Friday, December 16, 2005

Don't Panic

A recent study determined that Wikipedia articles average 4 errors per entry, Brittanica articles average 3. This is an interesting observation, although we should be cautious; though it's a difference of only one error per entry, that means there are 1/3 again as many errors in Wikipedia as a whole. Of course, Wikipedia still has the benefit of being instantly editable. It may not be more accurate at this instant, but as academia becomes more comfortable with online resources (inevitable, as more and more struggling university presses turn to e-texts), I'd like to see more experts trolling the wiki, making sure the articles in their own field are up to speed.

Anyway, a friend posted in his livejournal the following quote, posted on a forum conversation about Wikipedia:
In many of the more relaxed civilizations on the Outer Eastern Rim of the Galaxy, the Hitch Hiker's Guide has already supplanted the great Encyclopedia Galactica as the standard repository of all knowledge and wisdom, for though it has many omissions and contains much that is apocryphal, or at least wildly inaccurate, it scores over the older, more pedestrian work in two important respects.

First, it is slightly cheaper; and secondly it has the words Don't Panic inscribed in large friendly letters on its cover.

--Douglas Adams, The Hitchhiker's Guide to the Galaxy

That about says it all.

-.. --- -. - -.-. .- .-.. .-.. -- . .- --. . . -.-

I also went through the "talk" page associated with John Seigenthaler's Wikipedia article. The article has a section on the controversy now, but someone had apparently tried to add a line at the beginning saying that "He is perhaps best known for his recent attacks against Wikipedia, which have resulted in new restrictions being placed on unregistered contributors to the encyclopedia." The addition was nixed by editors. The writer tried to protest, and when other said rather haughtily that he was known just fine thank you "among people who follow First Amendment issues closely." Another poster snapped back, "I know I didn't check out his page because of his stances on first amendment issues; did you?"

It's true. In an attempt to save his reputation from people who hadn't looked him up enough to notice that his article was false, Seigenthaler has made himself look much worse than the (now no longer anonymous) Wiki vandal could. I know I'd never heard of him before, and I'm a relatively educated person. Now I even know how to spell his name.

I really need one of these

I desperately need a bacon alarm clock.

Failing that, the Ig Nobel-winning Clocky. It would have a field day in our room, which is currently quite overfull. (Bonus: The creator is hot!)

Failing that, Wallace's machine that dumps you out of bed right into your pants. I did used to have a Wallace and Gromit alarm clock; I wonder what happened to that.

Wednesday, December 14, 2005

Mashup, parts 2 and 3

First of all, I really wish someone could tell me why I used valuable time that should have been spent on my paper photoshopping this picture of Dan's best college friend (the squid one, not the Camus one). One of my quitting-grad-school fantasies involves getting hired to make photos for the Weekly World News (never mind how far downhill they've gone in recent years -- what's with all the bad puns? If you scan the masthead and bylines it's clear that somebody's brother/nephew/son is getting way too much free rein).

This dovetails with a bit of inspiration Dan and I had the other night about naming the new incarnation of the band. (Right now it's Dr. Rocksinger and the Age of Longing, but that was a joke originally and is a little too ridiculous to stay.) I wanted Slide Back, but Dan thought it heralded the wrong kind of sound. Then I thought Phi Ratio would be a good one, but Dan worried that it might attract New Agers. Then inspiration struck: Photoshop filters. Seriously, picture any of these stencilled onto a bass drum:
Difference Clouds
Dodge & Burn
Neon Glow
Reticulate Splines
3D Transform

Some are better than others, but almost all of them work -- leaving out things like "Smart Blur" and "Angled Strokes" that might get you a cease and desist.

-... .-. . .- -.- .-.. .. -. .

In the other part of Dan's blog post, under Trent's picture, he has a really kind of excellent bit of literary spam and the question, "are spam filters genetically optimizing computer generated text?" I thought that would be a perfect jumping-off point for something I've wanted to discuss for a while: Darwinian Poetry.

The Darwinian Poetry poems started out as randomly generated chunks of text (they sound a little like "Paradelle for Susan"). An example:
clothe mine you
kind peculiar or
day in it I me
the believe eddy glory though a
holidays in of

These randomly generated poems were pitted against each other in pairs for readers to choose between. Through a genetic algorithm, successful poems would pass on "traits" (sections of poetry) to the next generation. (You can find a description of the algorithm on the site.)

