Cholly on Software:

"We are a million-year-old beast," he said. "Interfaces will have to mimic how the real world works." One of the ways to do that is to use objects that represent real-world things, like radio dials and on-screen controls for a software application, he said.

In production from MSNBC and Disney: Turner in the Rye, or Gone with the Windows. Reporting for duty are Martin Mull as Ted, Ernest Borgnine as Rupert, and lovable "Jane-Bo" courtesy of Pixar Animation Studios.

Last Saturday, a friend told me about Microsoft interviewing techniques. Least annoying sample question: "If you're using base negative two, how do you distinguish between odd and even integers?" Well, there's nothing like an arrogant geek for making really user-hostile interfaces, which explains the state of most Microsoft products. But now we have a different mystery: Why is that all Microsoft products are not equally crummy? (IE5, for example, is the best Web browser since Mosaic.)

The answer is temps. It seems that Microsoft doesn't subject temps to such "high hiring standards," thus opening a slim possibility for the accomplishment of useful work. To avoid rocking the arrogant geek boat, these temps are not judged by their performance ("We've set a hard rule: 364 days and these people are out. I don't care if they are rebuilding Windows 2000 by themselves, they are not going to work in this company."), and so Microsoft can be assured that they won't drain much needed funds away from, say, three of the world's five richest people.

Headlines for a New Society: Hackers due to release new computer bug to allow takeover of Microsoft computers

Sounds like the Microsoft press releases are getting more straightforward....

After a week of gout, Cholly's in the worst of moods -- which makes it the best of times for punditry.

• It's nice to see so many sites returning to the pre-David-Siegal pre-one-screen-per-page pre-Acrobat days of content-and-reader-focused UI design. It's not because research has finally won out over buzzwords, any more than reasoned discourse passes laws and wins wars. No, the only way to beat buzzwords is with newer shinier buzzwords, like "liquid HTML."

  The "pre-"-prefixed villains above all shared the misconception that a Web designer can ease the reader's experience by cutting up a single lump of content into arbitrarily-sized and carefully-packed-away pieces. But the human eye and brain are designed to work together to find elements of interest in larger contexts: losing the current context during our scan simply slows us down. (Predictably, "accessibility" is one of the shiny buzzwords joining the fray, even though the only possible defense for bite-sized-Easter-egg hierarchies is that they might speed up browsing for the vision-impaired, who can't so easily scan through a long page.)

• Decades of evidence proves that general hypertext discussions can generate even more smoke than markup-specific discussions. Again, it's nice to see the occasional puncture. But again, the pertinent research is underreported: What readers-for-pleasure value is not passivity before a static presentation but instead their control over the reading experience. Readers' eyes dart to and fro over the pages rather than moving sequentially; readers pause, backtrack, flip forward, and remember.

  Hypertext evangelists attack linearity, but a work of literature is not a one-dimensional line: it's a two-dimensional object striving for the illusion of four dimensions. One of those dimensions is time, which we assume to be unidirectional, but, if anything, the classic literary text overturns that assumption by emphasizing the nonlinerarity of experience.

  Hypertext might be able to provide more thematically appropriate surface structures than traditionally printed text. (Or it might not. The hypertext I'm proudest of has gotten more enthusiastic responses from readers who've read the flattened-out printed version than from those who've encountered it only as hypertext.) Even so, the virtues of a particular hypertext can never be called virtues of hypertext in general, and making a piece hypertext doesn't confer on it any special positive quality. It's like saying that the wah-wah pedal created a musical revolution. Play it, Prunes!

How Programmers Stole the Web: Tog's favorite example of Web-programmer parochialism is JavaScript -- which has its share of problems, but which is, after all, for programming. My favorite example is the use of six-digit base-sixteen numbers to specify color. Have you ever tried to explain this to visual designers? (I have, god help me.) For that matter, have you ever seen a programmer's idea of visual design? (We all have, god help us.) Didn't any Netscape employee ever even use a graphics application, where 0-255 and 0-100% are the standard color-value scales?

To Err Is Human; To Give Up, Divine: Good riddance. I was a lot sorrier to see the Mosaic name go.

