Luis VillaRamblings on software, law, and the spaces in between.

Random bits from the course of the day:

Lessig explaining CC, giving some history, and asking for donations- they don’t appear to be running out of money, per se, but if some percentage of their money doesn’t come from a distributed base of donors, the IRS gets unhappy.

Just as Linus explained his abhorrence of specs the other day, Mark Pilgrim has a really great rant on overgrown ‘standards’ and how obsession with corner cases can really screw you. Should be a great read for the lesscode crowd.

Mark also detailed his first post-blog project; sort of a mit semantic web piggy-bank lite. That might be an interesting thing for the beagle and epiphany folks to keep an eye on.

Siva Vaidhyanathan has a great read on how the current wave of peer production is not revolutionary but rather restorative. Nothing new there, exactly, but very well written, and a meme that should be spread.

kikidonk made me happy with a keybinding for the magic applet.

Work was work; stared at an sql error for quite some time before realizing I was looking at the wrong error log, which was why I wasn’t seeing an error. Still no idea why http auth isn’t happening correctly for my little calendar installation that should have been off my plate a week ago.

Looking forward to summit, particularly to meet Nathan Yergler.

Put up some more schedule bits in the wiki. General philosophy for the weekend (as I propose it) is (to quote the wiki):

The emphasis this year will be even more focused on BOFs and hacking. Day 1 will be dedicated to speed talks and BOFs that might yield hacking on Day 2 and Day 3. Day 2, for those not hacking, will focus on BOFs for things like marketing, QA, and ISVs, which are generally more policy-oriented than hacking-oriented (not that the QA guys can’t hack :)
On the last day, we’ll spill over from the first couple days, and at the end of the day, review the results of the hacking and BOFing in the closing.

Hopefully that’ll work well; we’ll see.

This job is forcing me to refine and restate some assumptions I’ve built up over the last four years. Hopefully I’ll get those out into print at some point as a guide for anyone else who slips out of our little free software world and back into the real world.

New, interesting-looking toy of the day: synchroedit, a sort of yarr-like collaborative web text-entry thingy, that will eventually be open source and integrated into wikiwyg and kwiki.

Nobody wins any challenge until code is checked in. :)

Another big gnome deployment: Sun Wah Linux is deploying their GNOME-based Rays LX to 140+ thousand machines in China. We continue to win big, substantial deployments across the world, and that is really, really exciting.

Unfortunately, Sun Wah are shipping 2.6 with substantial i18n input fixes (as far as I can see); just as Sun has shipped 2.old with substantial a11y fixes for some time. This highlights the need (IMHO) to have a full-time (or several) GNOME developers whose main task is integrating the work of distros (both code and QA-wise) into HEAD so that instead of all repetitively doing QA and development on old branches, we can collaboratively and less wastefully work on new and better stuff. [To put this another way- think of all the QA and development effort that has gone into GNOME 2.6 since 2.8 was released. Think of all that getting lost, or being done repetitively by Sun, Sun Wah, and others. IMHO, the big distros need to get together and figure out a way to cut down on that repetition and waste if they want to move forward and become more competitive with XP and OS/X. I'm not generally a fan of the centralized Mozilla culture, but it does encourage all the major players in that development arena to stay on the same page and reduces waste substantially.]

Upgraded bugzilla at work yesterday; will start filling it with stuff soon. Emailed back and forth a bit with Olav about his incredibly awesome work to get b.g.o upgraded.

Brilliant idea for next year’s GUADEC, from a planning meeting for wikimania boston I’m sitting in- it would be really cool to have slightly-laggy-but-almost-live translations either streamed or in IRC, done by pre-volunteered translators who were watching/listening to the streams. They don’t have to be in the room- can be anywhere they can get to the

(As an aside, I had no idea Toronto was so multinational- I had my first ‘man that guy is an idiot’ look from folks around here when I claimed canada was bilingual, not really multilingual ;)

Deskbar applet rocks. I’m an obsessive run dialog reader, or was until today. No more. I’ve actually removed my foot menu for the time being, since it was in an upper-left corner and I can jam my mouse there easily (which will have to do until bug 317941 is resolved.) And that is even without the goodness of beagle, which I don’t have working ATM. I expect the combination of the two will make life very good.

Am at berkman lunch with Tom Steinberg, of the very interesting My Society UK. Fascinating guy, and his first job at a thinktank was as an admin, so maybe he’ll be a good role model ;) He mentioned the fairly cool Placeopedia; I added Hahvahd.

If you are in Finland, you are needed in the capital tomorrow to protest against a new copyright law.

I’d kill for a breezy package of the new deskbar applet. Hint. Hint. ;) Am glad to see that it can launch arbitrary apps and autocomplete on .desktop files- conceptually, at least, this would make a very nice replacement for the current run dialog. Still have yet to see how it works in the real world, sadly.

Glynn: You don’t want to be a Luis, trust me. :) I do hope I’ll have time sometime before law school to sit down and finish off my sketched out free software QA whitepaper, which should include some ‘how to be Luis’ tips, I guess. One big key is not to worry too much about mistakes- make up for imperfection with volume and overall impact on the database. If there are 1/3 as many bugs and it is 3-5 times as easy to find things (because there are many ways to improve discoverability besides closing other bugs), they’ll forgive you any small mistakes that were made along the way. Remember that the goal is not to have a perfect listing of all bugs (which helps no one), but to have a listing (however imperfect) that actually helps developers and managers get work done. Compromises in the first taken in the name of the second should always be acceptable.

Am done with standardized tests, hopefully for the rest of my life. I think I did well but won’t really know until three weeks from today. Discovered the night before the test that the paperwork I had to print for the test (1) didn’t display correctly in evince (2) had the same display problem in OS/X Preview and (3) caused old (but not new) versions of windows acrobat to barf on printing (displayed fine). That was a little stressful to find out so late, but I was glad to find out evince was in good company.

After a week of mental detox and summit I’ll start on applications.

Took the test in Miami to have fewer distractions; as a nice plus I got to raid my Dad’s wine rack for dinner on Saturday night. Yum- Penfolds and Caymus. As usual, am depressed about the difference in weather between Miami and Boston. ;) Saw my step-sister Deblois for the first time since her divorce; she looks happy and is about to start recording an album. I wish her luck. Other step-sister (Lucinda, no web page ;) is getting married in November and preparations are in full, insane swing- much discussion of the proper fabric for ribbons, planting of flowers (ceremony will be in my step-dad’s back yard), etc. Glad I’m not in any hurry to do that.