Inkwell: Authors and Artists

    
I've had the (mis?)fortune of having five different mobile phones in semi-
regular use over the last couple of years (largely for good reasons,
actually, but that's neither here nor there). All came from different
manufacturers (Sony-Ericsson, Motorola, Samsung, LG, Palm), and not one of
them had an interface that I could entirely stand. It appears that creating
a truly usable small-format UI is a damnably complex problem.

A question on an entirely different subject, though: I've noticed that
"design" has become, in the last year or two, a dominant metaphor/language
for organizational and business consulting. There was a BusinessWeek article
a few months back arguing that design is becoming as big a buzzword in
consulting as networking was at the turn of the century. Stanford University
is opening up a new D school, I'm told, with a strong push towards applying
design thinking to non-traditional design areas.

Have you seen this in your own work? Have you been asked to consult on or
talk about subjects with few direct interface design aspects? Conversely,
have you observed ways in which interface design processes and thinking can
have broader application?
  

    
"Curb appeal" is a great way of putting it. Sure, the house looks nice
from the street, but what's it like to live there?

It's tough to get over visual appeal when 1) we're wired to respond
strongly to that and 2) it's tough to know and demo what it's like to
use/live in something until, well, you use it. You can market
"ease-of-use" of course, but anyone can say that. I'm sure Comcast, my
cable TV provider, advertises its on-demand and TV listings services as
easy to use, when they are anything but.
  

    
A small format UI is extremely tricky. Not only do you have the
limited real estate to deal with, but the audience population is
incredibly massive. Nokia alone has reportedly *1 billion users all
over the globe*. Designing for that type of audience is daunting.

"Design" aka "Innovation" are definitely in the business consciousness
right now, especially with both BusinessWeek ("In Innovation") and
Conde Nast ("Portfolio") having competing new magazines in the space.
Designers are turning their attention to things like services,
processes, and even designing organizations themselves. And I do think
design brings a way of working and an attitude of prototyping that is
different than is traditionally taught in MBA programs, say. Some
people call this "design thinking" but I tend to shy away from that
term, as I explained earlier in this conversation.

In my own work, I've been involved on many projects that have required
a wider scope than simply designing the individual product. A lot of
my projects are like that these days, in fact. I have also done pure
"Design Strategy" work as well, but I enjoy making things a little too
much to get wholly involved in that. There are companies like Stone
Yamashita and IDEO's "Transformation Practice" that specialize in just
this sort of thing.

The design of services is the next big thing in interaction design,
some of which--perhaps many of which--won't have "interfaces" much at
all, at least not from the customer's point of view. Services can be
designed in a similar manner to a product, with some differing methods.
  

    
Designing for different cultures and cross-cultures is one of the
major challenges for interaction designers today, one which I mostly
skirt around in the book, since I definitely don't feel I'm much of an
expert in that area. 

Being appropriate is really important. What makes sense in one culture
simply doesn't elsewhere, and if you aren't part of that culture it is
difficult to know that. By culture here, there's a wide scale:
individual organizations can have unique cultures as can entire
countries or regions.

To my mind, there's only a few ways to bridge this cultural gap.
Research is the clearest one. Go to the environment, talk to the people
there, and find out what will be appropriate there. Secondly, you can
have your work reviewed by those in the culture for appropriateness. A
translator, in a way.
  

    

The interesting thing about the usage model clash that occurred on Fotolog
was that it would've been very difficult to anticipate.  It fell into a
category of what my friend, Gareth Branwyn at Street Tech has long
characterized as "the street will find it's own use for technology."

Fotolog had a simple and nice service which allowed people to open accounts,
upload a few photos a day (most people uploaded just one or fewer), and
facilitated a comment section below each photograph.  Friends could be
chosen by users, and the most recent uploads of these would appear as
thumbnails on the user's page.  All in all it was a nice, simple
interconnected community-based service.  A format that's fairly common
today, but a few years ago it was still relatively rare, when such
community-based sites were just beginning to emerge.

When the site was less than two years old, a popular Brazilian
journalist/blogger who had earlier joined Fotolog and loved it, began
writing about the site and sparked a huge wave of new users.  To be sure,
there was a wide range of skills and styles, but the largest bulk were young
people with cameraphones who focused more *on the social aspects made
possible by the comment logs* than on the photographs themselves.  This was
a perfectly legitimate usage of the system, but its dynamics overloaded the
site's servers and even two years later, Fotolog was struggling to keep up
with the continued huge influx of users with this different usage style.
But language and cultural differences also soon became apparent, and this
led , unfortunately, to some community rancor that took time to work
through.

