Windows 2020 Lifestreams and Objectlens's
Gregory Alan Bolcer
Sun, 06 Jan 2002 09:26:13 -0800
Gerlenter seems to think that Scopeware, a software
that does the "Slaughterhouse 5" view of a user's
activities will be the most useful metaphor for
the next shift. Some of the other metaphors
have been around for some time now including the
task gallery, data mountain, "urp", and others
as CHI people have been talking about these things
for at least a decade.
The thing that seems to be winnign though is the
virtual desktop. The metaphor is a small map overview
with some number of workspaces that have all the shortcuts
and documents you need user-self-selected for a particular
work mode or task. You access these spaces by moving your
mouse off the end of the screen into the next desktop. It's
not widely used, but almost every Linux user has theirs
set up that way with a lesser percentage actually using it.
There's even other companies like Teleffect that work across
all a user's desktops.
For instance, I am sitting at a keyboard and a mouse which
are(n't) attached to my SGI O2 (Logitech wireless) with
my whole virtual desktop environment. Using Teleffect, I can
install it on my Mac OS X machine, in the same room, my
XP machine at my office. Any time I want to use any of these
other computers, I want to just drag off the screen to the
"direction" of that computer and then automatically
use that one. If they can make that one work securely
and through firewalls and wrap that up into a desktop
metaphor similar to a 3D CA/Unicenter type metaphor
(or Jurassic Park one) you have solved one of the
first conceptual problems that the user faces, having
to interrupt tasks because different software and data runs
and lives on different computers.
The second major user metaphor problem new age OSs face
is the "data plurality" problem. It's something that they
KM people have been struggling with for ages and don't
have a good meta-phor. Typically, people like to
have things that are conceptually close to display
as physically close. The problem with that is, you
may have a file or data fragment that is conceptually
close to two different things. For example, a user
wants to keep a Web article about some subject in
one folder, but then also wants to keep a quote from
someone in the article in another folder. He may
want to do that because the article is about a
competing project and the quote is from one of the
potential prospects or business analysts. There's
a couple of solutions using current desktop metaphors:
1) put the same copy of the article in all
3 different conceptual folders they have on their system,
one for competition, one for prospects, one for
2) Put the orignal copy of the article in one folder
and put shortcuts in the other folders.
3) Save the URL in booksmarks or a bookmarks file and
pray the content doesn't move, get archived, or go away.
4) save the content to a single search portal or a knowledge management
system and hope that the user can sort out the content on
future access using automated or semi-automated parsing
and indexing, i.e. I want to know what tech analysts deal
with subject X which returns some KM or search results.
5) Have "Recent Documents" or "Recent Programs" somehow know about modal
tasks and cluster them somehow.
6) If you're realy clever you can manually assign closeness
indices or rely on some smart folder technology to populate
your folders using some consensus building across users
(even though some of the content may be private) and use
some Java applet springs and boxes diagram.
The bottom line is, there's no good user/desktop metaphor for
having some single piece of content being easily manipulated
and displayed in two places at the same time. It's even
worse if you're talking about intersecting folders. If researchers
can solve those two problems with a usable metaphor, I predict
that would be a strong foundation for overcoming the
2D point and click.