9.507 Java and society

Humanist (mccarty@phoenix.Princeton.EDU)
Wed, 31 Jan 1996 18:59:36 -0500 (EST)

Humanist Discussion Group, Vol. 9, No. 507.
Center for Electronic Texts in the Humanities (Princeton/Rutgers)
Information at http://www.princeton.edu/~mccarty/humanist/

[1] From: Willard McCarty <mccarty@phoenix.princeton.edu> (16)
Subject: computers & society

[2] From: Stan Beeler <sbeeler@quarles.unbc.edu> (24)
Subject: Java

[3] From: "Dr. Joel Goldfield" <joel@funrsc.fairfield.edu> (10)
Subject: Re: 9.501 computing & social models

[4] From: "C. M. Sperberg-McQueen" (90)
Subject: In defense of Java (re: Humanist 9.501)

Date: Tue, 30 Jan 1996 18:49:17 -0500
From: Willard McCarty <mccarty@phoenix.princeton.edu>
Subject: computers & society

Martin Mueller, referring in Humanist 9.505 to my question about computers
and social models, asks if "there [is] finally an interesting story in all
this or is it just the old story of 'individual and society' projected on a
technological screen?" My problem is with his "or", which I would rather
replace with "and". Perhaps there's not much of a point to be made other
than the apparent fact that computing systems mirror social models, and so
our idea of who we are, but it does seem to me that this point needs to be
made. Is it not crucial that we understand this mirroring relationship? If
the computer were not playing such a prominent role in the industrialized
world these days, then perhaps the idea of humanity it manifests would be a
less urgent concern. It approaches, however, the power of Archimedes' lever,
which to me means that we need to pay attention.

As an aside to all this, I also find myself musing over the metaphor of
"projection". Is that what we do?


Date: Tue, 30 Jan 1996 17:58:59 -0800 (PST)
From: Stan Beeler <sbeeler@quarles.unbc.edu>
Subject: Java

I am currently assisting our Computer Science Department in
teaching a seminar on Java because I believe that the language has
something to offer to our Internet community. There has has been something
of a tempest in a teapot about the focus of Java. From what I can see it
would not be practical to transmit a full featured word processor across
the already overloaded bandwidth of the Internet every time someone wants
to write an essay. Of course there are those who tout the "Internet
computer" which will cost next to nothing and depend upon the net for
almost everything. I believe that this project will be doomed to the same
sort of failure as the "disk-less workstation." It is much more efficient
to provide people with PCs or Macs that are connected to a network by
Ethernet or SLIP /PPP. However, Java can provide an easy means to give
client systems small, custom applications quickly. For example, I can
envision a Java applet that would allow a professor to massage student
marks at the end of the term so that it sends back information to the
registrar in an acceptable format. (our registrar seems to change the
acceptable format regularly). I do not think that recent developments in
the Internet will lead to a return to the mainframe / terminal style of

Stan Beeler

|Dr. Stan Beeler English Programme University of Northern B.C. |
|email: sbeeler@quarles.unbc.edu | WWW: http://quarles.unbc.edu/ |

Date: Tue, 30 Jan 1996 22:37:36 -0500
From: "Dr. Joel Goldfield" <joel@funrsc.fairfield.edu>
Subject: Re: 9.501 computing & social models

Hmmm, intriguing questions. Some observations: just because one
values and chiefly uses the independent PC with occasional,
or more frequent connections into the WWW does not mean:
1) that one is a Republican;
2) that one is paranoid.

However, it could reflect a "fuzzy" tendency toward favored
privacy and self-sufficiency prized by some Transcendentalists we've

Joel Goldfield
Fairfield University

Date: Wed, 31 Jan 96 10:51:07 CST
From: "C. M. Sperberg-McQueen" <U35395%UICVM.BitNet@pucc.Princeton.EDU>
Subject: In defense of Java (re: Humanist 9.501)

On Mon, 29 Jan 1996 18:56:51 -0500 (EST) Willard McCarty said:
>Many here will know about Java, the programming language that
>offers extension of interactive networking to transmission of
>application programs. ...
> At the extreme end of the scheme, users would run Java
>workstations incapable of any other kind of computing. Institutions
>could then determine exactly what software its members used.
> ...
>The platform-independence of the Java scheme is obviously
>attractive, e.g. for computer-assisted teaching and learning. No
>question. Less clearly beneficial is the potential for central
>control. It would seem at least arguable that the independent
>computer has been perceived as a serious threat by some people,
>that the LAN offered one kind of comfort, but that the Java
>workstation offers much more. ...

