June 01 - June 09, 2001
May 01 - May 31, 2001
April 01 - April 30, 2001
March 01 - March 31, 2001
February 01 - February 28, 2001
January 01 - January 31, 2001
December 01 - December 31, 2000
November 01 - November 30, 2000
October 01 - October 31, 2000
September 01 - September 30, 2000
August 01 - August 30, 2000
July 01 - July 27, 2000
June 01 - June 30, 2000
May 24 - May 31, 2000
May 1 - May 23, 2000
April 1 - April 30, 2000
March 1 - March 31, 2000
February 1 - February 29, 2000
January 1 - January 31, 2000
December 1 - December 31, 1999
November 1 - November 30, 1999
October 16 - October 31, 1999
October 1 - October 15, 1999
September 8 - September 30, 1999
August 29 - September 7, 1999
August 13 - August 27, 1999
August 6 - August 12, 1999
July 25 - August 5, 1999
July 17 - July 24, 1999
July 11 - July 16, 1999
July 01 - July 10 1999
June 09 - June 30 1999
June 01 - June 08 1999
All of 1998
Among the best papers at CHI2001 were the ones that folks
from Xerox PARC presented on information scent. And the best
of those papers was "Information
Scent as a Driver of Web Behavior Graphs: Results of a Protocol
Analysis Method for Web Usability" [PDF]. In it's
The development of predictive scientific and engineering
models of users cognition and interaction with the
World Wide Web (WWW) poses some tough and interesting problems.
Cognitive engineering models, such as GOMS, fit user interaction
with application software (e.g., word processors) when error
rates are small, tasks are well-structured, exploration
is virtually nonexistent, and
content is not a major determinant of behavior. Typical
interactions with the WWW, on the other hand, are very likely
to involve many impasses, ill-structured goals and tasks,
navigation and exploration, and substantial influences from
the content that is encountered. In this paper we present
an approach to the analysis of protocols of WWW use that
is aimed at capturing these phenomena and that is aimed
towards the development of predictive models that are of
use in science and design.
The purpose of this paper is to introduce a replicable
WWW protocol analysis methodology illustrated by application
to data collected in the laboratory. To support replicability
and reuse, we are developing a bank of ecologically valid
WWW tasks and a WWW Protocol Coding Guide, which will be
available on the WWW...
So, all this discussion of multi-tasking and attention
focus have set me on a couple of paths, one a thoughtwander
and one a research flurry.
Humans can attend to only one thing at a time. This doesn't
mean that they can't receive multiple channels of stimulus--we
simultaneously see, smell, touch, hear. Hell, we can even
receive multiple types of information from our eyes--that
which is the focus of our vision, and the periphery. And we
make decisions of action based on these multiple inputs. But
we're not truly multi-tasking. We can only pay attention to
a single thing... And if something in my periphery triggers
my brain, I then attend to that, and lose focus on whatever
I was dealing with.
So, we can have
displays with multiple readouts and successfully attend to
them, because, by and large, the information on them doesn't
change. And when something aberrant happens, we are drawn
to it because it DOES change. The problem would be if two
aberrances (is that a word?) occur simultaneously, because
we wouldn't be able to deal with both.
Now, while we can't
pay attention to more than one thing at a time, we can crudely
multi-task by switching our attention around a number of different
things at once. When cooking, you've usually got the a couple
pots on the stove and something in the oven and your chopping
something on the board, and you're bopping around, dealing
with all of this. Or, on the computer, you're reading email
while surfing the Web while chatting in AIM.
So, for all reasonable
intents and purposes, you ARE multi-tasking, because most
tasks don't require continual focused attention. To somebody
observing you from the outside, you seem to be perfectly handling
a number of parallel tasks with little trouble (of course,
until one of those tasks requires inordinate attention, like
an email needing immediate thoughtful response, or someone
in AIM telling you they broke up with their boyfriend, etc...
at which point you forget about all the other tasks.)
Kicking the search query "multitask
cognition" into Google lead to some delightful results
(and a bunch of less interesting ones).
whose comment started this whole thread over on Ev's site,
wrote about this over three years ago in JOHO, in a piece
Price of Multi-Tasking: Your Soul."
