Jump to content

Blog/YoBlog0802

From Wikimania

20060802, Yosofun's Blog (YoBlog)

4:23 AM. There are people frolicking about in the field adjacent to Dane, screaming like they’re jumping off the Grand Canyon… so I write to pass the time, in hopes of staying sane—that through writing I might somehow replenish the precious REM sleep that I’ve lost. Again and again.

After about three hours of discontinuous sleep, I forced myself to wake up at 9 AM. Although I’d been totally exhausted from forcing myself to stay awake most of the day to adjust to the schedule here, I didn’t really get to fall asleep until much later.

The living conditions at Dane Hall are… ghetto, in the purest of sense. It’s dorm life, but then again, it’s 0.25-star-rated dorm life. Once upon a time, I believed my term in a dumpy dorm with a splintered-broken table and a perpetually leaking roof was bad, but... Living in a monk cell that’s perpetually above 100 degrees Fahrenheit with only an open window and an old dusty fan to battle the impending heatstroke is definitely worse. The package of “Bad,” however, is not yet complete. There’s also the fact that the springs of the stained mattress I’m typing this entry on are probably as old as the building itself—such that every time you have you back on it for more than twenty minutes, you’re sure to get up feeling like your spine’s about to break up or your neck’s about to decapitate. It’s never quiet for more than about twenty minutes—twenty, if you’re lucky—since by day, there’s ubiquitous drilling from construction, and by night, there are these aforementioned screaming idiots convulsing outside. The 0.25 star’s really more like a quarter of a pity-star; I’m sorry to say that the place just doesn’t honestly earn it.

Ah, complaints. How great it is to simmer down my fury through passive writing in the quasi-permanent medium of blog instead of furiously typing out ephemeral words to be lost on IRC. But, given the longevity of my words, I might as well attempt to be useful, now, and document something more objective. Here goes what I can remember:

I attempted to find the OLPC room in Wiesner at MIT, but apparently they’d long ago ditched the place. I met some tall hitchhiker guy who can speak Afrikanese, whose name I can’t remember, who provided me with the useful information that OLPC ran off campus to some business building labeled 1 Cambridge on the MIT map. After a bit more of getting lost, unable to find the exact location of OLPC in 1 Cambridge, the hitchhiker guy called up someone who informed us that the Days are to be held on the 10th floor. Almost there. But, there’s actually a reception-desk “guard” guarding the entrance to the elevators. The guard in the building had an emailed list of names, and apparently, neither hitchhiker guy nor I am on it. Status now stuck, no longer lost. I decided to haggle the guard, since the list didn’t look that legit, anyway. The guard called up some guy on the 10th floor, who attempted to shoo us away. I guess neither of us looked like techies; I really should have brought more summer clothing, especially my collection of t-shirts with such remarks as “EMACS ARE BIGGER THAN BIG MACS!” So, we were on the verge of getting kicked out. Trekking through the humid summer heat past-dew-point all the way here, then nowhere, then back, really didn’t seem the best itinerary. Moreover, at this point, it would have been back, but with a growing migraine—ah, the pangs of sleep loss. I wanted in, since my migraines tend to disappear under the right kind of technical discussion. We finally figure out to summon S.J., who somewhat inaccurately validated our identity to the guy, but, nevertheless, the end result being the guy letting us in.

Hacking Days #2:

So we finally made it in. It was the middle of some guy’s talk on WiktionaryZ/Wikidata. Overall, the guy led a great discussion, but somehow my migraine didn’t go away. He kept on stressing on “defined meaning,” which is actually as self-explanatory as it sounds, but apparently most people in the room thought the term ambiguous. It’s, in a way, like how most people befuddle the difference between the language and the metalanguage, to use terms from the Abstract Logic grad course I took as a psycho freshman. Specifically, a “defined meaning” has to do with the proper defining of the nuances of a particular meaning in a particular language (hence, defined meaning). They do have their own format for data entry, however, but that’s really just trivial referencing up of the format protocol.

To my disappointment, they skipped the Entropy flow talk and went on directly to lunch. If whomever was supposed to give that talk started writing up equations of state, and doing proofs and derivations from thermo, my migraine might have lessened. (Thermo was from junior year, and I might have forgotten more than I should have, but then again working recklessly through physics proofs is a great way to stifle any migraine. Multitasking like crazy also works.) Except, however, I’m guessing it has to do with entropy with regards to information, modern stat mech. Still physics, but still, that’s probably not what the talk would have been about—I don’t see a direct relationship between that and WikiMedia. Hmmm… perhaps “Entropy flow” doesn’t have anything to do with entropy, but is really just some misused but somewhat catchy name for some obscure code. Well, whatever the talk might have been about, we shall never know.