The refining process is still going on. On my last visit, for instance, I was faced with a choice between poem 19122:
waves sang

by galaxies horizon
iron dark stars
stars derailing sure revealing
one dream
a love just water

and poem 16912:

first snowfall came in believing

in iron
with cold knowledge

on sweet torrent
light searched
made paler

It's no Marianne Moore, but there's no doubt about it: The poems are improving. They're more grammatical, and occasionally they're quite lovely.

What does this prove? Maybe nothing, but it indicates a few things. First, that if we make an analogy between poems and organisms, then their proving ground is the reader's mind. If clusters of words evolve into okay poetry when faced with a critical reader, then the reader must be analogous to a hostile environment, one to which the work must adapt. None of this authorial-intent nonsense: The development of poetic species has depended on pleasing the reader.

Second, randomly generated writing can "improve" (in regards to our usual standards for a good piece of writing) without deliberate human intervention. So yes, Dan, there is a Santa Claus... I mean, there is (maybe) self-evolving spam literature. If our increasingly draconian filters kill off all species that don't meet standards of interesting human writing, then yes, we're breeding a race of hardy little mutated literary pests. The difference between this and, say, a race of hardly little mutated bird flu virus is that the standards for spam survival are based on how well they mimic decently-written, grammatical, non-penis-enlargement-obsessed human expression, so the ones that "evolve" to slip through the filters are actually pretty interesting rather than deadly. I can't help but think of it like mimic camouflage, though -- the spam avoids filter predators by evolving to look just like the other messages in your inbox, like a moth blending in with the leaves on the forest floor.

Tuesday, December 13, 2005

Now playing: Resistance

One more post before I take my nap, because this one is time-sensitive: Today is Dean Grey Tuesday (link goes to Boing Boing), a day for trying to save the "American Edit" Green Day mashup album. Warner banned the album ten days after it came out -- unsurprising, given the impossible row mashup has always had to hoe, but collage music is an important creative force and deserves fighting for. If you like Negativland, if you like the Evolution Control Committee, if you listened to Danger Mouse's "Grey Album" (I didn't get to before it was suppressed), or if you're just in favor of using intellectual property law to stimulate rather than stifle creative expression, go download this. It is important, but it's also cool.

As long as it's about writing critique

Here is my final paper on the need for a digital archive for Christine de Pizan's Epistre Othea (that's the doc, here's the html). This represents my first attempt to write about anything so practical as actual text encoding, so please, XML/TEI experts, correct my bare-bones explanation. Especially those of you who were actually involved with the Blake Archive, which I deal with at length.

Panic kicked in weirdly early for this one -- I started worrying on Friday, even though it's not due until Wednesday, because I just didn't bother to count days -- which means that I managed to finish it on time even though, for instance, I spent almost all of yesterday sleeping. Now that it's done, I'm going to reward myself with a nap. Why is graduate school so tiring? (Okay, I know it's just me and my voracious appetite for sleep. And my need to refill my Lexapro scrip.)

Anyway, this is due at 5 tomorrow, but if there are major edits suggested, it's okay with Theresa if we turn it in Thursday. You'll notice a couple of incomplete footnotes because the Hindman book is in my car somewhere (oh, and one or two that I needed to find URLs for, which are completed now but aren't in the uploaded version), but I'll get to that today or tomorrow.

I hope I've made a friend and not an enemy

In a fit of writing-teacher enthusiasm, I responded to a post by the writer responsible for the Johnnie Ray lie detector article, who was simultaneously asking to be critiqued and begging not to be criticized. It's a conflict I'm familiar with, having just given out my teacher evaluation forms, and consequently having just told students that it'll be more effective to write "this, this, and this didn't work" than to write "please fire Jess," because they're not going to fire me. But of course, I don't want to hear "this, this, and this didn't work" either. I want to hear "Jess was the best teacher ever and I write good now. Well. Sorry, I write well." Anyway, so I feel for Maggie, and hopefully I didn't come off overly sanctimonious, since I took the request for useful criticism at face value and used my "end comment voice." If I used that voice and still came off sanctimonious, I kind of need to know, since that's how I talk to my students and I don't want them thinking that I'm patronizing them.

