All posts by Blake Stacey

A Survey for Curmudgeons

I have a simulation happily grinding away in the background, using one core of my spiffy new dual-core system, doing my work for me, so not only do I have a moment to procrastinate, but also I should be happy about new technology. However, the headphones which came with the iPod nano I got for Christmas picked today to fall apart. The earbug doodad is beside itself with the joy it feels at being part of a cultural icon, I suppose. Given that the iPod itself had to be reformatted twice and connected to three different computers before it was able to receive music, that the interface packs more absurdity into its purported simplicity than I would have imagined possible, and that consequently it has relegated itself to the status “device which plays “Mandelbrot Set” on demand,” having the headphones cheap out on me is rather like salting the fields after Steve Jobs has burnt the city.

All this to say that today I’m in a mood for appreciating old things which work.

Geoffrey Pullum wrote, four years ago,

Shall I tell you how The Cambridge Grammar of English was prepared? (I am not changing the subject; trust me.) The book is huge: 1,859 printed pages. The double-spaced manuscript was about 3,500 pages (yes, it actually had to be printed out and written on by a copy editor the old-fashioned way). It took over ten years to write. And it was done using WordPerfect 6 for DOS. Rodney Huddleston chose to upgrade to that around 1989, wrote a couple of hundred complex macros, and stuck with it. I learned the WP DOS macro language in order to collaborate on the project.

WordPerfect was basically in its final, completed form before Clinton first ran for office. It works. The file format is fine for authors, and records everything we need to record. Rodney and I are still using WP6 file format today to write our planned student’s introduction to English grammar. In all the years since the late 1970s, WordPerfect has not altered the file format: all the largely pointless upgrades in the program have been backward compatible. The format really does the job. But things are different with the WordPerfect program itself. The progress has largely been backward.

The things we have noticed about version differences are minor, but they all tell in the same direction: every upgrade is a downgrade.

Forget the Clinton administration: TeX basically solved the problem of representing mathematical equations as text, during Reagan’s first term. The LaTeX macro language, which handles document-scale organization, is almost as old. Perhaps we’re stuck at a local maximum, and with luck and pluck we could find a better way, and on some days, that seems almost mandatory. Still, we’re at a pretty darn good local maximum, as local extrema go.

(Something deep within me finds a resonance with PyTeX, an attempt to have Python sit on top of TeX the way LaTeX does, but the project seems to be moribund.)

The question for today, then, is the following:

What are your favorite Old Things That Work, and which changeless relics really do need a shake-up?

Previous surveys:

Comments on all the above remain open.

Geek Fuel

A couple noteworthy items:

Tim Farley, who spoke at TAM 6 about Internet tools which skeptics can use, has put the transcript for his talk online. There’s a great deal of stuff in there to geek over; we can at least have a lot of fun while we’re failing to save the world. Farley makes me feel shame at the weakness of my RSS-fu.

Via the diligent Peter Suber comes word of Open Education News, an aggregator site for open-education developments (new teaching tools, textbooks which are free as in speech or as in beer, etc.). It’s interesting to see how much younger OE looks and feels than OA, with none of the crotchety arguments about definitions and fireballs hurled at traditional enemies.

After Patchwork

The spores of discontent have infected the nervous system of Warren Ellis, driving him to higher ground so that they may burst open his skull and spread their meme sequences upon the wind:

That’s been the job of half the web, for the last several years — collating links from the other half of the web. Last year, I started getting a little itchy about this.

Wouldn’t it be nice if we could stand up now and say, okay, these are the post-curation years? The world does not need another linkblog. What is required, frankly, is what we’re supposed to call “content” these days. When I were a lad, back in the age of steam, we called this “original material.” Put another way: we like it when Cory and Xeni are the copy/paste editors for the internet, but we like it better when Cory writes a book and Xeni makes an episode of BoingBoingTV.

(In fact, if you read any of the abhorrent comments threads on BoingBoing, you could be forgiven for coming away with the notion that its readership would be happy if it shut down tomorrow.)

Trying to create “original material” does occasionally leave me with the impression that I should take lessons from my indie-rock friends on whining, “Won’t anybody listen?”