I then found the
of Harold Pashler, a professor at UCSD (an institution
that's lead in cognitive science research), and he has a page
for downloading, including a soon-to-be-published article
Switching and Multi-Task Performance" (PDF) which
discusses actual scientific 'n shit research in this field,
and is too drily academic, and has a big chunk (starting on
page 13) on Dual Task Performance, which starts:
We turn now to the limitations that arise when people attempt
to perform two different tasks at the same time. While there
is a large literature on relatively complex and continuous
dual-task performance, the focus here will be on discrete
tasks. The reason for this is that with more continuous
tasks interference and switching are easily disguised for
reasons that will emerge clearly below. Not surprisingly,
limitations on simultaneous mental operations
evidently arise at various different functional loci. Perceptual
analysis of multiple stimuli can often take place in parallel,
but when perceptual demands exceed a certain threshold,
capacity limitations can become evident (Pashler, 1997)
although non-perceptual factors (such as statistical noise
in search designs) often masquerade as capacity limitations
(Palmer, 1995). These limitations appear largely, but probably
not entirely, modality-specific (Treisman & Davies,
1973; Duncan, Mertens & Ward,1997). Similarly, response
conflicts arise when responses must be produced close together
in time. These perceptual limitations are often most acute
when similar or linked effectors are used, such as the two
hands (Heuer, 1985).
The most intriguing, and for the present topic, the most
relevant limitations are those that arise in central stages
of decision, memory retrieval and response selection. Intuitively,
most laymen assume that the cognitive aspects of two tasks
can be performed simultaneously unless one or both is intellectually
demanding. This seems not to be the case, however.
This is most clearly seen when people try to carry out two
speeded but relatively simple tasks, each requiring a response
to a separate individual stimulus. As Telford (1931) first
observed, people almost invariably respond to the second
stimulus more slowly when the interval between the two stimuli
This guy's got
tons more like that, but... I... just... don't.... have....
Somewhat of a tangent,
but related, are two Malcolm Gladwell stories from The
Art of Failure: Why Some People Choke and Others Panic.""The
wonders, "Can humans multi-task?" The simple answer
is, "No." The more complex answer is this:
Humans can *pay
attention* to only one thing at a time. Folks who claim to
"multi-task" are usually just rapidly switching
their focus of attention among a group of things. This is
what the Air Warfare Coordinator that Ev quotes is doing.
Now, what about
walking and chewing gum? Or being able to carry on a conversation
while driving a car? Well, humans habituate actions, such
that they don't require attention. In fact, humans can't NOT
habituate. Among other things, this leads to mis-use of tools
(be they physical or "virtual") when interfaces
To learn more about
this, read Jef Raskin's
More to read:
"Humans have told stories since the cave, and there is
a resurgence of interest in the art among todays business
leaders. What is new is the purposeful use of narrative to
achieve a practical outcome. In this seminar, four leading
thinkers on knowledge management explain why storytelling
will become a key ingredient in managing communications, education,
training and innovation in the 21st Century."
George Brett's notes
on the symposium.
Information Retrieval on the Web: a new direction
Social networks, auto-taxonomies, web-as-brain
Must Be A Better Way...
David Gelertner, new UI metaphors, coping with information
Leader of the Packie.
I received many refutations of the packie -->
Paki claim, some calling me racist. I'm aware that "Paki"
is a derogatory term for a Pakistani, but, well, if it were
the appropriate etymology, I wouldn't let PC get in the way
of discussion. As it turns out, enough people convincingly presented
the "package store" lineage that it's most likely
the root. Though, isn't it interesting (if perhaps a bit upsetting)
that some think it comes from "Paki"?
of the Global Brain
GlobalBrain-L FAQs (benefits
A Curriculum for Cybernetics and Systems Theory
Life in the Wireless Age" is James Gleick's take
on the present-and-future state of wireless networking and
its implications on our lives. Runs the gamut from GPS in
cars to personal digital assistants to Bluetooth-vs-802.11b
and more. Most compelling to me is the discussion of social
[Bernardo Huberman's] research consistently finds informal
communities making better decisions than any of their membersknowing
more and thinking better than experts. We now know
that society can work better than any individual,
he says. There is this notion of a collective mind,
a social mind, and today the Internet allows us to tap that.
We are distributing intelligence. We are creating social
organisms that carry out continuous computation.