This reminded me of a few important things. First of all, to quote my least favorite MA project source, "the net is vast and infinite." If you're going to talk shit about anybody, even minor shit that you think is harmless, they're probably out there and they can probably read it, if they're working hard enough. I've been poking fun at that lie detector piece for a while, and it never occurred to me that its author was inevitably out there and wanting to know what the hell was so wrong with her introduction anyway. I should probably earn my right to disapprove of writing, by only making fun of students' writing that I've already had to grade, critique, and hopefully improve.

Second, I wasn't sufficiently clear in my post about the level of blame I place on editors. Most of the people who read this know how I feel about writers and editors, but see point one; not everyone who encounters this blog will know me. So let me be perfectly clear: If a magazine has dug itself into a hole to such a degree that it needs to hire inexperienced writers, the editorial staff must be cognizant of that fact, and must keep the quality of writing under control. Otherwise, they're making not only the publication but the writers look bad -- and young writers' reputations can't necessarily brook a less-than-stellar portfolio entry. Seed is not only doing me, the would-be reader, a disservice when they fail to put out a worthwhile publication. They're also harming their writers by giving them insufficient editorial guidance (though now at least they pay, I guess). As a writing teacher, I can't condone that.

See point one: probably a Seed editor will show up now and want me to answer for my sass-mouth. Bring it on, [hypothetical] buddy; I'm after your job.

Saturday, December 10, 2005

If I get a B, blame Boing Boing

Thanks, Boing Boing, for alerting me that Samorost 2 is out. I just killed a good hour playing this when I should have been working on my Christine de Pizan paper. If you played the first one, this one is just as beautiful, with equally spooky music. The puzzles aren't my ideal -- a little too much sharp-eyed searching, not enough problem-solving -- but it's lovely to look at, and for those of you who don't do puzzles, it's a little brain exercise to ward off dementia. They're trying to push a full version, which is a little irritating, but hey, you don't usually get games this pretty for free.

Tuesday, December 06, 2005

The best book ever, free

Although I'm not sure why the blog post title is International Whatsits for "THF," Francis Heaney has announced some super-important news (also disseminated by BoingBoing): He is putting out one of the funniest books ever written as an eBook. Props to Francis for recognizing that free access will only help his sales. He realized that people frequently buy his books in twos, one for them and one for a friend (I did exactly that, in fact), but that this only happened on the infrequent occasions when people ran across it; making it available as an eBook, then, might reduce the "for me" sales but would increase the "for friend" sales by increasing the book's overall profile. Smart marketing, or "smarketing" as we like to call it.

Of course, it helps that Francis' book is eminently irresistible. This is further proof, then, that anybody who gets their panties in a twist about stringently controlling intellectual property is talentless (Celine Dion, anyone?)

Beware! User-editable information!

Laura brought to my attention this article by John Seigenthaler, pissing and moaning about his falsified Wikipedia entry. Her comment: "This guy is smart enough to track down an IP address as well as Wikipedia's founder, but he's not smart enough to click 'edit this page'?" I think it goes even beyond that. It looks to me as though he went after Wikipedia's founder not because he wasn't smart enough to figure out how to fix the information, but because he was more concerned with publicly showing that the info was wrong than with correcting it. But, of course, his personal vendetta got turned into a parable against internet research.

Today, the nominal director of the Freshman Writing program (her predecessor will always be the real Freshman Writing director to me) sent out a link to this NYT article, suggesting that we use it to make our students aware of the dangers of online information. Now, granted, I have an entire class period on carefully evaluating your online sources, although my real object lesson is not Wikipedia but But Wikipedia, I usually tell them, is okay. It's not the most high-ethos source, sure, and they should double-check any information they find there, but it's a fine place to start. Why? Because while misinformation can be found anywhere (including books, guys), and is more easily found the less editorial oversight you have, Wikipedia has a million eyes on it. It may not have official editors, but it has as many potential editors as it has users.

To work, though, it requires that people who find misinformation fix it, whether or not they feel personally slighted. This guy, out of his own umbrage and paranoia, is breaking the system that allows Wikipedia to oscillate slowly towards reliable fact.