I think I'm as eager as anyone to maintain the independence of the
individual desktop machine -- and as willing as anyone to note the ways
in which technology serves and shapes our social organization -- but it
seems to me that this view goes at Java from the wrong end.

Yes, Java could be used to run diskless workstations, but diskless
workstations have been on the market for a long time even without Java
to run them. They've always seemed to me somehow self-contradictory
machines, at the same time independent and tethered to their central
host. But while they haven't exactly driven workstations with disk
drives out of the marketplace, they are still around -- less, I think,
because the organizations which buy them want to exercise unlimited
power over their users than because they would like the PCs in the PC
labs in the university to remain functional and virus-free.

I've worked at sites where the PC labs were full of free-standing
machines with local disk drives, and at sites where the PCs lacked hard
disks and booted off a remote server. The free-standing machines needed
to have their disks reformatted at least weekly to cleanse them of
viruses and reinstall the publicly available software; because the task
was time-consuming, it didn't happen nearly often enough, and in a lab
of twenty machines it was not rare for one or two or three machines to
be out of service. The remote-booting machines make it easier to
upgrade publicly available software, to install special software for
classes, and to fight viruses. Our micros group occasionally runs a
virus scanner over the network-based C drives on which users can install
-- until the machine is next rebooted -- their own software from floppy
disks or the net; an appalling proportion -- half, as I remember it --
have viruses. Because the PCs in our labs boot from a central server,
however, those viruses go away when the machines are rebooted. (This
also means that the virus has gotten onto the temporary C: disk in a
very short time from our users' floppy disks or network behaviors.)

The difference between Java and older diskless workstations is that Java
will make it harder, not easier, to centralize control over the software
on a machine, even a diskless workstation. To install new software on a
diskless machine, I have to load it onto the machine's temporary drive
each time I boot. This isn't always quick or easy, particularly if I
have to compile the program. Java makes this much easier, since it's
designed for dynamic linking and network delivery of executable code,
and so I don't even have to install the software before saying I want to
run it. Java also makes it much safer to run downloaded software, since
it's designed to protect my machine -- whether a diskless workstation or
a standard PC or Unix box -- from viruses and other hostile software.
Since Java code can be loaded from anywhere on the net, fast enough for
interactive use, and since it removes much of the danger and fear of
using such downloaded software, I suspect it will lead to more, not
less, sharing of software across the network. The only way I can think
of for a site to control what software its users can run on their Java
workstations is to cut itself off entirely from the net, or to set its
local Java security at paranoiac levels. And, of course, users who
insist on self-booting free-standing machines have their own disks. On
the whole, I think Java is going to give much more freedom to the user
than it gives control to the central organization. Promiscuous pointing
and clicking will be much less risky with Java than promiscuous
downloading from bulletin boards without Java. (I can see the promos
now - "Java! For promiscuity without risk!")

Java will also make it much easier for small niche software products to
be developed and distributed. That should be of some interest to those
of us in academic life, since we do so much of our work in such niches.

Java will, I predict, make it much easier for a decentralized group to
collaborate on developing -- for example -- a set of inter-operable
tools for the kinds of textual analysis now performed mostly with
monolithic software like OCP, Tact, or Word Cruncher. Since I continue
to have hope that such collaborative development can become a reality, I
am rather cheered by the invention of Java, since it looks at first
glance (I've read the language spec but haven't got all the way through
the tutorial) like a much better tool for the job than anything else we
have available.

-C. M. Sperberg-McQueen
ACH / ACL / ALLC Text Encoding Initiative
University of Illinois at Chicago
u35395@uicvm.uic.edu / u35395@uicvm

All opinions expressed in this note (except those I have quoted with a
view to refuting them) are mine. They are not necessarily those of the
the University of Illinois, its board of trustees, its administration,
the State of Illinois or its legislature, nor of the Text Encoding
Initiative, its executive committee or other participants, its sponsors,
or its funders. Anyone who says otherwise is wrong.