Ah, memories, in the coroner of my mind.... Throwing Internet development into the Dark Ages by deciding to dictate new "standards" by whim was antisocial enough. But then came the Doughboy's astonishing claim that his (or NCSA's, who's counting?) browser was going to become an honest-to-gosh operating system and wipe Microsoft out. Oh right, I spouted off at the time, a cross-platform operating system that's going to fit without any problems in application space on top of another operating system (which in turn was on top of a third operating system, in the case of MS Windows).... It's exactly the kind of plan you'd expect from an arrogant college goop dropped straight from campus to boardroom.

And it worked exactly as well as you'd expect, too. As a business move, it brought Microsoft down hard on the browser market. As an engineering move, it turned the Netscape application into a breakdown-prone resource-tapeworm. And as for Andreessen's grander ambitions -- surprise! -- Microsoft's engineers had an easier time integrating with an existing operating system than Netscape's engineers had creating and layering a new one.

It did, however, shift enormous piles of money to those responsible for the debacle, thus helping to reorient software engineers' definition of "success" to match their CEOs': the timely manipulation of market gullibility.

Too bad Gates & Co. decided to keep IE5 single-platform; thank goodness they were dumb enough to get into legal trouble by using high-pressure monopolistic tactics to push an application which would be quite capable of walking on its own. And my very best wishes go to Opera and iCab.

If you can't take the seltzer, get out of the ring: If I was a better researcher, I'd figure out how many Web pundits who have protested the shutdown of The Dysfunctional Family Circus have also protested the notion of Third Voice users exchanging comments over the pundits' websites. (Personally, I think the idea of magically lifted-off graffiti is pretty neat.)

And I'd also look into whether the parody-as-fair-use legal argument is at all muddied by the captions that refer to the strip's drug-addled child-raping no-talent-bum dad as "Bil Keane."

Recently received via anonymous donation: A free-in-every-box informational pamphlet headlined "How to use 100% certified organically grown cotton digital tampons".

I remember back when it was a big deal just to be stereophonic....

The Information Supergayway

Jack Glicken, Chief Of Police for Midget Village...

... who bears a striking resemblance to a later resident of the Gayway, Jack Spicer.

pundit.org

The first generation of Web designers came either from the print world or from the software design world. The nice thing is that the two worlds tend to make different mistakes which can be played off against each other. The nasty thing is that they agree on the importance of the Big Dumb Intro Page (or BDIP): a graphically intense page whose only useful element is a "Click Here!" button or a timer script that eventually takes the user where they wanted to go in the first place.

• Print types like the BDIP because it reminds them of the cover of a book or brochure and of the splash page that introduces a glossy magazine article.

  Leaving aside the rare case in which the cover is content (e.g., comic books), these are essentially navigational aids. In the print world, they provide access; in the Web world, they delay it.

  • The big-logoed cover of the Oakland White Pages lets me know what book I have to reach for and open to get to the text-heavy content of the Oakland White Pages. On the Web, we don't need a big-logoed cover since we've got the link itself: I know I'm opening the Oakland White Pages because I damned well selected a URL described as "Oakland White Pages".

  • In a glossy magazine a huge Illustrator-shaded graphic and big shaggy display fonts let me know that I'm at the beginning of a multi-page article. Whereas in a browser all I have to do is hit the "HOME" key or drag the scrollbar to the top.

• The one place in the software world that BDIP-like-objects are used navigationally is the trade show, where the static introductory slide or screen-saver-like graphic lets the potential customer know what they're in for if they stray too close. Let's hope that only the most extreme of marketers would mistake a user's computer for a company booth at a trade show and move on....

• Software types think that BDIPs are like splash screens A splash screen is (barely) tolerable in a large application which may take a while to load, since it reassures the user that their request is actually being worked on rather than having (immediately) crashed the system.

  In contrast, on the Web a BDIP prevents the system from working on the user's request: what the user wants can't begin to be loaded until after the BDIP has been removed. The proper Web equivalent to a splash screen would be careful design of the home page to optimize loading: no big unsized graphics, no page-long tables, no frames....