This is all reminiscent
Brain ideas, such as the Principia
Cybernetica Web. The PC folks have a GlobalBrain mailing
list requires group acceptance to join--however, the archives
are posted on the Web for all to see (and, not surprisingly,
contain a pointer to the Gleick article!)
Some research into
Huberman turned up these links:
PARC Internet Ecologies (where he used to work)
paper on cooperative problem solving
overview page on cooperative problem solving
Huberman doesn't seem to have a page himself--links to one
on Xerox PARC turn up 404, and I can't find anything about
"Sand Hill Labs" at HP.com.
News from all
over. This post is all about email I've received on the
various topics discussed over the last week or so.
I love that my readers dig regionalisms as much as I do. It's
spurred more email than any topic I've written about in months.
Two folks wrote in to say that liquor/convenience/corner stores
are called "Offies" in the UK -- short for "off-license",
meaning licensed to take alcohol off the premises. In Ohio
and Connecticut they're called "carry-outs." I received
word that Masschusetts stores might be called "Packies"
not because it's short for "package store," but
because it's short for "Pakistani," the supposed
ethnicity of many proprietors of such stores. And Michael
Boyle wrote in with this:
Just reading your little entry about regionalisms on
peterme.com and I thought you might be interested to hear
that in Quebec everyone, English and French, uses the Quebecois
(it's not used in France) term for corner store (where you
can buy beer and wine, but not liquor) "depanneur"
which we shorten to "dep". So if I'm going over
to a friend's house on a Friday night I'll ask, "hey,
should I stop by the dep on my way?"
The really funny thing about this is that the word in
French (again, it's mostly used in Quebec) is odd even in
its native tongue - it's derived from the phrase "en
panne" which means "broken down". If you're
driving in France and you say, "my car is en panne"
it means you're on the side of the road with the hood up
waiting for help. So, literally, a depanneur is the place
where you de-panne - or fix yourself up.
Which makes sense considering that's where you buy your
Not as interesting, but for completeness' sake, we buy
liquor (and better quality wine) at the SAQ, which is just
the acronym for Societe des Alcools de Quebec - or the liquor
commission. It's a government-run retail outlet. It's pronounced,
in English, "sack".
"What We Do"...
added to the discussion of "What We Do", my meandering
post from April 15:
I think the other side to your argument is that the
Web *attracted* a lot of people who were generalists, 'connectors'
etc., in the first instance, but the commercial structures
of practice that were built have not encouraged the growth
of more people like that - a lot of kids who have been laid
off don't know a hell of a lot more than how to use flash5.
Matt also posted
his thought-provoking presentation called Survivability
in Networks, that addresses parallels in network architecture
and design team formation. Matt is big on generalists. I prefer
to think of us as "professional dilettantes."
Rothstein also contributed to the "What We Do"
I read your blog-blurb on "what we do" with
great interest. I've actually been thinking about this topic
a lot lately, especially with regard to what companies like
the one I work for can de described as doing.
It's a mess. Sometimes I wish that English had the capacity
of Spanish or French to simply add a suffix onto a noun
and have a business description. Perhaps next to the lavanderia
or patiserie we could simply exist as a weberia, or an interneterie.
Obviously, that's a little silly, but it begs the question:
Why hasn't a single word yet evolved to describe what web
design/programming/implementation shops do?
Unsurprisingly, I blame the marketers. And it relates
very closely to the "turf war" you describe in
your piece. It wasn't useful to distinguish such companies
by name (i.e. Harvey's) or location (North Avenue), or,
for lamentable reasons, reputation. So a series of increasingly
convoluted words were implemented to describe what should
be simple processes in a complicated way.
Imagine, if you will, that the word "bakery"
did not exist in English. I might discuss with my friend
where I like to get my Bread. Perhaps I favor D'Amatos,
because they are a Risen Dough Solutions Provider, whereas
my friend thinks that the offerings at Red Rooster are better,
because they are a Flour, Yeast, and Water Integrator.
Again, a little silly, but I think it illustrates the
fundamental absurdity the way we, as a profession, attempt
to describe ourselves.
