| |
July
27, 2000
Off to Austin. I'll be in the Live Music Capital of the World
this weekend.
In the
Middle of Web Design. Jakob's The
End of Web Design has fomented a debate/discussion/flamewar
on websites and message boards throughout the land. And what no
one seems to understand is that Jakob isn't *happy* about what he
has to say. Jakob isn't dancing in the streets over these things.
I've had the fortune to chat with him, and I know he's as frustrated
(if not more frustrated) about the state of design on the web as
anyone else.
The simple
thing that Jakob's article points out is how the world *is*. It's
currently a networked mess of hundreds of sites, with people adopting
conventions from general web use, conventions they'll apply to your
site whether you want them to or not. Sippey's article, while smart,
is unfortunately afflicted with a "how the world *should be*"
sentiment. And maybe how it might be a few years down the road.
But I've
got to design for the *right now*, and I can tell you, as we're
in the early stage of redesigning Epinions, Jakob's notions smack
a whole lot truer than Sip's. We have users who tell us, "I
always ignore the left-hand side of the screen." We have users
who want to apply shopping cart metaphors to things that aren't
shopping carts, simply because the terminology is kinda similar.
Also, we have to integrate our offerings with those of other Web
services, and we need to make sure we have a malleable (and possibly
generic enough) design that allows for a consistency of experience
as users move from our site to others.
The other
thing that seemingly everyone misses is "what remains in web
design", which includes task analysis, content design, and
information architecture. Which, to my mind *is* web design. That's
pretty much all I do. And it suggests to me that Jakob doesn't really
believe in the title of his own essay.
And what
really pisses me off are the attacks on Jakob's person. Disagree
with his ideas, fine. (Lord knows I have.)
Hell, I think Jakob's quite off-base with some of his contentions
in this latest piece, particularly around wireless. But this passel
of self-important Designers, incapable of coping with rhetoric,
and whining about Jakob's proclamations, and resorting to ad hominem
attacks because they have no actual reasoned response, well, SHUT
THE FUCK UP. Why bother going on about it? Why publicize it? Just
go design something, and leave the discussion to folks with something
to say.
July
24, 2000
Komplexity for Kids! Steven
points us to StarLogo,
a programming environment for modelling decentralized systems, based
on Logo. Looks like fun!
Everyone
else will link to it, so why shouldn't I? I'm a big fan of Jakob's
latest alertbox, The
End of Web Design. I particularly like it because A) many of
his notions are dead on, b) he knowingly betrays his own nihilist
title in the last part of the essay, and c) it's sending "web
designers" into apoplexy, always a good thing. For a well-reasoned
rebuttal, read Sippey's The
Beginning of Web Design. Though, I think Sip agrees with Jakob
more than he realizes.
July
20, 2000
Symbio-terrific! More poking around on the self-organization
front led to The Symbiotic
Intelligence Project: Self-Organizing Knowledge on Distributed Networks
Driven by Human Interaction, which I'll read, really, when I
have the time, honest.
July
19, 2000
An explanation of my apartment. When I have people visit my
unkempt apartment, I shouldn't say, "Sorry it's such a mess."
Instead, I should explain that it's a highly evolved self-organizing
system.
So, I've
been doing some research on self-organizing
systems. This has been spurred by some discussions at Epinions--our
CEO spoke at the Industry Standard's Internet Summit conference
on how our service is a self-organizing system--and why it means
it's not just a product buying guide.
My primary
exposure to self-organizing systems was through the WEBSOM
(Web Self-Organizing Map, I don't know why WEB is in all caps),
a data visualization technique for massive information spaces. The
basics are straightforward--a semantic processor analyzes all the
discrete bits of information and develops a rough 'understanding'
of them, and the algorithms of the self-organizing map group the
information bits into clusters of related relevancy. An interesting
application of this process is the PicSOM,
a way of navigating and finding pictures by selecting pictures similar
to what you're seeking. (The interface for the PicSOM sucks, you'll
probably have to read the user guide, but it's worth poking around
if this stuff interests you.)
The utility
of such a system is clear--human editors can't handle the complexity
of All Information; if we can get computers to intelligently categorize,
we can more easily find items of interest in the vast sea of data.
This
extremely accessible explanation of self-organizing systems
stresses the importance of feedback in the process. Though written
by a self-help guru type, seemingly to promote his services, it's
thesis that iterated trial-and-error is superior to coaching gets
at the crux of the matter--these systems work from a bottom-up,
evolutionary approach reliant on mutations and feedback that gradually
lead to improved results.
This is
one of those yet-another examples of how bottom-up approaches perform
better to top-down approaches. And reminds me of the work of Rodney
Brooks, as featured in Fast,
Cheap, and Out of Control.
Eat red
rocks! I'm an idiot because I keep not reading more of Phil
Agre's writing.