If You Eat Hamburgers You Don't Deserve to Buy My Vegetables

From the Tech Law Journal (via librarian.net):

The Librarian of Congress, James Billington, stated that the Library will not digitize books. "So far, the Internet seems to be largely amplifying the worst features of television's preoccupation with sex and violence, semi-literate chatter, shortened attention spans, and near-total subservience to commercial marketing," said Billington.

"You don't want to be one of those mindless futurists," said Billington, "who sit in front of a lonely screen."

"It is isolating. It is a lonely thing." In contrast, "libraries are places, a community thing."

Rest assured or uneasy that we here at the Hotsy Totsy Club will continue to point to digitized books and, when Ray buys a scanner and loses his job, may even be digitizing some of our own....

Even aside from the New York Times reporter's hopeless muddle over the word "virus," I have mixed feelings about this story. On the one hand, The Sims is proving to big business monkeys that the CD-plus-downloadable-behavior-changes software model I've been pushing (with no luck) for six years actually does work. On the other hand, I'm not programming The Sims.

"Like almost everything in the game, the guinea pig's function as a disease vector was carefully simulated, Mr. Wright said. For example, the guinea pig only spreads the disease if a Sims player neglects to clean its cage, and only if a player reaches into the cage to pet the software animal and is bitten will he get sick. Someone who has gotten sick sneezes and coughs and will infect other human characters in the game who come within several 'tiles' distance."

pundit.org

"Encapsulation, Inheritance and the Platypus effect" is a not-bad little keep-it-honest rant for programmers, but the author mistakingly blames object-oriented languages for some foundational trip-me-ups of software engineering. He almost says it himself: Object-oriented techniques aren't so much useful because they map the real world, as because they map how software engineers think. I've always done my top-level design in what was later called a "object-oriented" way. Using an object-oriented language just shortens the trip from design to implementation. The better the language, the shorter the trip, which is why I still miss ScriptX.

Having said that, it's always good to be reminded of our besetting sins as software engineers and humans. Premature overgeneralization, for example, which is just Our Gang's variation of humanity-at-large's besetting sin, rushing into categorization. We receive so many internal and external rewards for generalizing and categorizing that we tend to anticipate categories long before we've had enough experience to justify them, and then feel forced to defend our itty-bitty-witties against all enemies (i.e., facts). Thus, our anticipations lead us to ignore the evidence of the real world (and, admittedly, reap the often sizable rewards of ignorance) rather than preparing us for it.

Less philosophically and more engineeringly, the advice to "trust no one" is well-founded. I mean, don't waste time trusting no one, but try not to tie your fate too closely to code you can't examine. On the grossest scale, all inter-group (much less inter-company; viz. ScriptX) projects are doomed; on a smaller one, I'd refuse to use any framework I couldn't modify, particularly after my Microsoft AFC nightmare.

(Removing and adding display elements wasn't thread-safe. OK, that's the horrible unworkable one. But the so-horrible-it-was-funny one was when I did a performance analysis and found a bottleneck at the Microsoft-supplied sorting routine. How hard is it to do a sort in an object-oriented language? There's gotta be pointers, right? I guess it depends on which sub-sub of a sub-sub gets the job, but Microsoft handed it to someone who must not've heard of pointers, and Microsoft never had the code reviewed. Christ. One routine rewritten; three-hundred-to-one speed-up.

Not nearly so bad is the temporary-object-and-synch-locking overhead of Sun's StringTokenizer. But BAD ENOUGH!)

Man. And Christ. Sometimes that English-Philosophy double major looks good.

Until I see what English and Philosophy professors are doing.

Gosh Darn the Pusher

One of the few things I don't like about MP3-mania is the way that the big software concerns have positioned it as a way to pirate CDs. I mean, what I love about MP3 is that it's a convenient cheap durable way to preserve and play audio that's otherwise available only on inconvenient or not-so-durable media: cassette-only recordings, old 45s and 78s, very out-of-print LPs.... But Real and Microsoft and MP3.com concentrate solely on making it easy for moving CD tracks to the computer, and since CDs are already pretty convenient and durable and are less likely to be out of print, it's hard not to see that as encitement to piracy.

In fact, one of the first free pieces of software that did convert non-CD-audio to MP3 files, BladeEnc, was hassled right out of binary distribution by big business monkeys. Conspiracy theory, anyone?