The pyWikipedia project has to do with python bots written to do automated tasks in Wikipedia, such as spell-checking and link verification. Some of the scripts are rather poorly written, especially one of the earlier spellcheck scripts, and I wonder what sort of havoc they might have caused. The presentation was straight-forward, and there was not much to discuss. Since few people present have contributed to this, this presentation might have spurned further collaboration on the project.

I was actually starving by the time lunch started about an hour late. Still, I managed to refrain from piling up on massive amounts of sandwiches and wraps. (This was fortunate, since the wraps were way too oily, a source of a slight bit of indigestion. Then again, what had I expected? This was presumably the usual catered food as done by a large university, often prepared by incumbent students.) With food and water, I went back to my laptop, attempting to do a bit of reading to abate the migraine. I probably should have left and holed out in some study carrel in one of the libraries somewhere to catch a few hours of quiet sleep, but still, I’d rather be here to be, at least, somewhat here. Anyway, it didn’t work.

After lunch, the guys at Days got to let out gastronomical gas on a poor girl slated to give a presentation on usability to an audience of coders who probably comment their code only as an afterthought (and with dread, the sense of precious code time being wasted). Fortunately, though, the final version of MediaWiki that’s publicly downloadable is pretty well documented. Anyway, my point is that the guys at Days aren’t really frontend interface designers, but they’re coders, the guys who make the pretty frontend things work—so, the presentation seemed for the wrong audience. Then again, I suppose MW isn’t yet as segregated as to have people who focus only on designing interface. Still, much of what she emphasized on was that manual documentation for features like adding/uploading an image should be made more clear—like said earlier, since most present probably dread commenting their code, writing user-friendly manuals wouldn’t have been fit for them either. The guys heckled her a bit, but again, this might have been avoided if the presentation were given for the Mania part instead of Days. Then again, given the fact that her study only involved ten users, whose background seem poorly described, I don’t believe it’s ready for publishing. (The background ought to have included other information—such as whether the users have prior experience with HTML or another basic ML.) The part I distinctly remember happened in the middle of it all. Just out of the blue she was like, “That’s why people like me exist...” At that point, I realized she reminded me of my high school counselor... I don’t really understand people like her; why do they refer to themselves as such and yet not change it?

Next, there were Lightning Talks.

As always, the stuff I come for gets truncated or omitted. The BlahTex talk was basically a 30-second blah, mostly inaudible to audiences in the back. I heard something about LaTeX, but I actually didn’t hear the guy utter the abomination, “MathML.” I had to retroactively find out what it BlahTeX is by googling it up; it’s a LaTeX2MathML parser for MW. In sum, I’m not sure how useful it would be, since MathML has usability issues, whereas LaTeX (or the lightweight version of it, MimeTeX) doesn’t—the equations in the latter are rendered into graphic files easily viewable by all browsers. (Also, to those who do not display images, LaTeX2HTML conveniently quotes the tex equation in the alt tag; any other LaTeX to HTML parser can be easily modified to do the same.) Even using Firefox, which is supposedly a MathML-compliant browser, I have to go out of my way to download certain fonts (which actually come with Mathematica, which I don’t have installed on my borrowed laptop) to make it work. And even then, the equations look (almost) Microsoft-Word-equations-editor-ugly. Ew. In the beginning, Knuth made LaTeX, and it was good, now, however…

I had to leave to deal with indigestion during the LiquidThreads Talk, so I don’t know the exact details of that. It appears to be your basic thread commenting snipplet applied to the MW Discussions page. Other than a basic move-feature, I don’t see why it’s necessary, since users can already nest with colon-syntax to create threaded-like replies. Since I’m on a roll shooting out complaints, I have to say that the little I’ve seen of it, it doesn’t merit enough to be a SOC project—threaded commenting isn’t that hard to do and doesn’t take that long to implement. Making a multibrowser-compliant AJAX frontend for the commenting options, though, might.

MW Video support was interesting. It featured a neat javascript player that embeds (apparently) seamlessly into MW pages; each embedded movie initializes with an automated screenshot of a particular frame of the movie. Someone raised the question on why FlashVideo wasn’t used. This was where it got interesting. Apparently Macradobe now charges a fee for access to the .swf SDK (this is ‘’’what happens’’’ when Adobe buys Macromedia). (Incidentally, if anyone wants to view an old version of the SDK “for educational purposes only,” I can probably dig it up from one of my many dislocated hard drives; will take some time, so only if you’re actually serious.) I immediately cited OpenSWF.org—and then I heard from some guy in the back the situation on Adobe’s destruction of Macromedia’s old Internet2 ideology. Mentally, I was like, “no, no, that caaan’t be the case,” so I attempted to access OpenSWF. I found out they’d died, a long while ago, and then I forced myself to read the updated status on the SDK. Sometimes, it feels like I’m Rumpelstiltskin. Being away from something for six years is really more like a few centuries Internet-time. I am so impossibly old.

