So I finally updated x.org to 7.0 on my HTPC. I had been putting it off since the crappy proprietary matrox drivers required to use the card are poorly-maintained and probably wouldn’t be updated to install to the new file locations, if they worked at all.

Anyway, I did the 7.0 update, futzed with the ebuild to install the drivers to the right location, and started up X. Unresolved symbols in the video driver. Great. So I do some questionable manual shared library linking, and X finally starts. And then crashes right after showing the video card’s splash screen. Great.

I found a newer, unofficial build of the driver made by some random dude (a fact that points out how much Matrox’s support for the driver sucks), but he used a funky shellscript installer, and getting at the files easily is a pain since for some reason the –target cmdline option doesn’t work. Stupid. Anyway, I got tired of trying to force it to work.

So, I’m fed up. Matrox is now on my shitlist for Linux. I just ordered a $40 (minus $15 mail-in rebate; yay!) nvidia geforce 6200 on newegg, and that’ll be it. I’ll probably put the 6200 in my desktop, and move the 5700 in my desktop back to where it started in the HTPC. Or maybe I’ll just put the 6200 in the HTPC, because it requires less effort.

So, unless I want to downgrade back to x.org 6.8.2 (I don’t), my HTPC is a very large doorstop until the new card comes. Well, actually, it’s also a cross-compile box. I finally figured out how to get crossdev to properly compile a cross toolchain (including glibc and g++, which I was having trouble with before) on my two x86 boxes, so now I can compile updates for my ppc PowerBook on three different machines. Which is good, because the PowerBook is kinda on the slow side.

If anyone wants a Matrox Millennium P650 for about $60 (pricegrabber says they cost anywhere from $120-$230), let me know. It’s a pretty nifty dual-DVI card (no VGA at all), aside from the shitty Linux support. Otherwise it’s going into my Pile O’ Unused Crap.

Cardinal Poupard, head of the Vatican’s Pontifical Council for Culture, says:

This is a shocking and worrying cultural phenomenon that reflects, on the one hand, the ignorance of millions of people and, on the other, the voluptuous pleasure the media take in promoting products that have nothing to do with the truth.

Irony surrenders. It just can’t take any more.

I present Exhibit A.

I was reading a post by Luis, who has just joined Facebook, after trying Orkut and finding that more or less no one in his social circle actually uses Orkut.

I’m most active on Facebook myself, though it’s hard to really say why. Part of it might be that it was the first social networking site I used (aside: for some reason I resisted Friendster and the others like the plague). Some time after joining Facebook, I also joined Orkut and Friendster, though that was because I was invited to by friends; I didn’t seek it out myself. I don’t quite recall if I found Facebook myself, or if a friend invited me. Probably the latter.

I guess Facebook is more relevant to me since it has a college focus (though they’ve expanded that to include high schools and employers). Granted, I’m not in college anymore, but more than half of my closest friends are still in college or still related to one in some way. (Hah, grad school: suckers! ^_^) Or maybe I just never gave Orkut or Friendster a chance since I was devoting energy to Facebook. On the other hand, I have around 100 ‘friends’ on Facebook, and less than 20 each on Orkut and Friendster, and I don’t think I’ve actually invited anyone on any of the services. So all my friends on any of the three are people who were already on the service that I found by searching, or who joined and found me.

So I guess, either by chance or by design, more people in my social circle tend to end up on Facebook than the other popular social networking sites.

Unfortunately, though, I don’t really get the feeling of any kind of participation level. I tend to participate passively: I’ll check out my friends’ pictures when they post new ones, skim their profiles when they update them, etc. I join ‘groups’ on Facebook not because I want to participate, but because it seems cool or funny, or it’s a topic I identify with. Even if I wanted to ‘participate’, I’m not even really sure what that means.

So I see Facebook as a window on some friends I don’t really keep in touch with as well as I’d like, as well as a way to reciprocate and let people know what I’m doing. But for the people that I see often in person, or talk to regularly on AIM or IRC, Facebook really does nothing for me.

Now, what would be cool is if there was some involvement with OSS-related people that I know. Unfortunately, I’m pretty sure Facebook is more or less US-only (though I have seen Oxford and Cambridge on there), so that leaves out the majority of the Xfce and Lunar guys I know.