I wish I had an insightful conclusion to offer, or the
great idea for the one word that would allow us to get around
all of this nonsense, but I do not. I fear we will all be
working under this linguistic obfuscation for some time.
finally, Group Forming Networks...
wrote in about Group Forming Networks, discussed April 12:
I've been marvelling at the power of this phenomenon
also --- something reproduced in my weblog experience in
a way that I haven't seen before --- weblogs are a much
more effective way to form these networks than bulletin
boards or conferencing systems were in the past --- I think
because of the architecture of weblogs, which involves linking
(forming new associations) but doesn't encourage flame wars
or as much noise (you only read the weblogs you find interesting,
and you kind of ignore the rest)...
This rule of 150 is interesting... I have also noticed
a "rule of 10" and a "rule of 30", however.
The rule of 30 has to do with the number of people I think
you can really keep track of at a time in a social group.
For example, I lived in a coop in college with 35 people,
and everybody knew everybody else, and we all felt pretty
close to each other. My friends who lived in coops at other
colleges of 60 people or more, however, said that those
communities tended to break into factions, ironically much
smaller than the maximum, usually only around 5 to 10 people.
In the dorms, some people only knew their immediate roommates.
So part of the reason I lived in the coop I did was so I
could know and hang out with more, not fewer, people.
The rule of 10 seems to be about the optimal size for
a tightly focused development team, however. 30 is too large.
If your project needs more than 10 it seems to me you need
to break it into pieces that can be handled by teams of
10 or less.
Which brings me to thinking about organizational group
size. It seems that if all of these rules hold, organizational
networks ought to form naturally into these sorts of subgroupings.
Collaborative projects ought to have subgroupings or teams
of 10 or less (including, if there is a "management"
team, 10 or less there) --- maybe people can belong to more
than one team at a time (a team lead could belong to a management
team as well as the team they lead). Then there might be
informal "pools" of 30 or less people, maybe people
that they regularly collaborate with. Then larger groupings
of five or more of these things into networks of 150. Scaling
beyond that I think requires carefully building interfaces
to avoid too much scattering of attention.
The more you folks
write, the less I have to think!
regionalisms. I found out last night that what most people
call a 'liquor store' or a 'convenience store,' Michiganders
call a 'party store.' Like, 7-11 is a 'party store.'
This just in: Meg
informs me that in Massachusetts, liquor stores are called
"packies" short for "package store."
headed with "experience design," "user experience,"
"information architecture," whatever. Over
Hack, Christina's been writing about the relationship
between IA and usability. Coincidentally, I've been mulling
the "What I Do" question actively for the last couple
weeks, pretty much since CHI2001. I've written lots of notes,
and talked to lots of people. I'm having trouble developing
cogent thoughts about it all, but I figure there's no harm
in sharing the muddle in my head.
Let me start by
saying that this problem, this trying to come to grips with
digital product design and all its ins and outs, the various
parties necessary to make it work, the infinite viewpoints
for getting it done--it's big. The more I think about it,
the harder time I have getting my arms around it.
In the past year,
I've attended three professional conferences: AIGA's Advance
for Design, ASIS&T's Information Architecture Summit,
and CHI 2001. At each of them, I've heard the same question
asked--"What is it that we do?" At each conference,
the participants have the same "Them" -- marketing.
And at each event, I've witnessed the same head-scratching
over turf wars in the design process -- for instance, at CHI,
someone said that it sounded like "information architecture"
was the same stuff that the folks at CHI had been talking
about for 20 years, and calling it "interaction design."
So, I've been thinking
Here's what I believe
happened. For the last 100 or so years, a series of design
disciplines have developed into fairly well-defined professions.
You've got industrial designers, architects, graphic designers,
interaction designers, information scientists, etc. etc. They've
all been toiling away at their various problems, understandably
ignorant of the works of other designers, because the products
these designers made were all so distinct. A automobile dashboard
is not a building is not a subway map is not a software interface
is not a search-and-retrieval system for massive amounts of
then the Web happened.
(Cue chorus angels, and light streaming down from above.)
And all these different
design professions ran headlong toward this new medium, because
they saw the value they could bring to this new medium. And
they butted heads with a resounding THUMP, and, staggering
back, wondered who these *other* designers were, and why they
were all jockeying to do similar work.
And it became apparent
that, all these seemingly different designers were actually
more alike than they'd realized. If you abstracted up a level,
they all did pretty much the same thing: attempt to solve
complex problems through a process of design.
