LINGUIST List 4.37

Fri 22 Jan 1993

Disc: A Permanent Linguistics Server?

Editor for this issue: <>

Directory

1. Stephen P Spackman, Permanent Linguistics Server?

I think we need a permanent Linguistics Server to see to the needs of
the field. It could provide:

 An archive site serving the needs of linguistics, storing
 programmes, corpora, bibliographies and papers;
 Permanent electronic mail addresses not impacted by changes of
 employer or internal policy;
 A permanent home for such institutions as Linguist and the
 Linguists address server; and
 Perhaps even a platform for formal electronic publication, as
 standards appear and the notion becomes more attractive.

Is this a good idea? Is this the time? Does anybody want to do it?

 ~ * ~

Here's how I think we could handle some of the practicalities.

The most important thing that such a server would provide is a permanent
*name*. There are plenty of institutions in the world that can and do
find the resources to support us from time to time, but to guarantee a
project lifetime measured in decades, with any continuity of policy, is
very difficult. Even such a simple thing as keeping a single computer
permanently available for a given task is a lot to ask of a single
benefactor. Fortunately, the internet is so designed that names are by
and large independent of location; once an address is acquired, the
machine it designates can be moved or changed as necessary.

The address itself should be chosen carefully so as not to be
contentious; a surprising number of organisations seem to be upset by
the public suggestion that their employees have external national or
commercial affiliations; this would argue for an address in the
top-level .org domain.

>From a technical perspective, a permanent server might initially require
a computer of large workstation scale, a gigabyte or so of disk storage,
and, of course, its own top-level address. With luck, we might get most
of the hardware donated.

To run it, we would also need a room, a power supply, a good physical
network connection, reliable backups, and part-use of a system
administrator - in short, we would want to rely on the good graces of
some company or university to keep it fed and watered. With luck, we
might get this donated (perhaps the more easily so if the arrangement is
renewed from time to time).

In operation, it would need volunteers and (as we are now seeing with
Linguist) perhaps some paid staff to look after each of the services
that it provides, some being more editorial in function, some more
technical. One would hope that the sense of long-distance community that
services would acquire by sharing resources would somewhat reduce the
burden on these good souls.

All of this would ideally be done under the auspices of an international
scholarly or professional organisation, if an appropriate and willing
one could be found, if only for the benefits of name-recognition; though
I am not certain that one exists.

(Computer science has the benefit of a professional society that has
come to believe that it has international scope and a need to
demonstrate its commitment to pioneer this field. The Association for
Computing Machinery now has a (logical) site "acm.org" which provides
members with email forwarding services for a nominal fee (I believe that
for an understandably greater sum they will even provide 800-number
modem access for those without computers to call home, though I could be
wrong). They have recently begun serious exploration of the possibility
of publishing their main journals in electronic form.)

 ~ * ~

In the last year or so I have asked a number of people how they felt
about the notion, and response has been quite positive. A few
institutions have made guarded, informal, off-the-record reply that they
might be persuaded to help out with day to day operation of such a
machine.

I think that something like this is quite feasible. Is it a good idea?
How could it be better? Is there an organiser - perhaps one with the
right contacts - in the house?

stephen p spackman +49 681 302 5288(o) 5282(sec) stephenacm.org
 dfki / stuhlsatzenhausweg 3 / d-w-6600 saarbruecken 11 / germany