Part of the problem stemmed from the ease at which people could open free
memberships, and other things which hadn't been previously taxed, such as
unlimited comment logs.  It was rocky going for a long time, but eventually
the founders and management were able to build up and adjust the system to
accomodate a much larger scaled usage and added Portuguese language support.
 As a user, I found the service slowdowns and outages frustrating, but the
overall phenomenon was fascinating and enlightening.

One thing that seemed to emerge as a lesson from this is that complex
systems, no matter how well the *intended* users are researched beforehand
or tested in models, can veer off in directions that are impossible to
anticipate.  This seems to point to a need to be able to adjust and tune the
system within the real world context.  Fotolog suffered because it fell
behind in its ability to adapt as quickly as the community and its usage
model was evolving.

Dan, you deal quite a bit with service design.  There are so many more, and
more complex systems and services in our future.  What are the various
issues involved and strategies that interaction designers can bring to bear
in systems and services like this, where users may adopt very different
usage models from what was intended?
  

    

>One thing that seemed to emerge as a lesson from this is that complex
systems, no matter how well the *intended* users are researched
beforehand or tested in models, can veer off in directions that are
impossible to anticipate.

"Can go wrong, will go wrong" still seems to apply after all these
years.

Although perhaps, "can't anticipate, will happen" in this case.
  

    
>One thing that seemed to emerge as a lesson from this is that complex
systems, no matter how well the *intended* users are researched
beforehand or tested in models, can veer off in directions that are
impossible to anticipate.<

Might individual humans, society, the world and the universe qualify
as complex systems? 

BTW, Rather than concentrate on why complex systems "veer off" it may
be more interesting to investigate why some complex systems work so
well without (or perhaps because) humans consciously running them.

"Things don't work as we plan
  And isn't that grand!! 
  

    

That's an astute observation, <robertflink>.  I think that's true, but my
question wasn't about whether interaction designers should concentrate on
*why* complex systems make unintended behavior or usage likely, but rather
the issues surrounding how interaction designers can best prepare to alter,
adjust, and adapt the designs after and as they're being used for real.

While we can model and test certain aspects of complex systems beforehand,
it's not feasible in all situations to model them in ways that will reveal
how they'll work or human users will respond and behave in the real world
and at larger scale.  So I'm curious about the development of strategies to
anticipate more ongoing adaptive design.  Particularly in complex systems
and services.

Much of what we concentrate on in design involves the period before the
introduction of a product or system.  And we've learned the value of rapid
and multiple iterations and cycles of testing and refinement.  But once
systems are deployed, it's often more difficult to continue to alter and
adjust them for a wide range of reasons (legacy issues, fixed
infrastructures, or simply that it was not anticipated and therefore not
budgeted for).  Today we have version releases, but those often come at long
intervals, which in the case of services, could cause people to abandon the
system before it can properly adapt to their evolving needs.

Might we not only do good design as we've known it in the early stages, but
also begin to "design for ongoing design"?  Adopting strategies that will
make it easier to respond to changes as they occur?
  

    <scribbled by jet Sat 7 Oct 06 10:08>
  

    
[scribbled and reposted to fix a grammar error. --jet]

Coming in late, lots of projects due at school this week...  Managed
to read "Designing for Interaction" on the bus last week and thought
it was a great introduction to interaction design for non-designers.
It's the sort of book I could give to my mom (or my boss) and say,
"This is what interaction design is" or "this is why we need to hire
an interaction designer and stop letting the engineers develop
interfaces".

I just started the undergrad Design program at Carnegie Mellon this
year after ~15 years working in the tech industry doing everything
from low-level coding on supercomputers to consumer electronics.  Most
of my classmates are fresh out of high school and have grown up with
the Internet, the WWW, mobile phones with cameras and laptops; the
Korean students have also grown up immersed in a seriously advanced
mobile communications culture.

It's safe to say that the entire class is pretty technically advanced
in one way or another, from already knowing Photoshop and SolidWorks
to having created web sites, blogs and shown work in online
portfolios.

Right now we're all freehand drawing on paper with Prismacolor, making
3d models out of bristol and learning design fundamentals the way they
were (I'm guessing) taught 30 years ago.  We'll do this for a full
school year, from what I understand, then split into CD
(communications design) and ID classes and go on our merry little
ways.