Non-Missing Metropolis Footage

And while we’re talking about classics being reinvented, here’s a bit of cinema news: it looks like additional footage from Fritz Lang’s Metropolis (1927), lost since the original version was pulled from the Berlin theatres, has turned up again. According to Die Zeit, “The rediscovered material is in need of restoration after 80 years; the pictures are scratched, but clearly recognizable.” The Guardian reports,

The uncut version is said to solve the mystery as to why Maria, the workers’ insurrectionist leader, is mistaken by a baying mob for her doppelganger, a female robot.

OK, cinema buffs, here’s a question: did you find this part of the movie problematic or difficult to understand? I, for one, missed whatever is supposed to be the matter here. (Mobs are not noted for their fine powers of perception.) Perhaps the next time I watch my Metropolis DVD, I’ll smack my forehead and exclaim, “How silly of me not to have noticed that gaping hole all along!” Still, I’d be much happier to find out why Joh Fredersen wanted to let the mob into the Heart Machine. Did he just have a fantastically poor grasp of the consequences?

Extra details from the Guardian:

Schmale, a spy who is sent by the autocratic leader of the futuristic city, Joh Frederson, to pursue his son, Freder, plays a minor role in the cut version, but a significant supporting role in the original. “The role … can finally be understood,” Rother said.

A scene in which children are saved from the workers’ underworld is also said to be “much more dramatic” — and more violent — than in the cut version.

Yay violence!

Being the sort of nerd who listens to DVD commentary tracks and watches all the special features, I recall that the contents of many of the missing or damaged portions of Metropolis were deduced from the film’s censorship certificate. This document, from the Head Office of Film Censorship, listed the contents of all the title cards (I presume to certify that the film was acceptable for viewing audiences). It looks like the newly rediscovered footage thus contains either ambiguous dialogue or none at all.

Two Bardolatrous Items

I was in the basement of Building 14 several weeks ago, where MIT hides the “humanities” books, and I found the following:

The notion of Shakespeare entertained by an age affords an index to thought in general. If men re-create God in their own image, they are constantly remodelling their effigy of him whom they insist on regarding as the most God-like of men. Après Dieu, il créa le plus.

— Hazelton Spencer, Shakespeare Improved (1927)

Most of the things I’m seeing on the Internet these days have been making me go all stabby. I want to find press agents and revoke their science licenses. I want to smack George Lakoff for spreading neurophrenology and mainlining stupid into what should have been academic discourse. I want to make Ben Goldacre release the photos. . . yeah, he and PZ know what I’m talking about.

But I can’t do any of those things, so I’m driving them out of my mind with Chimes at Midnight (1965). Damn, but that soundtrack is catchy.

In Which Blake Fails

Not noticing the tiny, unobtrusive switch on the side of your laptop which is labeled “WIRELESS ON / OFF” and wondering why you are no longer detecting any WiFi networks: FAIL.

Trying to troubleshoot your switched-off WiFi by digging through kernel module configurations: FAIL.

Attempting to connect using the Ethernet card and a hub which turns out to be non-functional: FAIL.

Finally switching the WiFi to the ON position, connecting to the Internet and realizing, “Hooray, now I can get back to work on that proceedings book for the conference which happened four years ago” — EPIC FAIL.

Currently Reading

Xiaotie Denga and Li-Sha Huang (2006), “On the complexity of market equilibria with maximum social welfare” Information Processing Letters 97, 1: 4–11 [DOI] [PDF].

We consider the computational complexity of the market equilibrium problem by exploring the structural properties of the Leontief exchange economy. We prove that, for economies guaranteed to have a market equilibrium, finding one with maximum social welfare or maximum individual welfare is NP-hard. In addition, we prove that counting the number of equilibrium prices is #P-hard.

Found by citation-hopping from here.

Pretentious Document Format

I Love the Blogosphere!I love the Interblagotubes.

Where else could you dump thousands of grumpy and disaffected words on unsuspecting readers and have them come back asking for more, and in a more easily printable format? Seriously: if you know a better venue for long-winded, gloomy rambles, I’ll go there instead.

Recently, Scott Hatfield asked me if I could convert one of my lengthier essays into a PDF or some other such format. In the spirit of excessive and unwanted generosity, I did a quick-and-dirty conversion job on not one, but two posts. Here they are:

The next time I need to procrastinate on something dreadfully important, I might try this again.