The fact is, Wikipedia is peer-reviewed. In the print world, there's a well-documented process of error correction: hopefully one of the editors catches it in a draft stage, and if not, there's a process for issuing errata, publishing new editions, or -- if the author and the editors are just really wrong -- responding with your own article or volume. With Wikipedia, that's streamlined. Don't like it? Fix it right away. Does that mean there are more errors in WIkipedia, or does it mean that we're just more comfortable with the clunkier process of print accuracy? After all, responding to an error in a print book is part of intellectual discourse, not cause to question the whole field of publishing.

Now, granted, we're talking about libel here, not garden-variety inaccuracy. (At least, we're supposedly talking about libel... the lady doth protest an awful lot for me to really believe there's no truth in the original biography.) But again, isn't libel to some degree an artifact of print? Can something really be considered libel if it can be removed by the person to whom it refers, or for that matter by anyone? Supposing I wrote "John Seigenthaler is a whiny prink" in pencil on a bathroom wall? Supposing I peed it in the snow? (Okay, I can't, but you get the picture.) Wikipedia is correctable with near-zero effort, so can a libel accusation really hold water?

The real problem, as I see it, is with other sites lifting WIkipedia's entries. This effectively takes a snapshot of something that needs to be dynamic in order to be reliable. Minimizing that process is something I could get behind (though considering the way Wikipedia works it'd have to be done by honor system or Google-combing volunteers). Condemning a user-editable site for ephemerally presenting incompletely accurate information? That, I can't.

Monday, November 28, 2005

It's a shame about Seed

I was thrilled when I heard that Seed magazine was coming back from bankruptcy. You have to love a publication with the tagline "Science is Culture," especially when you spend time arguing from exactly that position. It's become abundantly clear that I need to update my scientific knowledge, and Seed used to be an exciting way to do that.

What I forgot was that when the magazine went out of business, it stiffed a lot of good science writers. I mean, that's what happens when you go bankrupt... you owe money all over town, and nobody in town gets it. So everybody in town is pissed at you. In short: Seed can't get any decent writers, since it owes its previous contributors thousands of dollars. It now reads like it was written by some of my less skilled freshman comp students. Even my normally brilliant boyfriend was stumped; I asked him to explain a short article about information escaping from black holes, and he could only give me a quick precis of Hawking radiation before having to say "if the article were written better, I might be able to tell you what they were actually getting at."

It's a little hilarious -- my mom forwarded me one faux-noir intro to an article on lie detectors that was comical in its awfulness -- but I don't want it to be hilarious. I want it to be good, and it's not. It's hackish. Call to arms, publishing industry: Make me a "science is culture" magazine I can actually bear to read.

Friday, November 18, 2005

Thanks Jill!

My new favorite hypermedia theorist (and unlike when Scott says it, that's not code for "girlfriend") pointed me to a discussion going on at WRT about evolution vs. intelligent design theory of electronic literature. I was worried that I should be worried, but it's just another reflection of the way that the science and literature paradigms are interpenetrating -- and e-lit, unsurprisingly, is more sensitive to scientific shifts. That said, it seems to me that framing it as "evolution vs. intelligent design" ought to be a no-brainer... if we consider textual approaches to reflect the scientific zeitgeist, then we should have left intelligent design behind. The question, instead, should be how literature evolves: Saltatory? Darwinian? Memetic?

Even that, though, strikes me as having too broad a scope. A literary work is not analogous to a species, so it seems inaccurate to compare the author to an intelligent designer (or, as my students are fond of malapropizing, an "intellectual designer"). Instead, shouldn't we be thinking about the genesis of a literary work as analogous to reproduction? (Click the link for my photoshopped pregnant Spenser.) The question, as far as the importance of authorial intent, might not be ID vs. evolution, but sexual vs. asexual reproduction: Is the work an offshoot of the author, or a mutual creation by author and reader? (In regards to which, see my earlier post on "Skin.")

Now, granted, that's not strictly compatible with my other ideas, but I do get the feeling that the scope of that question is wrong. To firm up my thoughts a bit more, I think I'll need a good infusion of biology, as I said before... I don't mind, but I'll have to dredge up the time.