If you were on the undergraduate curriculum panel for a top design
school, what would you want to see included in the core design program
to develop skills needed for interaction design?  Do you think the
fundamentals are still the same fundamentals?  Or do you think that
there are new or different core skills that students fresh out of high
school should be learning?

Going a step further, should "interaction design" be awarded some
capitalized letters so it can stand beside CD and ID?
  

    
We use the initials IaD, but I don't know how widespread that is.

Jet, do you have any human factors, psychology, huamn-computer
interface (HCI), or anthropology classes? As both Dan and Jim
mentioned, gaining an understanding of how people actually use
technology is key to designing things well, and more than any other
discipline, IaD is smack dab at the intersection of cognition and
emotion.

One of the most informative and scary presentations I've seen was by
BJ Fogg on Persuasive Technology -- there's a Stanford lab and a book
of the same name. How a site like Amazon manipulates you and what they
know about how people choose paths through interactions was really
unsettling. And this wasn't merely graphical paths, but things like
creating deliberate delays after certain clicks and fast response
after others.
  

    
Not many.  Checking the ID requirements to graduate, I find one human
factors class in the second year, a psych class required of all
freshman, and a couple of design history and design/social
responsibility classes.  There is a specialization (a few classes less
than a minor)in HCI available for some design and CS majors.

I don't know how (if) one could teach some of the things Dan discusses
in an undergrad program, which got me to thinking about what would be
the core skills needed to perform those tasks.
  

    
I typically use IxD (the acronym that the Interaction Design
Association uses) as shorthand for interaction design...

Some schools have already started undergraduate interaction design
programs and I know Carnegie Mellon is considering having it as a third
track alongside communication and industrial design for undergrads.
One of the reasons I wrote the book was that I thought it could be used
as a textbook for undergrads (probably upperclassmen). I taught an
intro to interaction class at CMU for two years and was frustrated
there was no good textbook--I kept having to pull pieces from different
books and online sources to make it work.

Spending a year learning to draw well--especially with teachers like
Mark Mentzer and Karen Moyer, both of whom are wise and inspiring--is
time well spent. I hate the fact that, despite several drawing classes,
my skills there are still poor. Being able to draw concepts rapidly
and well is (and probably always will be, even when our whiteboards
become giant digital screens) a core skill.
  

    
Jim, it sounds like you're talking about adaptive design: objects and
systems that change their form and content over time to better meet the
needs of their users. This is a really fascinating area to consider.
We're starting to see a little of this, with objects like Adidas'
Adidas 1 running shoe that changes it shape based on how the wearer
runs, and on a number of websites that serve up news based on topics
the user has read in the past. (Curiously, software--which, because
it's embedded into our personal devices (the PC and mobile phones) and
has access to lots of user behavior and memory--has lagged behind in
this area.)

I do think these adaptive changes to products and services need to be
subtle and occur slowly over time. You don't want your digital devices
responding to one-off occurrences and whim. You want them to slowly
change to your way of working, the same way a well-design house
eventually conforms to the people who are living in it--and visa versa,
I should say.
  

    

I think you're right that many types of adaptive change should happen slowly
and be driven by long-period feedback from mass usage and behavioral
evolution, though quicker responsive adaptation on an individual scale (user
preferences, etc.) can be used to better serve the needs of individual
users.

As for products and devices, it's true that much of the software in them
today is embedded and unable to change or evolve.  But as more products
begin to have networked and upgradable/changeable software, we'll need to
have effective adaptive design strategies there as well.  Lots of companies,
such as cable operators and phone carriers would like to be able to add and
modify the software of their devices to suit the usage patterns of their
individual users.  Sometimes this could be simple things that we're already
seeing, such as presenting an offer for media that matches or compliments
the user's past choices.  Another way future devices might adapt, would be
based on a user's individual device interactional usage patterns, involving
speed or navigation.

Shifting gears a bit, maybe we can spend some time noting products, systems,
and services that have successful interaction design.  Dan, could you
describe some of the things out there that you thinks embody exemplary
interaction design?
  

    
I recently came up with a list of what I thought were the top ten
interaction designs for a discussion on the IxDA's mailing list. I'll
flesh them out a little here. Most of these have been around for at
least several years, if not a few decades, so they have been refined
over time (time being the only real judge of good design, really):

In no particular order:

1) The automatic teller machine (ATM). Used by hundreds of millions of
people, even those with severe disabilities.