For alternatives, go to The Sonic Spot and Transferring LPs to CDR. My current toolkit: Wave Repair for recording and LAME for compressing.

David Auerbach economically adds to our belly-down crawl around the thundering canon:

For myself, it's wonderful to discover that, thanks to Salon Premium, I am now refusing to support Camille Paglia! (Admittedly, this is one of those "speaking prose all my life" thrills, but a thrill's a thrill.) On top of which, I can also actively not shore up the tottering incomes of Salon's CEO, editors, movie reviewers, and so on. Talk about win-win!

My attitude towards subscription might be a little old-fashioned, though, since I don't subscribe to newspapers but I was glad to throw some money toward the NosePilot kid and I'd be glad to pay Lynda Barry directly for her watercolors. What galls are the extravagently wasteful layers of plastic and cardboard pimping that wrap the product. The web doesn't need prejudging editors so much as postfacto pointers, and the web doesn't need high-salaried executives or designers at all. What the web (still) needs is a reliable way to handle genuinely micro micropayments and a reliable way to protect creators from being bankrupted by unexpectedly popular creations.

The dreadful commingling of the overpriced software industry and overpriced entertainment industry loaded huge amounts of unnecessary cost onto a model kept afloat till now mostly by inflated valuation and partly by advertising. Advertising alone can't come close to maintaining it. Good riddance in the long run, but in the short run, the tumbling mountains of garbage are, as is their wont, sweeping lots of great stuff away.

Services I would gladly pay to keep alive simply vanish without being given a chance, their wanna-be-like-the-big-guy owners using the same reasoning by which corporations are supressing the history of cinema: the copyright holders, not caring about what they hold copyright on, consider the cost-to-benefit ratio of giving permission too high to deal with. (You have to hire someone to give permission; you don't need to hire anyone at all to ignore requests. Suing for infringement, of course, is always worth the money.)

Instead of archiving and cataloging their own work, writers depend on crash-by-night magazines to collect and maintain material, when the realities of both print and online publication is that magazines work, at best, as initial publicity. Comics artists waste time on brain-dead Flash loops when they could be making full-color serials. Newspapers, rather than storing their articles as highly compressible dirt-simple text and collecting the small fees that would be otherwise fed to library photocopying machines, are closing access, increasing "reprint" costs to unrealistic levels, and publishing more material on expensive paper than on cheap diskspace.

After its jerrybuilt business district finishes collapsing, the web may find itself only set back five years or so. But even in the midst of the swirling dust, there are encouraging signs. Weblogging seems to have already spread to zine scene levels, without paper zines' constraints on further growth. And there are finally signs that some academics are ready to push against the utterly unnecessary waste of traditional journal publication....

Dan's Data:

... plenty of times, when I've been discussing some goofy, badly flawed Web site business model with its merry exponents (at various different dot-coms, not just News), they've said to me 'So what do you want us to do? Just give up?!'

Well, yeah.

Department of Homeland Security

Near the end of the workday, network performance seemed to stall. Moving through the active processes, I brought my FTP window to the front and saw it emit the message "Connection privacy lost" just as there was a burst of modem activity. Holy shit, I didn't even know it was possible to break into an open FTP session, much less for the FTP client to recognize the intrusion! It's amazing what computers can do nowadays.

In a fatigued flurry I stopped the process, rebooted the modem, ran virus checkers, inspected my system files, got my backups together, and fruitlessly searched the web for more information about losing FTP connection privacy...

... and after about fifteen minutes I remembered that I was working on a website for a privacy advocate and that its FTP bookmark was named "privacy."

"I think it’s guaranteed that for the rest of my life that she’ll never ask me about myself or my work."

When prying at the purported origins of art in neurosis, it's good to remember that we've implicitly defined this here "art" of ours more by its publication than by its mere production. Setting dynastic brats aside (hopefully to be crushed by speeding trucks or eaten by bears), perhaps one reason so many twentieth century artists come from unsupportive families is that they're less worried about familial reactions to their art, either because their family is uninterested in anything they do or because their family is unlikely to venture near its published context or because who cares what those losers think? (And similar for low-to-no-status day jobs with uninterested employers.)

