's what Science's Information for Contributors page instructs, under 'Conditions of Acceptance':

Archival data sets (such as sequence and structural data) must be deposited with the appropriate data bank and the identifier code should be sent to Science for inclusion in the published manuscript (coordinates must be released at the time of publication).

(Bolding mine.)

However, as this LA Times story reports, that's not what's happening. Instead, Celera is going to hold on to the sequence themselves, and is apparently planning on requiring people to agree to not use it for commercial purposes before allowing them to view it. Read the LA Times story; there's lots of good reaction from leading people in the field. I'll just include the quote from David Lipman, head of NCBI, and my boss's boss:

"For two decades, the policy of requiring submission of sequence data to Genbank has been a tremendous success and has been critical for much of the progress we've seen in genomics," said David Lipman, director of the National Center for Biotechnology Information, which manages Genbank.

I had a spirited discussion about this issue with a cow-orker earlier tonight. He was arguing that this was an acceptable compromise, which allowed Celera to get the data out there without having to take the major step of attempting to patent it all. I (unsurprisingly) disagree. I'm not against Celera making some money back on their large investment in obtaining and annotating the sequence. However, if they want to do that, they shouldn't be attempting to publish the data, in any form. If they are going to publish, they need to follow the rules that have been established by this section of the scientific community, just like everybody else.

My colleague argued that if they didn't publish, they would run the risk of other firms beating them to the punch on desirable products derived from the sequence. My response to that? "Life is hard." Business is risk, and by attempting to beat other firms in the product development race, Celera is taking a risk that they might get beat.

Ultimately, I'm not surprised that Celera hatched this deal. From their point of view, it's really the best thing to do -- they get the best of all worlds: protection from competition by other companies, the chance of some measure of good will in the academic and public communities, and the ability to argue prior art on anything other companies try to produce.

I am, however, completely disgusted with the editorial board of Science. By rolling over for Celera, they've lowered the bar for future publications from drug companies, and they've made life more complicated for thousands of academic researchers. Science is published by the AAAS, the American Association for the Advancement of Science. This decision doesn't advance Science, and it doesn't do a whole lot for the American public either. About the only thing being advanced here is Celera's opportunities for profit. I'm not, nor have I ever been a member of the AAAS. After this mess, I'm not sure I ever will be.

more beer >>> Thanks for the kind word, Steve (and thanks for the continuing top-notch reporting -- you're really going above and beyond "weblogging" with what you do).

However, I do have full Anchor Xmas's from '98 and '99 (1 bottle and 5 bottles, respectively). Some day a bit closer to the 25th, I'm planning on starting out with the '98, moving into the '99 for a bottle or two, and then finishing off with the 2000.

Oh, and the beard is fully back (why, yes, I am a hairy lad!). It seems to have stopped at its usual one to two inch length. I'm thinking about giving it a mild trim to see if I can encourage further growth...