2) The email client. The first and (arguably still only) killer
internet application. While it has its flaws (such as not scaling well
to hundreds of email per day), it's still amazingly useful and usable.

3) The word processor. It simply changed how we humans
write--something that had remained fairly constant for hundreds, if not
thousands, of years.

4) The spreadsheet. Aside from changing economics and accounting
forever, the spreadsheet (VisiCalc and Lotus 1-2-3) was the killer app
that unlocked the personal computer for non-computer scientists.

5) Amazon. The first website that showed us that a website could be
much, much more. Aside from the most powerful recommendation engine in
existence, Amazon also introduced so many innovations that is difficult
to list them all: everything from the first social network to
one-click ordering to the Elastic Compute Cloud.

6) Tivo. A product that has changed another communication medium. It's
 simplicity and power and how the system works is a marvel.

7) Instant messenger, particularly AIM. Now standing practically side
by side with email as the great communication methods via the web.
What's so amazing about IM isn't only the instant messages themselves,
but also the buddy list: the ability to detect presence online and the
second communication channel of the status message.

8) Airport check-in kiosks. Like the ATM, another simple to use device
in what could potentially be a high-risk, frustrating situation.

9) The karaoke machine. A way for friends to communicate complex
emotions via music. If you don't believe me, check out the karaoke
scene in Lost in Translation again. :)

10) The web browser. Pretty self explanatory.
  

    
I'm not so sure about #'s 2, 3 and 4.  Maybe it's better to say, I
don't understnad how they can be on the same list as simple devices
like ATM machines.

Last night I spent an hour explaining to my mom how to do the most
basic of things in Mail.app because she kept accidentally clicking on
the sort-order triangles or tripping over other tiny UI elements. (I
also got to explain what the "dot", "ballon", and "two-way traffic"
signs in the display mean.)  In a computer skills workshop last week I
watched a bunch of Freshman struggle with simple tasks in Excel and I
regularly get Word attachments from hell sent by engineers who should
have just sent plain-text email.

Using any of those three items correctly or well requires a fair
amount of technical sophistication and training.

Maybe email, word processors and spreadsheets are great examples if
you're already technologically savvy or have domain-specific
experience?
  

    
I'd add Google to that list.
  

    
Not everything can--or should be--simple. Simplicity reduces control,
and sometimes you need control in order to accomplish the tasks you
need to do. While I agree that Spreadsheets and the Word Processor have
become bloated with features over the years, you can't deny they are
still used by hundreds of millions every day--perhaps with some
frustration, but also with a lot of success.

I'll agree that Google is a technical marvel, but less so a design
one. Their innovation was a technical one: displaying the best search
results. Which is not to underestimate that, of course. Perhaps "The
Search Engine" as a class could be added in there.

Love to hear other people's lists. I bet Jim has one...
  

    

Alright, now we're getting to the fun part!  I'd also love to read other
peoples' lists.  Before I post a list of things that I think embody good
interaction design, I'd like to add my comments to the issues <jet> brings
up.  Dan hits on two specific points I agree with, the first being that not
everything can, or should, be simple, and the second being the problem of
"feature bloat."  The second has degraded the user experience of a number of
products that I initially found to be exemplary.

Unarguably one of the major pioneering heroes of user experience and
interaction design is Douglas Engelbart, who along with Bill English,
developed and tested the earliest mouse-driven computer-based graphical user
interface back in the 1960s at Stanford Research Institute (SRI).  In 1990
or 1991, I attended a presentation he gave at the famous Xerox Palo Alto
Research Center (PARC), and something he said has stuck with me many years
later.

To paraphrase him, he pointed out that in many other fields of human
endeavor we design tools and systems that enhance the capabilities of the
most expert practitioners or users.  But that in software and computers, we
often design *only* for the lowest common denominator. He posited that
society would greatly benefit if we began designing a much wider range of
interactive systems, including those that required a much greater amount of
learning and expertise to use, and would be much more powerful and useful as
a result.  And to demonstrate, he brought along two of his grandchildren,
both younger than ten years old, and had them demonstrate their amazing
speed in using chordic keypads.  Chordic keypads are different from regular
keyboards, in that they have five keys, matching the fingers and thumb, and
utilize combinations of key-presses to enter characters.  They're
significantly faster than regular keyboards, and I'd point out that this
method would allow entry using devices a user could hold out of sight, which
would enable unique usage scenarios.  The chordic keyboard is just one
example, and there are a myriad of reasons they've not been adopted, but I
think Engelbart's insight is an important one.