Friday Geek Update

My aged and broken laptop is still broken and has not grown any younger. Moreover, the USB key on which I had a decently recent backup of my work appears to have died as well. Furthermoreover, the server on which I also had my work backed up is suffering from a bum RAID array. Mission for today is to extract the drive from the old laptop and wire it directly into the dilithium recrystallization coils — er, I mean, connect it to my new Sony VAIO C420. I note that Micro Center sold me a laptop with Windows Vista on it, but I forgive them, since Ubuntu gutsy (the installation disc I had on hand) installed without any trouble. Audio, wireless and all those goodies worked without extra effort; I haven’t yet had much success with the Bluetooth support it automagically detected, but the only device I’ve had to test it with has been a cell phone which doesn’t play well with anything else, either. I found a status-bar tool which displays the current weather conditions as reported on the Intertubes, and unlike the previous version I’d used, this one can display temperature in kelvins. A year in Lyon followed by a change to my laptop settings went a long way to making me “internally metric”; this may be the logical next step.

(By the way, I booted into Vista just once — so I could say I knew the enemy, and all — and it sucked. It took the duration of an entire Pinky and the Brain episode just to decide how best to phone in to the mothership and report the music library I hadn’t yet put on the blasted thing because I’d just taken it out of the box. Neil Gaiman was right to consider XP an upgrade.)

All that aside, it is now Friday afternoon in Cambridge, Mass. (which is across and down the river from Newton, Mass. — there’s gotta be a physics joke in that). Outside, it’s a partly cloudy 302 kelvins. Inside, it’s time for the Dandy Warhols, with “I am a Scientist.”

Incidentally, we like to have music playing while we cook dinner here at Château Sunclipse, and this was the song we had going when we discovered that enchilada sauce with a dash of hoisin made an excellent base for beef soup.

What Science Blogs Can’t Do

No cosmic law says that when you gaze into your navel, you have to like what you find.

My thesis is that it’s not yet possible to get a science education from reading science blogs, and a major reason for this is because bloggers don’t have the incentive to write the kinds of posts which are necessary. Furthermore, when we think in terms of incentive and motivation, the limitations upon the effects of online science writing become disquietingly clear. The problem, phrased without too much exaggeration, is that science blogs cannot teach science, nor can they change the world.

INTRODUCTION

Notice how short the “basic concepts in science” list is, compared to the “basic concepts” which we know are the foundation of our fields? It has eleven entries — count them — for all of physics. Translated into lectures, that might be a couple weeks of class time. Chemistry is even worse off, and while the biology section is big, it’s also remarkably scatter-shot. Such introductory lessons as get written don’t get catalogued, and thus become damnably difficult to find again.

And, the problem hardly stops there. As the magician Andrew Mayne recently pointed out,

People only know what they can understand. There’s a lot of great information out there, but not enough is being doing to make it widely accessible to the masses. Most science entries in Wikipedia read like they’re written by graduate students for other graduate students. Even the basic science stuff is written that way.

We need to put ourselves into the perspective of someone who hasn’t had the science exposure that we’ve had and find ways to help make this information more accessible.

Why is introductory material so poorly represented?

Well, what do we science bloggers write about, anyway? This is how I caricature what I see:

0. Fun posts about random non-science stuff — entertaining, humanizing, but not the subject I’m focusing on right now.

1. Reactions to creationists and other pseudo-scientists.

2. Reactions to stories in the mainstream media, often in the “My God, how did they screw up so badly” genre.

3. Reports on peer-reviewed research.
Continue reading What Science Blogs Can’t Do

Woe

Las Vegas is a town of bottled water, not just because they’re hawking it on the street corners, but on general principles: the city takes things which should not be encapsulated — risk, chance, sex, scenery — packages them and sells them at exorbitant prices. I’m glad I’m out of it, except that on the way home, I was caught in a monsoon downpour which lasted almost exactly the duration of the walk from the T station to my front door. The next day, my trusty laptop developed an interesting new behavior: when I turned it on, it turned itself off. I suspect the containment around it failed, to use a Star Trek-ism, and its power supply had a delayed allergic reaction to the rainwater. It remains to be seen whether I can extract the data from its hard drive.

Fortunately, the so-called social circle in which I move is a giant geek support system.

While I try to get my act together, Maria Brumm has a couple good posts on the troubles mathphobia causes in a geology education. To an aspiring high-school science teacher, she writes,
Continue reading Woe