The gut reaction
to see these foreign specimens claiming similar ground was
to engage in a turf war. To claim their way was the best way,
and who are these others trying to take my work away from
This is an unfortunate
response because what's clear to me, more than anything else,
is that we need to draw from backgrounds as varied as can
be. The Web has created a space for a uniquely synthetic type
of design, and a need for processes to address this.
It's also calling
for generalists to lead project teams. The analogy that's
most obvious is the film director. Directors typically come
from a particular specialty--writing, editing, cinematography,
acting, etc.--but must be able to work with a vast variety
of craftspeople, from set designers to sound effects specialists
to costume designers, etc. etc. The role of the director,
more than anything else, is to provide a guiding coherent
vision for the film being made. This does not (necessarily)
mean being a micromanaging ogre, telling people how to do
their work. The best directors know to give their team autonomy
in their individual areas.
From what I understand,
most architects work in this fashion, too. They have a vision
for a building, but the actual work is done by draftspeople,
structural engineers, plumbing consultants, interior designers,
etc. etc. The architects job is to hold it all together.
So, if nothing
else, there seems to be a role worth distinguishing in Web
design, an Experience Architect or Experience Director, an
acknowledged creative lead for web projects, a single person
responsible for holding the vision of the final product. This
person should probably not be the Project/Product Manager
(in the same way that a film's director is often not its producer)
-- dealing with the administrivia of a project and dealing
with it's vision are two very separate things.
Um. This thought
train has ran out of steam. Good night.
what I meant. Nick,
The Smartest Person On The Web (Oh, wait,
maybe that's Steve.
Anyway...), wrote in about the Semantic Web:
Your point, sirrah, is modern linguistics: semantics
is all very well, but when is someone going to come up with
an adequate take on the Pragmatic Web?
Otherwise, it's SGML repeated as farce.
(langue and parole, langue *and*
I don't know what
languid parolees have to do with the Web, but it might be
worth looking into.
the individual. I've recently been reading about the power
of Group-Forming Networks. GFNs are "Networks that support
the construction of communicating groups create value that
scales exponentially with network size," says David Reed,
creator of the model, in his essay "That
Sneaky Exponential--Beyond Metcalfe's Law to the Power of
Community Building." This stuff has interesting relationships
to the Rule of 150, which Malcolm
Gladwell discusses in The Tipping Point --
up to 150, there is a "community memory"
that works such that either you know how to find something
out/do something, or you know the person who does. Beyond
150, you no longer know whom to turn to, and need to go through
intermediaries, and a communication breakdown occurs. Research
in cognition supports this, primarily Robin
Dunbar's work written up in Grooming,
Gossip, and The Evolution of Language.
has a point. Winer
called me out on dissing the Semantic Web, particularly
dismissing Engelbart and TBL. And he's right... Whatever their
intentions, those guys invented products that have enormously
benefitted the world. I don't know exactly what *my* point
is, except to say that I guess I just hope that pointy-headed
academics would spend more time addressing the real needs
and concerns of people, and less time hand-waving about how
Things Should Be.
of Daves... I had a delightful lunch with David "JOHO"
Weinberger, perhaps best known as a co-author of The
Cluetrain Manifesto. We talked about all manner of
things, from personalization, to The Semantic Web, to riding
the Clue-Gravy-Train for all it's worth, to how he and I can
travel and, thanks to the Web/email, find folks we "know"
in almost any city we're in, to how communities subvert designers'
intent (he'd never heard of the riotous
Family Circus reviews on Amazon), etc.
many conferences, so little time. Via
Rock Eater I've learned of "Social
Dimensions of Engineering Design," the Mudd Design
Workshop III, taking place May 17-19 at Harvey Mudd College
in Pomona, Ca. With topics like "Social Issues and Themes
in Design," "Collaboration in Design," "Design
In and For A Complex World," it all sounds very interesting.
Semantic Web--Who Cares? The
May issue of Scientific American features an article
outlining the Semantic Web, Tim Berners-Lee's latest drum
to beat (see the W3C papers on it here.)
The article explains:
The Semantic Web will bring structure to the meaningful
content of Web pages, creating an environment where software
agents roaming from page to page can readily carry out sophisticated
tasks for users.
