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

IE and Firefox appear to have agreed on something, but Tim Bray is (as usual) totally correct: a little orange button somewhere in the page, even if everyone agrees on what the button is, is still a suboptimal user experience. (As usual) Tim also has a good suggestion for what the experience should be- something that it shouldn’t be too hard for anyone to implement. If you’re, you know, writing a browser, you should probably click and read the (very short) post.

link

I have been saying that wiki and open source succeed because of statistics and probability for a long time, but I’ve never sat down to explain it really well. This blog post on the long tail blog does. Really good job; I recommend people read it and think about the implications.

gnome

Bugzilla is dead; long live bugzilla. More here soon, hopefully, but… wow. Congrats to Olav for really making this thing rock.

Ray Strode’s post to foundation-list about logos has been sitting in my inbox for a few days now, waiting to be loved. I saw jrb briefly early this evening; he said ‘oh, I nominate you to respond to Ray. (pause) (pause) Wait, wait, no I don’t.’ jrb is wise. He of course says that because, well, my response will be something along the lines of ‘The foot should effectively be like the elephant, or the donkey. There is a platform and a policy, but we don’t use trademark to enforce it. So have at it; if you’re an ass, we’ll lart you publicly.’ This is, I should note, completely not the official policy of the foundation.

I once upon a time wrote a much longer rant about this, explaining how our trademark license suffers from an extreme mismatch with our copyright license, which in turns means that our trademark license suffers from an extreme mismatch with our community and our means of collaborative production. That was before I stumbled upon the elephant/donkey analogy; saying ‘trademark is bad for us, for all these reasons I have a very hard time making succint’ is generally much less effective than ‘here is an example where very high profile movements have discarded trademark and have survived without it.’ So, by all means, read the (never quite finished) paper if you’re a license/social structure geek, but otherwise- just go meditate on how and why the non-trademark approach works for national political parties, and whether or not you think it makes sense for us to follow in their stead.

I keep seeing all this blather about what desktop Linus uses. Now, I of course disagree with Linus; I won’t go into the details why because they have been rehashed 10,000 times already, though if you must know, the two images in Jimmac’s blog capture it better than anything I can say.

That said, it just doesn’t matter. GNOME, KDE, Windowmaker, whatever… none of them are actually innovating in a way that will help the Free desktop win. Hands down, everyone seems to agree right now that OS/X is the most powerful, most stable, and easiest to use desktop on earth. And you know what? No one except a couple million digerati cares. MS sells more copies of Windows in a good month than OS/X does all year, and I’m sure more copies of Windows are stolen in a good week worldwide than OS/X sells all year. The only reason OS/X is at all relevant right now has nothing to do with software- it has to do with the little white things in millions of people’s ears. So… so what if GNOME is easier to use than KDE, or if KDE is more powerful than GNOME? If the magic code fairy created KNOME tomorrow and gave KNOME 10% more power and 10% more ease of use and 100% more Freedom than OS/X, we’d be the best end-user operating system in the world, and we’d still be losing. And we’d still continue to lose, because OS/X and Windows are good enough, and we still don’t have a killer app- a killer innovation- that will make people switch, much less that will make people overlook our problems with wireless drivers, legacy software, .doc/.xls/.ppt import/export, DVDs and mp3s, etc. In the end, yeah, maybe GNOME is better than KDE, or maybe KDE is better than GNOME, but neither is very persuasive- best case scenario for both for the vast majority of users is ‘well, that is good enough so that I can sacrifice and get away from the hell that is Windows.’ That is pretty sad. Until one of the desktops crosses that magical line, and makes large numbers of people really excited- really passionate- about the Linux user experience, the difference between the two doesn’t really matter all that much.

So if you’re tempted to pound out another line of argumentation about whether GNOME is better, or KDE is better, or what have you, go write code for something that actually makes Linux a clear, innovative leader in something, even if there is a risk that innovation will fall flat on its face. Think of an idea that will make your mother or your father or your non-technical siblings say ‘wow, that rules- I must switch to Linux so I can have that!’ And then go implement it. Hell, do a half-assed job. Just do something new and different, and share it with the world so others can run with it, adjust it, or maybe just worship it because you got it right the first time. Don’t go tweak KDE a little more to beat GNOME, or tweak GNOME a little more to beat KDE. Do something new. Do something exciting. Get us out of our rut.

[As an aside, because I didn't blog about it when I saw it- this is exactly the kind of thing I'm talking about. This was new, this was different, it was doing things no other OS is doing right now. Was it perfect? I'm sure not. Can it be improved, possibly by merging it with other features, adding here, trimming there? I'm sure. Is it something lots of people should be experimenting with whenever the code is released? Absolutely. And I listed several other ideas like it way back in April- things that could be explored, implemented, played with- things that would make us different and potentially better than everyone else. What are you doing to make that a reality?]

Research-y Bits

Wikipedia is basically as accurate as Britannica on a selection of science articles, according to Nature. Wikipedia has slightly more errors, but only slightly. And of course they are more up-to-date, though Nature does not cover that.

Interesting-looking research on different types of wikipedia contributors. [Via sj.]

Work

Spent four hours in meetings today; rest of the day I read rails docs, pretty much. Interesting stuff- should be able to build the (very simple) site I need pretty quickly, if rails is how I choose to do it. Should know by the end of the day tomorrow.