I dunno. While it’s somewhat fun to be a part of these little online communities, I don’t feel like my life would be in any way diminished without them.

Clearly the best answer to a geometry problem ever.

I was reading p.g.o today, and came across this little snippet from Murray Cumming:

Show how much better [Save] is than [Yes] in a “Do you want to save changes?” dialog, with a “Really Discard Changes?” dialog as the punchline. This emphasizes our attention to detail, and to the user experience, so they don’t need to pay attention.

I’ve given this a bit of thought on various occasions over the past 8 months or so, and I can’t really come up with a conclusion as to which I like better.

I think, for users unfamiliar with a new piece of software, using descriptive titles (“Save” instead of “Yes” in the above example), is good: it helps prevent possible data loss, and removes any amount of confusion. It also helps in the case where multiple applications have different conventions: perhaps one says “do you want to save changes?” and another says “really discard changes?”, and you haven’t used either of them enough to quickly remember which is which. Having a “Save” button in the former dialog, and a “Discard” button in the latter quickly disambiguates and avoids confusion. (Although, you could make the case here that app developers shouldn’t use the negative form ever, and should always use the more positive, data-saving “do you want to save changes?” form.)

However, for the advanced user with a good memory, this is a pain in the ass. I know which apps ask what on close. For some applications, I get into the habit of using the keyboard to press the buttons. When I want to quit the app and ditch what I’m doing, I hit ctrl+q, which then brings up the “save changes?” dialog, where I press ‘n’. Or if I do want to save, I hit ‘y’.

Now, maybe that’s not such a big deal. I guess it’s ok to remember to hit ‘s’ or ‘c’ instead.

But the situation changes when you bring in a bunch of other dialogs. Perhaps I’m using a file manager. If we had the “old” way, I’d have ‘y’ for affirmative actions (“yes, delete that file”, “yes, move that file”), and ‘n’ for negative actions. Now I have a bunch of different keys: ‘d’ for delete, ‘r’ for rename, ‘m’ for move, ‘s’ for save. And it’s not just the file manager: these dialogs with a bunch of different keyboard shortcuts are a pain in the ass. Not to mention that there are still “old-style” apps around that use ‘y’ and ‘n’.

So I’m still on the fence. Designing for general usability seems to indicate that the new way is better, so that’s what I’ll follow in any apps I work on. But a small part of the advanced, keyboard-shortcut-loving user in me dies every time.

I’ve noticed recently that a lot more TV shows than usual are doing story arcs instead of being composed mostly of standalone episodes. Which I like, because I enjoy the idea of an ongoing story that can’t be wrapped up neatly in 43 (or 22) minutes.

However, it seems that the writers (or producers, or whomever) seem to think that every episode in a story arc needs an extensive 2-4 minute “previously” segment where they recap what’s happened in the story thus far. Is it really necessary? It’s annoying, and wastes precious time that could be used for actual story development. Take a hypothetical half-hour show, which really has 22 minutes of actual airtime. If they spend 2 minutes at the beginning doing previouslies, that’s 9% of the show they’ve wasted. Lame.

And are people really that stupid and forgetful that they can’t remember what just occurred last week? I can understand a short “previously on” segment if the show has been on hiatus for a month, but week-by-week?

I suppose there’s the issue of people who missed last week’s episode, but really, why waste time catering to those people? There are plenty of ways to catch up on missed episodes (legal and technically-illegal), and most popular series have their episodes re-aired later in the week.

Then again, I do like that more series are doing the extended story arc thing, so I guess I shouldn’t complain too much.

It doesn’t happen often, but sometimes there’s a truly insightful comment in a Slashdot post:

Honestly, I think virtual worlds will set us free and give us the strongest dose of reality check we’ve ever experienced. After a while you notice that you are valuing utterly imaginary things above actual real things and then you start thinking, “Well, Jesus. What is the value of real things? Maybe the ‘real’ things in my life aren’t even real. Maybe the real things I bought are just as hollow as so many bits on the ether. Maybe that’s a problem that I should address.”

Or maybe it won’t turn out that way for most. My perspective: there’s as much virtual crap at the local shopping mall as there is in the Flavor of the Year online game. It’s all the same hat.

Cool, I have mod points today.