This is one of
those Extremely Noble and massively complex endeavors wherein
academics, removed from the real world, attempt to solve a
problem nobody has. (I fear MIT's
Oxygen project will suffer a similar fate.) The only reason
the Semantic Web gets any press is that Tim Berners-Lee, the
"Father of the Web," is working on it.
History has shown
us that technology inventors often haven't the faintest clue
about the device's actual use. Did folks line up to hear what
Philo T Farnsworth had to say about television?
tend to have Extremely Noble intents for their technology.
Douglas Engelbart was
obsessed with "augmenting" intellect, and the
first development of the WWW was definitely along those lines
(for academics to compare notes). And the Semantic Web is
no different. From the SciAm article:
If properly designed, the Semantic Web can assist the
evolution of human knowledge as a whole.
no one, apart from some self-appointed Bringers Of Fire, wants
their intellect augmented, nor really cares about the "evolution
of human knowledge." The Web, this extremely exciting
hypertext platform, serves other human needs and desires--primarily
to communicate, also for sexual release (porn!), and for finding
information of personal relevance (what's the weather where
I'm traveling? how can I do my job better? where's my favorite
face will freeze like that.
pic captures how I feel much of the time. The loverly
lady on the right is the inestimable Jennifer Kilian, with
whom, after this shot was snapped, I proceeded to get into
a delightful shouting argument about the various roles of
Boy with Blue Dog." And other great works can be
found at the Museum
of Depressionist Art. Worth many chuckles.
from... Providence! So now I'm in Providence, RI, home
of Brown University, RISD, and, most importantly, Jen
I'm staying here while attending and speaking at the Seybold
Seminars in Boston.
in... comics? Yes, some kooky kids at CMU experimented
on 'the best way to put comic books in electronic form' and
wrote about it in the paper "Comic
Books: A Case Study for Redesigning Traditional Media and
Assessing Entertainment Value." (PDF) It's actually
an information-rich paper on measuring responses that aren't
it's Friday... It must be SF. But for less than 24 hours.
At which point I get on another plane and head to Boston for
Seybold. If you live in Boston and want to play, let
Quick Notes on CHI 2001. This website began nearly three
years ago with my
notes from CHI 98. I'll probably write up something more
fully, but in the meantime, stuff to chew on.
social navigation. A bit back I mused on the notion of
leaving notes via wireless devices for others to pick up.
GeoNotes allows users to mass-annotate physical locations
with virtual (multimodal) 'notes', which are then pushed
to or accessed by other users when they come into the vicinity
of the location. It is based on location positioning technology.
GeoNotes employs a number of social filtering techniques,
which all rely on logging of usage rather than content.
What's great is
that they're using social
navigation techniques to help users sort through the inevitable
clutter of such a system.
Some kids at Microsoft are using the massive data repository
that is USENET to inform the design
of information visualization systems to help visitors understand
community dynamics. It takes some clicking around to get,
but there are many good ideas here. For the thoughts behind
the tool, download "Visualization
Components for Persistent Conversations," they paper
they submitted to the conference. Folks interested in online
communities would do well to read Marc
Smith's home page.
Go with the
VisualFlow. Sony's VisualFlow
was demo'ed. It's a super-nifty media browsing tool that uses
Interface models to allow people to quickly sift through
their media collections (typically photos).
know, it's hard to keep up with ~200 email messages a day
while you're travelling. I'm at CHI 2001. I simply don't
have time to write. Lord knows when I will. Lots of ideas,
last night. A flick very much worth seeing. A treat for the
mind, but not so cerebral that it's distancing. The narrative
conceit (revealing the story back-to-front) works amazingly
well--it puts you in Leonard's head. Great for post-viewing
discussion. Though, I found myself quite disoriented for a
spell following the show--when you spend 90 minutes (or so)
consciously reversing the order of witnessed events, it takes
a bit to get your mind out of reverse and back into drive.
work. XPLANE redesigned
their site, to great success. Alongside their inestimable
xBlog they've added
the bBlog, focusing
on business issues (like I need more links to follow). Niftiest
is the full site map they include on the bottom of every page.
I first saw this kind of thing at Peter van Dijck's Move
To Colombia site. I think it's a great device. Peter was
good enough to write
about the map's success. I'd love to hear what kinda clicks
it gets for the XPLANE folks.