July
17, 2000
In other news. I'm hosting Danes (who *love* to shop), and have
piles of work, so blogging will likely be less than normal. Happily,
my readers are looking out for me. Evan
points us to "Bushwhacking
Through the Computer Ecosystem," an article on biological
models of engineering. Poking around, I found the SFI
Bulletin is likely to be of interest to y'all.
Afflicted
with Small Penis Syndrome. And my dad points us to "Was
Freud a Minivan or S.U.V. Kind of Guy?" a New York Times
article (registration required, I guess) on the difference between
otherwise demographically-identical consumers. Unsurprisingly, it
turns out that SUV types tend to be anti-social small-minded selfish
unhappy people with penis-size issues.
July
15, 2000
AA is not just a support group. Read my "Deconstructing..."
piece on AA.com. It's quite heavily edited, mostly for space
(remember, Internet World is a print pub!). Here's how I
had submitted it:
The user goal AA.com has to support is a fairly straightforward
one-buying an airline ticket to a desired destination. And the motivated
user can successfully achieve that end with the current system.
However, there are significant details worth exploring.
--Registration required. I bet you can purchase
tickets over the phone without having to sign up for frequent flier
miles, yet you have to register with the AAdvantage program to make
a reservation online. This forced registration detracts from the
goal of buying an airline ticket, steering users down a form-filled
chute before letting them return to the task at hand.
--Broadvision silliness. AA.com suffers a number
of usability issues typical of sites built with Broadvision
- Typing "aa.com" takes you to a page with a two-paragraph message,
but you're redirected before you can finish reading it. Clicking
'back' to see if you missed something important throws you into
a redirect loop
- Being inactive for more than 20 minutes (an plausible circumstance
given that people use this site at work, where distractions are
many) causes a "timeout," as a "security precaution" even if you
haven't done anything to compromise your security (god forbid
another coworker see your travel itinerary!). This timeout loses
all progress made, forcing you to start from scratch.
- Once after trying to use the site after a timeout, I a web
page with this error message: "NSAPI Error: Bad browser input."
No clue as to what caused this.
--Interaction hits and misses.
Hits
- The calendar interface for selecting dates is far superior to
the standard pull-down method (though the calendar has no obvious
affordances for clicking).
- Full city names are used, not arcane airline codes. (New Orleans,
LA means something. MSY doesn't.)
- It was also easy to toggle a search between American Airlines
flights and all flights.
Misses
- On the first ticket finding interface, there was no way to select
time of flight, returning awkward departure times. Clicking around
more revealed a nearly identical interface, except with the ability
to select time ranges. Why wasn't that offered in the first place?
- In the reservation process, to move forward after selecting
your flight (but before entering personal details) you're meant
to click "Purchase." In shopping transaction process, 'purchase'
is a strong word, meaning, "When you click this, you'll be giving
us your money." Since that's not what it means, users could be
discouraged to continue, thinking they'll have to pay before they're
ready. Seemingly small details like this can cause big problems.
The other
Deconstructor was much friendlier. For me, the thing was that while
AA has a pretty good interface,
I regularly use United Airlines'
site, which offers a far superior experience.
July
14, 2000
You amaze me. I just want to say that I'm proud to have readers
as fascinated with "poontang" as I am. The word "poontang"
that is. Yes. Ahem.
What's
it all about. We all know that Automatic
Media, the hip new content network formed by merging Feed
and Suck, is just ChickClick
for grad students, right?
July
11, 2000
Even more on poontang. Dan
Lyke chimes in with a pointer to this
page, which suggests, among other things, that poontang
is a contraction of "Pudding Tang," which would accord
with my dad's theory that it's origin resides with a baked good.
A search for "pudding tang" on Google just turns up a
lot of pointers to the Jell-O product and the drink of astronauts.
July
10, 2000
No! Don't listen to him.
*I* am Jason Kottke.
July
9, 2000
What to do when you're in a mood. See a Buster Keaton film.
(I just saw Steamboat
Bill, Jr., with live accompaniment by the Alloy
Orchestra. The show was prefaced by a slide presentation from
the author of Silent Echoes,
a delightful book that uses Keaton's penchant for filming on location
to provide an historical glimpse at life in the 20s.) Drink a chocolate
malt. Finally get around to reading McSweeney's
#4.
Blah
reading. So, I've recently read through a number of disappointing
works of non-fiction. In order:
Uncommon
Grounds: The History of Coffee and How It Transformed Our World,
by Mark Pendergrast. I bought this on the strength of how much I
loved his previous book, For
God, Country, and Coca-Cola, and how much I love a good
cup of coffee. Unfortunately, in taking the exhaustively detailed
approach that worked wonders in For God and trying to apply
them to the history of coffee, Pendergrast produces a ponderous
tome, filled with unnecessary details that require rubber pants
to wade through. And even so, because the history of coffee is so...
big, the book has no focus, and these details simply sit there.
