schricke@mist.cs.orst.edu (T Karl Schricker) (05/05/88)
The following posting is taken from comp.society. I am also interested
in this subject, and I believe the article should be cross-posted to this
group. Please note the follow-up is directed to the original source so
Mr. Jellinghaus can also benefit from any discussion.
Also, please address any books, articles, etc. that relate.
Thanks!
Karl Schricker
-------------------------------------------------------------------------
>From: jellinghaus-robert@cs.yale.edu (Rob Jellinghaus)
With all this Apple/Microsoft/HP unpleasantness going on, and with
all the debate about the current generation of windowing interfaces,
I have started to wonder what the NEXT generation of interfaces will
look like.
Right now, all the current interfaces are loosely based on the Xerox
PARC model--windows, icons, pulldown/popup menus, etc. My question
is: what's next? Is this paradigm the best of all possible worlds,
or is there something even more flexible and easy to use? And if so,
what is it?
This isn't just an idle question. Apple first popularized the PARC
style interface with the Macintosh, and the whole personal computer
industry is following suit. The workstation folks started the whole
ball rolling. I would think that the company (or individual?) that
comes up with the next great innovation will reap tremendous rewards
as the rest of the world tries to follow its lead. (Admittedly,
this wasn't true with Xerox, but hopefully the next innovator will
have more business sense. Imagine if Steve Jobs had been hired by
Xerox in 1975 or so...)
So what might this interface be? Well, there are a couple of trends
that may point the way:
- Technology is continually improving, pushing the limits of what
we can do. Advances in 3D, animated, interactive graphics (as
in Silicon Graphics' IRIS workstations and Evans/Sutherland
flight simulators) will change our desktops from static piles
of windows to spectacular moving, interactive displays. Work
in full-motion video processing will continue, bringing the
advent of video E-mail ever closer. (Of course, video E-mail
is just about the most boring thing one could possibly do with
such technology. What will the exciting uses be? That's what
we need to discover.) And sound will become much more signif-
icant to the user interface; when we have computers which can
synthesize the sound of whole orchestras in realtime (picture
a handheld Kurzweil 250), our computers will be as exciting
to the ear as to the eye. All this power may be only a couple
of years from entering the mass market.
- Of course, our computers will continue to get smaller, faster,
and more powerful, with optical storage media putting un-
precedented amounts of information at our fingertips. Cellular
phone links will become standard equipment, and ISDN and other
improvements in our communications systems will speed the
networking of the world. And such innovations as Sun's SPARC
chip and the ongoing work in parallel processing will give our
computers the muscle to put all these resources to use.
- There are other changes coming in the way we think of data.
Ted Nelson, in his revised version of "Computer Lib" (Microsoft
Press, 1987), discusses the obsolete notion of the "file".
The concept of the "file" as a single place to store something
will disappear. Apple's Hypercard and HP's New Wave are first
steps in this direction: Hypercard jumps from stack to stack
to find what it needs, hiding the details of which file contains
what from the user; New Wave has "hot links", which enable
one to, for example, paste part of a spreadsheet into a report;
if one then changes the spreadsheet, the figures in the report
also change, without any action on the user's part. This trend
will continue. In the future, we will find that our files
are no longer isolated; all the data in the system can be woven
into a web of interconnections, with all the nasty details
handled by the computer. If you want to work on something, you
just ask the computer for it; the computer can find it for you,
with no need for you to worry about exactly where the data is
stored. You spend no time on "managing your desktop"; the
computer handles all that for you. How exactly will you tell
the computer what you want to work on? Well, that's what the
next user interface will do for you. How? You tell me!
- And the next user interface will be, above all else, config-
urable. Xwindows has this, to a point; you can modify your
.uwmrc, .xrc, and other files to your heart's content; you can
even rewrite the window manager. But you have to be a hacker;
it's not trivial to write your own window manager. There is
a neat hack for the Mac II that enables you to easily change
the color of any part of the user interface, in a completely
intuitive way (I'm talking aout the "Kolor" cdev). And, from
what I hear, HP's New Wave is easily modifiable; Apple apparent-
ly modified it to look like the Mac. The next generation
interface will let the user easily change any part of it to the
user's own preference. How far can this concept be taken? We
will see....
With all the innovations described above, it's fairly easy to see
that our current windowing metaphor is not necessarily the best.
Many other possibilities present themselves. Perhaps something like
William (_Neuromancer_) Gibson's "cyberspace" will become the
prevailing metaphor; you'll move through shimmering spaces of data on
your screen... or perhaps it'll be nothing so far-out. Whatever it
is, though, I doubt it'll look much like what we have today, just
as what we have today doesn't resemble the command-line interfaces
of VMS, CP/M, et al.
I hope you didn't mind the (extremely) long posting, but these
questions of designing for the future will become ever more crucial
in the next several years (especially with all the infighting over
the current generation of interfaces), yet I've seen very little
speculation on where all this is going. If you think I'm full of
bunk, say so, but note that nothing I've said above relies on
unproven technology; it's all there, it just needs to get smaller,
cheaper, and faster, which computers have a habit of doing.
Here's to building the future!
Rob Jellinghausklee@daisy.UUCP (Ken Lee) (05/12/88)
In article <4450@orstcs.CS.ORST.EDU> (T Karl Schricker) writes: > With all the innovations described above, it's fairly easy to see > that our current windowing metaphor is not necessarily the best. > Many other possibilities present themselves. Perhaps something like > William (_Neuromancer_) Gibson's "cyberspace" will become the > prevailing metaphor; you'll move through shimmering spaces of data on > your screen... or perhaps it'll be nothing so far-out. Believe it or not, this kind of use interface already exists. Check out the NASA virtual environment work (reference: Scientific American, October, 1987). Also check out Myron Krueger's work at the U. of Conn. (reference: "Artificial Reality", Addison-Wesley, 1983). For the latest and greatest user interface work, go to the CHI'88 conference in Washington, DC this month. Great stuff. Proceedings of past conferences are available from the ACM. Ken Lee Daisy Systems Corp. -- uucp: {atari, nsc, pyramid, imagen, uunet}!daisy!klee arpanet: atari!daisy!klee@ames.arc.nasa.gov STOP CONTRA AID - BOYCOTT COCAINE