We can easily test this hypothesis. A pink portrait of bestiality is safe from Ma & Pa Philistine on the local café's walls; a "My Sister Was Raped & All I Got Was This Crummy Poem" can be comfortably tucked into a little-teensy magazine, or a hilariously flip sexual boast into a fanzine interview -- but as soon as your boss or cousin grows unaccountably tired of watching the Game Show Channel and decides to try your name in a search engine, the jig is jug-jugged tereu.

Thus, assuming that the web stays the course, we might predict:

1. Even more extreme levels of dysfunctionality among those who dare to publish under their own names, unless those names are index-swampers like "Jack Smith"
2. Much second-thought yanking of web-relocated material (I myself have been asked to remove a hiliariously flip sexual boast or two)
3. As in past centuries of self-publishing, greater reliance on anonymity or pseudonymity

Obviously not, so jot a winner beside predicted result no. 1!

Existenchal

Eclogues spends a lot of time being right. As for example the April 9th, 2002, entry which firmly and properly positions anti-Oprah-Book-Club tightlipped sniffery into a long tradition of attacks on "women's books," to which I can only add a) in English, the tradition actually extends at least to the late 1600s, when women first became professional writers, and b) well, I actually don't have anything else to add, 'cause when I tried I realized that Eclogues had already said it.

Being at present irritably unemployed with an Internet connection, I also followed Eclogues' rustyc pypes [February 22nd, 2002 entry] to an irritating Internet time-waster that purports to disclose conflicts in one's philosophical assumptions, and which, true to my smug preconception, made a fool of itself by telling me that there was a conflict between my passing of moral judgment and my admission that morality is culturally (as well as, to a lesser extent and although they didn't mention it, individually) relative. They might as well find a contradiction between "All men are mortal" and "Some guy will outlive me." Of course I pass moral judgments; I, as an individual, am part of a culture (or two or three), and that's one of the things that us individuals in cultures do to pass our irritable time. That doesn't mean I'm blind to the existence of differing views. It just means I want mine to win.

Even more irritating, given the popup ads on every page, was the Ethical Philosophy Selector which after a tedious questionnaire selected "Sartre" as my ideal thinker. Which is merely to say that the questionnaire showed me to be no philosopher. It's like testing my singing voice and then telling me that my favorite musician must be Mark E. Smith. Philosophy isn't a genre of belief, but of discussion.

That's what's always been wrong with all this automated matching crap, and why search engines, library browsing, and meeting folks at parties still rule supreme:

What's required in a friend isn't resemblance but recognition.

What's desired isn't identity and simultaneity but knowledge and occasional amazement and dependable support.

And as Blake and Newton agreed (hands across the damnation!), there's no support without resistance.

Anchorites

"again pondering time and date stamping. It seems such triviality but has a strong presence in the weblog." - nqpaofu

It's understood well enough nowadays that web-browser-targeted writing and reading tends to steer us in a different direction than book-targeted, pulp-magazine-targeted, or glossy-journalism-targeted writing and reading. But the legacy voculabulary used to jerryrig this jalopy tends to veer us into the ditch.

If a website is a book, then a page would be... a page? Thus early punditry decreed a web page should be at most "one screen" long. As we also understand well enough these days, this was awful advice. The paper page and the HTML page only coincide for the publisher of short poems.

Weblogging fulfilled the promise of web self-publishing because it was the first form whose conventions directly supported the natural tendencies of web-browser-targeted writing and reading. Here, the web page is more of a "chapter" or a "topical discussion" than anything a prose writer might call a printed page.

Which makes the web page unsatisfactory both as bog paper and as a way to point to a more-or-less arbitrary manageably constrained span of source material in its original context.

Permalinks were accordingly welcomed as a natural replacement for the convention of page numbers. And I can't deny that we publish serially, or that references to serial publications generally include the date, or that a single-author serial publication generally won't release an unwieldy span of source material on a single day.

But, as Jouke indicates, there's something creepy about it. Dates carry connotations outside the realm of publishing, whereas I'm unlikely to feel a pang of remorse or nostalgia when someone mentions page 1062.