Life

Got several applications actually done last night. More tonight, hopefully, and then a nice stiff drink…

Woot! We have gotten a verbal commitment from my favorite candidate for one of our two postings, and after some internal discussion, have re-written the second posting to be more client-centric. Sadly, that means ‘more windows centric’, but what can you do- if you’re tackling spyware, you have to go where the spyware is :) At any rate, if you or a friend have a solid windows client background and are interested in working on fighting the bad guys on a number of interesting fronts, stop by our job posting and take a look. Berkman is a pretty awesome place to work, and we hope that in the next year we’ll be building a pretty dynamic tech organization here that someone (maybe a reader of this very blog ;) can be part of from day one.

Great weekend in New York City, as usual, with the excellent Eily Hayes and Matt Beckett, watching Duke beat the tar out of Texas and generally coasting around the city. Bumped into Joe in Penn Station, so the ride home was nice too- good to catch up with him. Only negative part of the weekend was my knee acting up, unfortunately.

Was pleased to see on my return that I’ve been re-elected to the board; was flattered to see that I was the top vote-getter. Am optimistic that (forced by our recent downsizing) this will be the year the board learns to delegate and advice instead of doing so much in-house. Hope that those who did not get elected (particularly dynamic newcomers like Quim) will not be discouraged by this and instead will become even more active in the work of the community and Foundation.

Read two papers on the way to and from NYC. The first was ‘The Generative Internet’, by one of my bosses. Really thought provoking paper that should be read by anyone interested in the future of technology policy (not just the internet; he makes the point that from a policy perspective separating the PC and the internet no longer makes sense). I’m not sure I can yet agree with all of the conclusions, but it frames central questions very differently from how they have been approached before, and I believe very fruitfully. I’ll probably write more on it later, after I’ve had a chance to re-read it and digest the implications more thoroughly. The second was on ‘The Regulatory Challenges of Virtual Worlds’. A really interesting thought experiment on the implications of competition between virtual worlds- i.e., what would happen if you could take your virtual goods with you when you moved from, say, WOW, and ultima online. The authors hypothesize that the games would start competing on the structure of their governments and property regimes, and they do some interesting analysis based on that assumption. However, it seems like in the end the article is just a thought experiment- the basic assumption (that it will be possible to move data objects between virtual worlds) seems unlikely to occur to me. We can’t even get data to move between word processors, and the market by itself is clearly not interested in fixing that problem.

OOo 2.0′s dictionary does not include ‘blog’ or ‘wiki.’ That amuses me for some reason. And yes, I’m writing a design spec, in OOo 2, that discusses blogs and wikis. That does not amuse me, unfortunately.

I thought I was healthy. I woke up this morning kicking ass; by the end of the day my brain was toast and I’ve now gone from ‘brain is toast’ to ‘big headache and am completely congested again.’ And of course, the way the day worked out, I was doing my most important work between 5 and 6 pm. I’m sure it is unreadable, but hopefully I got enough of the point across to do what I needed to do. More of same tomorrow before escaping to NYC for the weekend.

Spent the evening with brain turned off- lots of catching up on Daily Show, and some time reading the very awesome Presentation Zen. Makes me want to give talk again so I can practice.

Have wanted for a while to take a microecon course, so when I saw this CC-licensed Intro Econ textbook by a caltech econ prof I was pretty interested. He talked about printing it out in one of his pages, so I emailed him last night and suggested lulu- and voila, hard copy now available. Pretty cool to think that we really all can be publishers now, in every sense of the word.

Have spent a fair bit of time at work listening to Magnatune’s CC-licensed electronica stream. Will probably end up buying some CDs there after christmas. Probably makes sense for RB to include the various magnatunes streams by default in future releases, too… supporting CC and all that. I should get off my ass and file a bug :)

Mostly good couple days. Very productive at work (both for myself and hopefully for helping others at JOLT and abcd) and decently productive at home, moving the law school apps forward. Everyone I talk to in person at Harvard is sure I’ll get in (sadly haven’t seen the new dean of admissions lately ;), but I can’t help but feel that I look underwhelming on paper. The total meltdown my last semester+ of undergrad just looks worse and worse every time I glance at my transcript. I suppose that is part of the anxiety that is keeping me awake tonight- can’t sleep, so am trying to finish the poignant guide to ruby and listen to Sketches of Spain.

Am very tempted to do my next work project in Ruby on Rails, since it has been ages since I’ve tried out a totally new technology, and because for a variety of reasons my next project will likely only be live for a short time before it is replaced, so it doesn’t have to be perfect- just able to sustain a brutal slashdotting once, hopefully collect lots of data from those folks and others, and not die in the process.

Had a nice talk this afternoon with a gentleman from JOLT, HLS’s Journal of Law and Technology. JOLT is doing a day of speakers and panels in March, and are looking for interesting folks. I gave him some names of relevant folks in the open source, patent, and innovation spaces- sounds like they should be able to land interesting speakers, so hopefully I’ll be able to attend. As we were walking out the door he mentioned that they’ll also be talking about technology in education, which is my new hobbyhorse; hopefully I’ll be able to give them useful pointers there as well.

Have been using gnotime for two full work days now. Not perfect (I can’t figure out how to edit logs, which is a PITA since this evening I forgot to stop the timer when I left work) but still I think overall a huge success for me- both helping me understand how I spend my time and forcing me to think consciously about all my context switches and hence reducing them quite a bit. Lot less email processing during the day the past two days; lots more focused consciousness. Seems to have had a positive impact on how much I get done already.