9.704 intellectual aspects of managing information

Humanist (mccarty@phoenix.Princeton.EDU)
Mon, 8 Apr 1996 21:58:53 -0400 (EDT)

Humanist Discussion Group, Vol. 9, No. 704.
Center for Electronic Texts in the Humanities (Princeton/Rutgers)
Information at http://www.princeton.edu/~mccarty/humanist/

[1] From: "Paul R. Falzer" <prf@callnet.com> (120)
Subject: whatever happened to the global VILLAGE?


I read with interest your comment on James Fallows's article, "Navigating
the Galaxies," which appears in the current issue of *Atlantic*.
Principally, the article briefly describes (perhaps only intimates) a few
developing techniques that some day may assist us in coping with the
information glut. Your comment was responding to part of the second paragraph:

"Precisely because no one can keep up with all the discussion groups, all
the new Web sites, and all the on-line libraries, people who will do
preliminary screening and point other toward promising sites have an
increasingly valuable service to sell....The most popular on-line discussion
forums tend to be not purely democratic but quasi-authoritarian spirit, with
an active "Sysop" (systems operator) who both steers and stimulates debate."

I doubt that many who subscribe to the Humanist would dispute your reply:
"Quasi-authoritarian? Me??? If he really means "quasi"...then perhaps the
sentence is not so bad. I think our perfectly good terms "editor" and
"moderator" are much closer to the mark than "sysop" and remove the need for
the imputed authoritarianism, quasi- or otherwise."

Jim Fallows is likely to take some heat for appropriating the term, "sysop",
which I believe originated with Fidonet. What he says probably applies
better to Fidonet discussion groups than to moderated listserves, Usenet
news groups, and CompuServe forums. Perhaps he should know better, inasmuch
as Kevin Grantham, the moderator of the CIS forum that Jim frequents,
performs much like you: Kevin participates, facilitates, and stimulates
discussion. I assume that you and Kevin censor postings that are frankly
commercial; thankfully, the Humanist and CompuServe's Agenda and Magellan
forum are not blighted with pyramid and Ponzi schemes. But, it appears to me
that just about anything else is fair game.

Nonetheless, I hope that Jim's essential point does not get lost. If I
understand it correctly, the article is suggesting: 1) that there is a
surfeit of verbiage on-line already and the quantity is bound to increase
dramatically. 2) One way to manage the information glut is to have what
amounts to a cadre of editors. 3) Another approach is to develop more
sophisticated means of indexing and searching. Jim's article focuses on the
third solution. He does not expound on what it means to "moderate" a news
group or discussion list. Perhaps for this reason, the reference to sysops
was hasty and should have been excised. If the topic is important enough to
mention at all, it warrants a more considered treatment.

Readers may notice other instances that display Jim's penchant for grinding
axes. For instance, he says: "THE natural impulse of the computer culture is
to look for ways to automate everything." I suggest that if there really is
a single computer culture, its desire is not to automate but rather to
control the content, access, and flow of information. I call this tendency
"revenge of the right fielder." Fortunately, the sheer volume of stuff
combined with diversity of the Internet and related information environments
is working against the geek polis. The threat is genuine and its
implications should not be underestimated. Still, there is only so much a
person can do from a dusky basement or a congressional cloakroom.

Later on, Jim Fallows shows his disdain for anything that smacks of
Microsoft. He says: "Under [Cooper's] direction researchers [at the MIT
Media Lab] attempted to use computer graphics not as a substitute for text
(the Macintosh-Windows approach) but as a way of making text more
meaningful." Here as elsewhere, Jim makes no secret of his preference for
the OS/2 environment. To parrot an old comedian, "preference" is a clean but
"prejudice" is a dirty, and the one tends to lapse into the other.
Significant issues get submerged in the process, in this instance, the
complex relationship between form and content.

These weaknesses (and others) notwithstanding, I think that Jim Fallows has
done a service by noting the limitations of current approaches to indexing
the contents of the Internet and previewing a few burgeoning technologies. I
was especially drawn to his discussion of Sun's indexing project which, if I
understand correctly, has the potential for becoming an inheritance based,
text parsing, engine. Jim correctly depicts the current algorithms, such as
Digital's Alta Vista, as search and retrieval tools. By contrast, the Sun
project might provide a means to _analyze_ the contents of a storage device
-- be it a local hard drive or CD-ROM, a local area network, or the
Internet. To my knowledge, there are no extant means of creating a
hierarchical database from word symbolic data in their native format, then
permitting users to do more than merely inspect the contents.

Perhaps my knowledge is out of date, and if so I hope that others will show
me an analytic search tool that already exists. The best examples that come
to mind are Quarterdeck's Web Compass and the Folio Retriever. The two work
differently, but both are essentially client-side middleware solutions. Web
Compass uses a variety of extant search engines to retrieve Web and other
Internet documents, then stores the results in a Microsoft Access 2.0
database. The Retriever requires the user to specify a URL or set of URL's,
captures the contents with or without graphics, then follows as many
hyptertext links as the user desires. This procedure results in a single
Folio Flat Format file, which can be indexed and displayed as a Folio
infobase. Besides some technical limitations that daunt both programs, users
either must have a very good idea of what they are looking for or they must
break rule one of conducting a search and just go fishing.

Writing on the Agenda and Magellan forum, Gary Oliver reports that he easily
filled more than 1 GB of hard drive space using the "go fish" method, and
less than fifteen percent of what reposed there was of any use to him. But
even if local storage did not pose a problem, these two approaches are
cumbersome at best and comprise only part of a viable solution. Microsoft
Access has its place, but not as a tool for analyzing textual information.
Folio Views has a powerful query function, but its .NFO format is
proprietary and unreadable without the commercial program. More to the
point, both programs function as mediators -- they come between the user's
system and the place where information resides.

Jim Fallows and I have a common experience with a software program that was
created by Lotus Development Corp., when the company's name actually
described its principal thrust. The program, Lotus Agenda, was a means of
creating, collecting, and organizing information. But it did more. In the
words of Gary Oliver, "Agenda is an information transformation product.
Whereas most products on the market perform information selection, Agenda
allows us to create and produce" (and I say, "and analyze"). Unfortunately,
as Lotus came under the helm of a guy who made his mark hawking Pepsi Cola,
innovation came to a screeching halt. Manzi was last seen floating on a
golden parachute; the products that were vitiated then summarily thrown into
a dust bin during his tenure included Magellan, Improv, AmiPro, Agenda, and

Jim Fallows and I continue to use Agenda, but we recognize that its useful
life is coming to an end, and not because it will no longer run on our
systems, but rather because it was conceived as tool for individual users
and its scope its limited to what can be stored locally, in a proprietary
format. Jim can speak for himself, but I am fairly confident in saying that
he is yearning for revitalized and contemporary Agenda -- a means of
bringing together the processes of creation, aggregation, and dissemination,
with active mediation as the province of the user. Concomitantly, I think he
would agree that barring a significant breakthrough, our sundry efforts will
be plagued by contending with myriad of formats, platforms, and storage
locales. Today, information management (in the widest sense of the term) is
a tower of babel -- an anachronism that gives the infogeek culture a
foothold, but leaves the rest of us frustrated. Jim is trying to precipitate
solutions. In this vein, I applaud his efforts.
Paul R. Falzer