Date:         Tue, 27 Sep 1994 13:25:36 -0400
   From: Susan Jacobson <[log in to unmask]>
 
     In reference to Mosaic, Eric Siegel writes:
 
     Is this really what people are so excited about? I
     personally really prefer the speed and flexibility of text.
     But then again, I like a command line interface ("what good
     is it learning how to use a computer if it doesn't seem
     cryptic to the uninitiated?")
 
     I once read a *very* compelling editorial in Wired magazine
     by Negroponte, the director of the MIT Media Lab. In it, he
     says that transmitting huge bandwidth over the wires is a
     waste of computing power and infrastructure money. What
     should be transmitted are "clues" that allow the local
     computer to reconstuct full motion video or images, or
     whatever using its own computing power.
 
   You are absolutely right. Another problem with Mosaic is its
   ONE-WAY nature. It can talk to you, but you cannot talk to
   (talk back to?) it. (Unlike the MUSEUM-L, for example!)
 
   Mosaic is an extended form of publishing, in my opinion.
   It is not about having a dialogue with an audience.
   It is a one-way transmission of information from those who
   are "in-the-know" to those who are not. And, personally,
   given the way it is set up, I do not see it evolving into
   a dynamic platform.
I have set up a number of Mosaic servers inside our company, and
maintain one of them. [So much for the appeal to authority part of
this argument :-)] I tend to view Mosaic currently as a transaction
based interaction mechanism. Not totally interactive, but having an
interactive flavor. The two major paradigms are hyperlink following
(click on this highlighted term or image, which starts a transaction,
which returns another page, possibly from a different site), and form
submission, where the client is presented with various input widgets,
like type-in fields, radio-buttons, etc. In this second flavor, the
client fills out the input-fields, and begins the transaction by
submitting the information, which again, typically returns another
page.
 
So for instance in a tic-tac-toe game, when you click on the square you
wish to place your mark in, what really happens is that your machine, sends
a request (one implementation would be to send a url, with the x and y
coordinates), and the server computes which square you clicked in, and
what the server's move should be, and returns to you a page, with your
mark, and it's next move. The thing to notice about this form of interaction
is that all the computation is done on the server, which is very limiting
for interaction, but is also fairly secure for the client (the mosaic user),
since arbitrary things can't be done on the local machine.
 
It isn't clear to me that this is an inherent limitation to WWW, which like
most of human enterprises is in a state of flux. There are a number of
proposals to expand html (the markup language used in WWW documents).
One of these vrml (virtual reality markup language), which aims to
allow people to serve 3D interactive environments, is clearly going to
require local processing, and probably changes to http (the protocol
used to communicate between client and server).
 
I believe speed problems are a temporary problem. There are at least
four places where speed constrictions can occur. I've already
discussed problems with the current protocol. The other places are the
server, the client, and the connection between them. It is clearly
possible to deliver full motion video in realtime to the home. The
existence proof is called cable.  How digital transmission to the home
will play out is anyone's guess, but the existence of cable indicates
that high bandwidth transmission to the home is possible, and if you
believe the industry hype, it is highly likely, though who knows
exactly when. I believe most of the current speed problems come from
tying to stuff megabyte sized images down 1200 baud pipes. Client and
Server speeds seem to be following the computron growth path of
doubling every 18 months, so they are likely not to be the stumbling
block.
 
Today, sitting at the end of a T1 (1.5 megabit) connection to the
internet connected to a server with a T1 connection, image-rich WWW
pages arrive at a usable rate (at least for me). The time to aquire
movies (mpegs) is too slow. Obviously with slower transmission
bandwidths WWW becomes less usable. There are questions about what to
do in the interim. To wait and see how things shake out, or try text
WWW experiments, or to go for it, and only have a limited audience.
The most interesting socialogical aspect of WWW is it's growth. [In
many respects, the current discussions about citation is an indication
of one of the major difficulties a historian of this technology is going
to have. So many of the first html pages have already disappeared.]
One of the web crawlers (programs that walk from page to page over the
net) reports over 500,000 html pages that it discovered.
 
With respect to Negroponte's quoted remarks (which original I haven't
read). I think for most museum uses, his comments are extraneous. If I
want you to see an image of my nephew (a very cute kid) or a Miro, I
don't see much alternative but pumping down all the bits to you.
Passing down the 15 bytes "a very cute kid", while fairly speedy just
doesn't capture the experience of seeing a photo of a three year old
smiling, which of course doesn't capture the live experience (good and
bad!)  too well either, but that's another can of worms.
 
There are many other problems with the current WWW. Your graphics artists
will have fits over the limitations of what can be specified on a page
using html (no way to say center this image, for instance).
 
Now some personal notes from maintaining a page. Since June, I've
maintained a page that points to as many sailing information sources
on the net as I could find, as well as some personal sailing
information, I've collected.  [I'm going for a Masters in maritime
history, archeology and conservation at New York University, at night
- which is why I listen in on Museum-l] Since June, we've had
somewhere between 1000 and 3000 visitors (yes most servers track which
hosts connect to them, and which pages are requested. Only to the host
level, not user). Besides being a lot of work to find resources and
present them (and I've really not had the spare time to think hard
about the presentation), the major relevation was the steady stream of
email from people visiting the pages offering additional bits of
information, requesting information or just saying hi. One of my
collegues, Will Hill, claims that people don't really want more
information, just better relationships and I think there is some truth
in that.
 
Finally, my bias. I find the Web terribly exciting. From being able to
access the library of congress to check my bibliographic entries, to
grabbing images from the Louvre. From being able to easily pull over
friends latest papers, to seeing the literally hundreds of artists
exhibiting on the Web. From getting daily BOC updates (a sailboat race
which is totally invisible in most U.S. news media) to discussing 17th
century seamanship with someone in Swedan. From seeing the current
version of a bill before the U.S. House, to providing information on
my server on opportunities for sailing in square rigged sailing
vessels. Currently, the Web just touches the periphery of my work and
interests. If I stand on tiptoe though, I can just barely see how it could
become a major intellectual adjunct. We'll see.
 
Mark Rosenstein.
=--=--=
The open society, the unrestricted access to knowledge, the unplanned
and uninhibited association of men for its furtherance - these are
what may make a vast, complex, ever growing, ever changing, ever more
specialized and expert technological world, nevertheless a world of
human community.
  --J. Robert Oppenheimer