Finally, there was something on quote-notation for bold and italics. I was on the verge of revealing a very simpleminded way of fixing that. Avoid regex, just use a flag counter and strstr() twice, the flag being used the second time to generate the closing HTML tag, flag unset after the closing tag count. It would have parsed the example the guy gave correctly. ("'hi"hello"'hi"'hello"hi"' as {b}hi{i}hello{b}hi{/b}hello{/i}hi{/b}) There was apparently a point I missed about nesting(?), so I didn’t bother going for clarifications. It didn’t matter; by this time, my migraine had proven that it was here to stay. And, moreover, there was Siggy next!

We got the Exhibits Plus pass, which is better than just an Expo pass, but isn’t the full conference. (Then again, all the good lectures have already happened, since we’re coming in on the last day of SIGGRAPH… so w/e) The thing also came with a tour given by names known only to graphics-in-groupers. The tourguides hyped up the event to the innocent hackers (believe it or not, basically no one’s been to S) and implicated the SIGGRAPH conference caste system to be much more than it’s worth. I wandered astray from the flock only to find that basically all the customized-souvenir options (2d/3d printing) from the Guerilla Studio have been signed-up for days before. They did have a pile of clay available, to be used for scanning a clay model into a 3d model. So, I took that home to play with to take back tomorrow.

Since we came the hour before they closed, I didn’t get to see much, but I did get to find out in depth about realtime hologram generation using 6 standard web cams and a standard (though high-end) LCD screen with scrim. The only problem was that was that the software was probably written with a rather high-end language, since it took (ahem) 4 or was it 6 AMD 3.x GhZ’s to run. This totally deadens its commercial potential… unless Playstation4 turns out to be quadcore…

I wanted to basically hole out in the Boston Convention airconditioning until late night, but the guards and janitors basically kicked everyone out. (Never underestimate the janitor.) It’s always the same story; the janitors always kick you out at the end of the day. I got back and ended up in Pound Hall. I downloaded a few clay-modeling tutorials from the net and started playing with the piece of clay from the Guerilla Studio. (It’s always fun to try out new stuff, except I didn’t have the tools.) I did happen to model a piece of crap that looks vaguely like the bust of some demonic bald old man with a Neanderthal skull. I might fix it up if they have tools at the studio or then again I might just have that scanned up for sentimental value. After a few hours of that, another migraine came up, and I figured I ought to try to sleep.

Of course, you know how that ended up. Recall the 0.25 star rating of Dane Hall.

Envoi

The night ends with four armed campus policemen barging into Dane, 2nd floor at around 4 AM. Somehow, the artillery didn’t fit with the “Veritas,” patched on their sleeves; then again, truth is often the by-product of force and armed-threats. I thought I heard them knocking on my door, and in fact, I’d been lying awake for the past hour or so, stuck in indecision between trying to attempt to sleep or just waking up and risking the advent of some tremendous headache due to chronic sleep-loss. So, I let them force me into the latter decision.

I get up, and I open the door. The dorm doors of Dane are of a class with the campus police in that they stifle choice; they do not give the occupants the privilege of commanding the particular door to lock—instead, the doors lock automatically for you… such that if you take a leisurely but keyless 10-step walk to the kitchen and have your door open, a draft of wind can easily blow it shut, thus locking you out. (Observe, the brilliance of campus facilities.) So anyway, I find that the police are concerned about the guy down the hall who had been sleeping with his door open—a wise decision, especially in the arena of applied fluid dynamics, i.e., heat flow efficiency. Apparently, they thick he’s some vagabond who’d found his way in and wanted to kick him out.

The whole thing is just so outrageously stupid that I find myself yelling at the police. What surprises me is that I manage to yell at them without screaming and raising my voice to match the volume of the frolicking screamers outside. I tell them that what they ought to be concerned about is not about people having their doors open in a locked dorm floor, but rather, that there are these morons running around yodeling across the lawn for the past six hours or so. The police with the Hitler-mustache tells me that it’s hot and most people are outside (this is 4 AM), his voice sounding resigned, but then again he has on this pleading expression. It makes no sense to me why they’d be bothering some guy with his door open in the dorm but not caring about these screaming dorks outside. Who knows, these screaming dorks could be the very vagabonds they had feared and armed themselves to confront here at Dane Hall. (Then again, perhaps they are. And maybe the truth of these fool campus policemen prancing around with these false badges of “Veritas” is that they really wouldn’t know how to properly protect themselves if they’d tried confronting the real vagabonds outside.)

But, alas, there’s something wrong with my ability to convey my full point verbally these days. Instead of screaming my head out at how incredibly irrational and unjust they are in ignoring the grotesque H2O screams emanating from the field less than a long-jump away from the perimeters of Dane, I merely walk away passively, though not failing to glare in rage at them first, then shaking my fist at them like some aboriginal shaman casting a curse.