[10427] in Public-Access_Computer_Systems_Forum
Re: Grumpy Old Cybrarians
daemon@ATHENA.MIT.EDU (Craig A Summerhill)
Thu Jul 18 20:08:15 1996
Date: Thu, 18 Jul 1996 14:01:59 -0500 (CDT)
From: Craig A Summerhill <craig@cni.org>
In-Reply-To: <PACS-L%96071017231693@UHUPVM1.UH.EDU>
To: Multiple recipients of list PACS-L <PACS-L@UHUPVM1.UH.EDU>
Reply-To: Public-Access Computer Systems Forum <PACS-L@UHUPVM1.UH.EDU>
----------------------------Original message----------------------------
Karen,
Re: your comments on the increasing use of the .PDF format:
On Wed, 10 Jul 1996, Bruce Gilbert <bg7601s@acad.drake.edu> wrote:
> >
> > Maybe I'm completely off-course here. Maybe I shouldn't be worried
> > about city workers with XT machines,
>
> Well, you wouldn't be alone in not worrying about such people.
> Clearly, people like Bob Gates and Marc Andreesen(sp?) could
> care less.
For some reason, Bruce, I have this gnawing feeling that you're not
really in a position to speak to the personal motivations of either
of these individuals.
If your point is that these people both have a vested interest in a
corporation that stands to gain by selling cutting edge hardware and
software, then just say that. (Actually, I don't think you even
have to say it -- most of us already recognize that.) Your insinuations
(or personal attacks in the case of your response to my earlier
comments re: Vannevar Bush and Memex last month) do nothing to
reinforce your argument and detract from otherwise interesting and
valid points you make.
On Fri, 12 Jul 1996, Michael Silverberg
<michaels@salieri.informatics.sunysb.edu> wrote:
>
> Having said that I think the issue of the bleeding edge versus the lumpen
> proletariat is one that will get worse but at least needs an airing. What
> should one do when every piece of software you own has an update every 6
> months? Have you even caught yourself wondering if a company that hasn't
> offered you an update recently has in fact gone out of business?
This is, IMO, a less-inflammatory description of the issue at hand.
A very real problem we (librarians) face is that the Internet is a
living laboratory.
It serves as an infrastructure for both the R&D needs of for-profit
corporations attempting to develop new products and also as a
production infrastructure for publishing efforts and libraries trying
to deliver information products to "real-world end-users" as Karen
characterizes them. Most end-users can't afford to constantly upgrade
their equipment to the latest and greatest, but neither can any company
competing in the computing industry afford to ignore the potential of
the cutting edge technologies. As libraries occupy the terratory between
the groups developing these products and the groups using them to do real
work, libraries are affected by the forces exerted by both groups.
Cutting edge users / vendors force us to adopt technology too rapidly
and trailing edge users force us to retain technology which has long
outlived its usefulness.
Prior to the commercialization of the Internet (read: the dissolution
of the NSFNet backbone), I was fairly forthright about my viewpoint
that much of this technology was not ready for prime-time. In the
months / years since the commercial cutover point, I have seen very
little if anything to convince me I was wrong in that point of view.
Speaking directly, I can assure you the commercial ISP now handling the
CNI-DOM gateway isn't providing half the service for twice the money
that SURANet used to. Nevertheless, the political realities forced
this premature cutover from a totally R&D / educational network to the
capitalist hotbed of small business and industry we now traverse daily.
And now we all have to figure out how to deal with it.
(BTW, Michael, there was an interesting article on the general theme of
"creeping featurism" in software published in _Byte_ magazine a couple
years ago. I might be able to find an exact cite if you can't find
it with the phrase "creeping featurism". Personally, I think it should
be required reading for *anybody* making purchasing decisions in
libraries.)
Bruce Gilbert <bg7601s@acad.drake.edu> continues:
> >
> > or about printing stacks of paper because you can't read a file, or
> > about using a whole lot of user education to reading a file format
> > on a medium designed to make document-access synonymous with
> > document-discovery.