Feature bloating, which plagues many products and systems today, is in large
part an unfortunate artifact of the predominate product marketing strategies
and assumptions.  Many features in products are added in successive versions
so that this feature list can look longer and ostensibly more impressive on
the product's box compared to competitors.  Many products that start off as
simple and elegant solutions to a problem become unwieldy after they're
complicated with accumulated features.  This is also related to the fact
that most products and services are "me too" copies of things that have
gained popularity and value.  This feature bloating often masquerades as
innovation, when true innovation would really involve deeper changes that
would better fit emerging technological, form factor, and usage evolution.
Or better yet, involve the development of completely new and novel products,
systems, or services altogether.
  

    

Here's a short and incomplete list of products, systems, and services that I
think embody excellent user experience and interaction design, and my
reasons for choosing them.

1) The automobile.  I still find it miraculous that the automobile gained
its powerfully simple and efficient user interface so early in its
development. The steering wheel, accelerator, and brake working together so
seamlessly and transparently in a manner that allows vehicles to perform as
an extension of the driver's will.  Imagine if instead of this simple,
interrelated set of controls we had a vast array of buttons, each performing
one specific function (as many remote control devices have).  I'm a big fan
of simple interactive systems which support a complex range of usages.

2) The trumpet.  I'm a trumpet player, and marvel at the simple three valves
(used in a chordic manner!) from which all notes can be configured.  The
expertise and artfulness of playing comes largely through the embouchure (or
mouth and facial muscle usage), but the fingering skills are played out upon
a very elegantly simple physical interface.

3) The harmonica.  And even simpler musical instrument interface, the
harmonica yields a complex and pleasing sound from merely blowing into it
and moving it back and forth sideways.  Again, the skill is in breath and
mouth control, but I find the harmonica exemplary in that it's usage is very
elegantly intuitive and one of the few musical instruments that can produce
a pleasing result for non-expert beginners.

4) The wall thermostat.  These have unfortunately begun to get more
complicated, as they've become electronic and seen the addition of buttons
and software that requires navigating hierarchical software systems, but I
hold up the thermostat for another, more fundamental reason.  One of the
best types of user experience is one where there's almost no interaction
required.  In it's simplest embodiment, the user simply sets a desired
temperature and the system takes over from there, maintaining that
temperature either by turning on and off either heating or cooling, or
alternating between both.

5)  A washing machine utilizing fuzzy logic.  In Bart Kosko's excellent
book, "Fuzzy Thinking," he introduces the powerful technology of "fuzzy
logic," that's inherently adaptive and utilizes many sensor feedback points
to regulate processes.  He gives many examples, but one that stood out for
me was a washing machine that had only a single "start" button.  From there,
the system took over measured the load's weight, analyzed the colors or lack
thereof of the clothing, and constantly monitored the clarity of the water,
adding just enough soap and regulating the washing cycle to match the load
most efficiently.  Compare this to even good jobs of designing more complex
control panels for washing machines.  This is a perfect example of a user
experience greatly improved through the adoption of a technology that
completely does away with the user interface.
  

    

I should hasten to add that the above list is limited to products, so I'll
post a list of systems and services later...
  

    
Dan and Jim both said that not everything can or should be simple.

I agree.  But email, spreadsheets and word processors should be simple
by default (or at least have an easy way to get into a simple mode)
but I'm not aware of any currently on the market.  How many more
people would use these three items if there were simple versions that
didn't require much in the way of computer literacy?  (Says the guy
who still uses Emacs/LaTeX to write term papers...)

I'll pick an arbitrary value and say that my mom doesn't need 2/3 of
the features in Mail.app.  How do I make them go away so they'll stop
interfering with her attempts to send a simple message?  

So that's why I question them being in the same list as an ATM, kiosk
or TiVo.  They're great tools once you learn how to use them, but it's
not just a walk-up-and-use-it experience.  And no, I can't defend my
bias that all things should have a walk-up-and-use-it mode.
  