A For God-like book on Peet's or Starbucks would have likely
provided much more manageable book fodder given Pendergrast's predilection
for minutiae.
Why
We Buy: The Science of Shopping, by Paco Underhill. A breezy
treatise on what people do in stores, from an anthropologist whose
devoted his life to studying commerce. This work proved to simply
be a number of facile observations of how retail environments should
be set up, filled with interesting statistical trivia and revealing
stories, but not adding up to any real understanding of the consumer
mindset. Most shockingly, Underhill doesn't even bother to address
that which ethnography can best uncover in this whole process--a
person's motivation to shop. Underhill's work begins from the moment
someone walks into the store, seemingly ignoring that person's life
outside. The subject of another book perhaps, but, in my mind a
subject far more interesting and thought-provoking than a catalog
of admonitions for silly shopkeepers. Underhill does anthropologists,
and his readers, a disservice with his aggressively popular approach.
The
Making of 2001: A Space Odyssey, compiled by Stephanie Schwam.
As part of the Modern Library's new "The
Movies" series, I looked forward to this book as a thoughtful
reflection on the development of what proved to be a watershed in
cinema.While the first half delivers pretty well on the book's title(most
entertaining are Arthur C. Clarke's diary entries and other recollections),
the second half is little more than a bunch of movie reviews, many
of which say the same thing, and none of which have anything to
do with film craft. It also left me wanting for a lot more about
Stanley Kubrick. For those interested in how movies get made, I
strongly suggest Leonard Leff's Hitchcock
and Selznick, detailing their fascinating history together,
and Round Up The Usual Suspects: The Making of Casablanca
(which is out of print, but many copies can be found through ABE.)
Referer-surfing.
In my logs, I noticed a link from http://werbach.com/news.html.
That name rang a bell, so I traced it upstream to the homepage,
and a bit of poking around led me to what triggered the memory--when
I was a web developer, I swore by Kevin Werbach's Bare
Bones Guide to HTML. Well, it seems Kevin has done remarkably
right by himself, writing and editing the influential technology
newsletter Release
1.0. I'm also drawn to Whirrled,
a work-in-progress (abandoned?) on life in the 21st century, cowritten
with his brother Adam. Yes, Adam
Werbach, who was the youngest president ever of the Sierra Club.
Active family, that.
Pissy.
I'm in a mood.
And I don't like it. Don't know what to do about it, though.
July
7, 2000
From other blogs. Geegaw
points to this page of top
ten movies lists of filmmakers and critics. Some great lists
in there. And some interesting trends. The obvious ones are there
(Citizen Kane, Rules of the Game), but I was delighted
to see some personal faves mentioned frequently (High
and Low, one of Kurosawa's lesser-known flicks, Au
Hasard Balthazar, Bresson's donkey flick). Clicking around
Combustible
Celluloid, you'll find a number of good resources for the avid
filmgoer. It also turns out that the site's proprietor is one of
Epinions' top
movie critics. Brittney,
in her quest for further film
geekdom, would be well-advised to poke around there a bit.
Boing
Boing links to a reprint of John Marr's classic "Death
at Disneyland" essay from his zine Murder
Can Be Fun.
And Maura,
in email, alerted me to the recent publication of Zot!
Online, Scott McCloud's revival of his comic book character.
July
3, 2000
More on poontang. Steve pointed me to these definitions
of poontang. (Though, to be fair, those definitions had been
mailed to me earlier by Laura Kehoe, reader of Jessamyn,
and librarian-type person.) I had always understood the term to
mean, simply, a vagina, used interchangeably with 'snatch' or 'pussy.'
I suspect the OED is barking up the wrong tree with it's French
etymology (putain, meaning prostitute), because, well, I've
never heard it used an anyway to suggest it's about whores. (And
the OED says "prob" suggesting they know they're simply
taking a stab in the dark.) And if it were French, then it's American
origins would be Louisiana or the French West Indies (which one
dictionary suggests). I'd love to take a few months off and travel
the south, researching the origin of "poontang." Um. That
sounds odd, doesn't it.
It's
Hell-errific! Though I go on and on about them, I'm not much
of a comics reader. They're by-and-large insipid dreck, and I find
even those with some substance are often dull. But thanks to the
Rak Daddy, I've found myself
on a Hellboy kick. Like, spending
hard-earned money on over-priced collections. Hellboy is definitely
a descendent of The Watchmen, a marvelous melange of super-hero-ish
fantasy, Victoriana-horror, and densely packed story-telling and
artwork that rewards very close reading and re-reading. The graphic
nature of the artwork is exquisite. I suggest that you, unlike me,
read the comics
in order--the story builds over time, and much of my confusion
could have been allayed.
|
|