>
> I guess I have to wonder about this statement. http, as a protocol,
> was designed to make file transfer easier and more robust; I doubt
> that the designers had "document access = document discovery" in mind
> as their primary goal.
What Karen has written is another expression of the same ideas you
didn't get (or rejected) in my earlier writings re: Vannevar Bush,
Memex, and hypertext.
Speaking as someone who was at the first "living documents" BOF Tim
Berners-Lee called during the March, 1992 San Diego IETF meeting
(23rd IETF), and as a painfully "matured" participant in the gads of
"uniform / universal / unified resource *" meetings that took place
over the following three-four years of IETF gatherings, I can tell you
Karen's characterization of "document-access = document-discovery" is
absolutely correct.
Furthermore, your characterization that HTTP was "designed to make
file transfer easier" is wrong. HTTP was *not* designed to simply
facilitate file transfer. In fact, I'm not even sure that was a partial
motivation in its development. There was a completely separate thread
of individuals within the IETF working on the enhancement of file
transfer issue (IAFA templates, FSP, etc.) at the time the HTML / HTTP
work began. And this idea of document linking leading to resource
discovery is very much what Berners-Lee envisioned in his initial
proposal to develop "document identifiers" (later URLs) as presented
in San Diego.
At the time of the San Diego meeting, I was highly skeptical of both
the scalability and the extensibility of the the Web architecture.
On the other hand, Berners-Lee armed with the real-world experience he
gathering in deploying this stuff within CERN, knew something about the
social aspects of hypertext that I believe Bush foresaw some fifty years
ago. Most of the rest of the IETF participants at that time did not
fully appreciate it either. Too many of us had our heads buried in
IAFA templates (FTP descriptive work) and glossy Gopher interfaces
and efforts to retrospectively document the Gopher protocol. CERN's
work in the social aspects of hypertext, which at the time was mostly
accessible to the vast majority of the Internet only through a really
lame vt100 interface developed at CERN, wasn't intriguing to most other
people -- excepting maybe Joseph Hayden and Marc Andreeson (both at
NCSA) who secured NSF grant funding to develop a more attractive
interface to the architecture (Mosaic).
It is only partially gratifying for me to read 1995 / 1996 interviews
with Berners-Lee in which he states the Web has not fully developed in
the direction he had hoped it would. This is largely because four years
later it has scaled up much better and proven to be a much more extensible
infrastructure than I (and a number of arguably more intelligent IETF
participants) ever thought that it could. No, it isn't perfect, but
there is a tremendous amount of unfulfilled potential in it still.
Finally, Bruce Gilbert <bg7601s@acad.drake.edu> writes:
>
> In the case of .pdf files, that means that we have to figure out
> strategies for users to access this stuff; if we have to "kludge
> it out" sometimes, that's fine, but we also need to be diligent
> in our quest for even better ways to get what is needed to "the
> reader."
On this, we are in agreement. The unfortunate reality of occupying
that critical middle ground between vendors and users is that we're
putting together a lot of "real-world" applications with chewing
gum, spit, and chicken-wire. It would be nice if no user (nor a
single librarian) were ever frustrated because the infrastructure was
universally robust, the information architecture(s) were meticulously
and intelligently crafted, and all the data available to us was of
the highest quality, lacking bias, and perfectly formated for our
immediate local needs (whether those needs entail printing, downloading,
reading on screen, or cutting and pasting).
Personally, I can't envision a day within my lifetime when such a
description will apply to the networked environment we've thus far
crafted... but, hey, I'd be happy to be pleasantly surprised.
It would make *me* less of a "grumpy old PACS-Ler". B^)
--
Craig A. Summerhill, Systems Coordinator and Program Officer
Coalition for Networked Information
21 Dupont Circle, N.W., Washington, D.C. 20036
Internet: craig@cni.org AT&Tnet (202) 296-5098