    

I think you make a great and valid point, <jet>.  We would certainly benefit
by having very simple, stripped down versions of many of our most core
applications and services.  For example, while I have Microsoft Office
(which contains the mega-powerful word processing application, Word), I
almost exclusively use the very simple Apple OSX text application TextEdit
for nearly all of my writing.  I appreciate its simplicity, speed of
launching, and freedom from feature bloat.  I'm sure Dan's got much to add
to this issue.

Below are five more products, systems, and services embodying great user
experiences and interaction design.

6) Intuit's Quicken checkbook and financial software.  Quicken, particularly
the early versions, were real lifesavers for me and my business.  I'd always
been fairly organized in my financial and record keeping activities, but at
the end of the year I would have to use a spreadsheet program to produce the
lengthy report.  But once I discovered Quicken in the early 1990s, I really
appreciated its simple and intuitive interface, which was based on the
traditional checkbook.  Plus, generating reports, and looking up that last
time you subscribed to a magazine or how much you paid for an item, was a
snap.  Even though it has suffered from a great deal of feature bloat over
the years, it's still one of my most valuable tools.

7) The Palm PDA.  Before the Palm PDA arrived on the scene, I had to work
very hard to maintain my list of contacts.  From the earliest Macintosh days
in 1984 I kept lists of my contacts and their information, and would print
these out in super tiny text that I'd keep folded in my wallet.  In the
early 1990s I began to use a small, slim Sharp organizer, which I liked, but
which had a very dangerous and annoying flaw.  There was no way to simply
back them up.  This was a device that allowed you to spend untold hours
entering data, but if it died, all of that was lost.  I finally ended up
buying a much larger Sharp organizer (for something like $300), an expensive
cable that allowed my small organizer to be connected to it for data
transferral.  And then an even more expensive kit that allowed the large
Sharp organizer to be cabled to a PC for data transfer and archival.  In
other words, I was willing to spend several hundred dollars and go to great
lengths simply to insure I didn't lose my data!  When the Palm PDA came
along, it was a godsend.  It did everything the old organizer did and much
more, and it had many excellent innovations, including downloadable
applications, infrared beaming, and simple and painless synchronization with
a user's computer.  I was love at first use.

8) The latest generation of vehicular GPS navigation devices.  I worked on
the prototyping of one of the earliest wearable GPS units in conjunction
with Texas Instruments for the U.S. Army in 1986, and since then have had
numerous consumer GPS devices.  They've improved in ease of use and value
steadily with each generation and the latest models are really fantastic
aids while driving.  Earlier in this conversation I described how one really
saved me when I got lost near Boston's Logan Airport.  And my small Garmin
handheld unit allowed me to wander through the winding medieval city streets
of Montpellier, France without fear of getting lost.  I merely followed the
"breadcrumbs" of my wandering path from my hotel, and marked key waypoints
as I encountered them.  Really brilliant technology and an ever-improving
user experience.

9) Flickr.  I've really enjoyed the power unleashed by the use of "tags," or
keyword metadata in a number of recently emerging web-based community sites
and services.  I became interested in the power of metadata in databases
back in the late 1980s, when I began work on my InfoSpace whitepaper and
conceptual network browser model.  Keyword tags, one form of metadata,
enable powerful and flexible categorization, grouping, association, and
interrelationships in large sets of data and this is particularly useful in
an enormous and growing database of photographs.  Flickr's not the only
photo community site that utilizes tags, but it also has open APIs
(Application Programming Interfaces) that allow third parties to create
additional applications that use Flickr, and has many other innovative
features that make it very powerful.  Google also benefits from having open
APIs, and this has emerged as a powerful strategy for allowing the larger
world to continually build value into systems.

10) Wikipedia.  Wikis, or websites that allows the visitors to add, remove,
or edit content have emerged as another powerful model for harnessing the
power of users.  While many people are content to merely use Wikipedia as an
encyclopedia, it's wiki-based structure also makes it easy for people to
participate in the building and refinement of its content.  The result is an
encyclopedia that has grown with amazing speed, and the various checks and
balances in place insure that mistakes, inaccuracies can be edited out over
time.  And another valuable quality is that obscure topics and subjects that
might not make it into a traditional edited encyclopedia, can have
references.  I think a lot of emerging internet services are supporting and
giving simple access to what WIRED Magazine editor-in-chief, Chris Anderson
has dubbed "The Long Tail," or the huge number of things that fall outside
of the bulging part of any bell curve, which only contains the most popular
or common denominator elements.
  

    
I just started reading John Maeda's new book, "The Laws of Simplicity,"
which is very appropos to this conversation.