LINGUIST List 3.602

Tue 21 Jul 1992

Disc: Citing LINGUIST

Editor for this issue: <>

Directory

1. Claudia Brugman, Re: 3.598 Citing LINGUIST
2. Paul Saka, RE 3.598 citations
3. Stephen P Spackman, Re: 3.598 Citing LINGUIST

Message 1: Re: 3.598 Citing LINGUIST

About PSYCOLOQUY (sic), mentioned by Philip Swann, 3.598. This is an
electronic JOURNAL--its contents are refereed, subject to editorial and
stylistic constraints, etc. Its citation procedures might not be relevant
to something that is, and should in my opinion remain, an informal forum
for discussion, as LINGUIST is.

Claudia Brugman
brugmancrl.ucsd.edu

	I agree with David Stampe that in citing LINGUIST the streamlined
format is appropriate. Not only is it more consistent with the style of
traditional print citations, but as a practical matter the wordier format
contains unnecessary information. I would think that virtually every
professional linguist (at least in the technological countries) would
either know about LINGUIST directly or else know someone who does.

Message 3: Re: 3.598 Citing LINGUIST

David Stampe asks, in essence, why I believe it is useful to supplement
traditional bibliographic information with electronic addresses, given that
these are so ephemeral. In the case of mailservers and such entities, I believe
that the situation is fully equivalent to that of a minor publisher where many
style guides DO suggest that one supply additional information to help in their
identification - typically a city if not a street address. Clearly, however,
listing a city for Linguist is inappropriate; its address (perhaps not forever,
but then a journal may not stay at the University of Chicago Press forever, and
books have been known to change publisher between editions) really IS at
tamvm1. The argument for listing an archive site is weaker, but motivated by
the observation that many (to be frank, almost all) libraries are proving very
slow to adapt to the new technology; they are often not equipped to retrieve
information from electronic archives, let alone interested in providing stable
local spools. The reason it suffices to give the title of _Language_ in a
citation is that that title suffices - in conjunction with a run of the mill
academic library - to locate the original source. _Linguist_ does not have that
same social prominence, and even with much wider use and greater stability of
services like archie and wais we will have to wait another decade or three
before librarians _universally_ take them seriously.

Worse yet, of course, is the problem - increasingly common in computer science,
at least (my thesis refers to several such sources) - of technical reports
published _only_ electronically. Here the archive site itself is the primary
source, and the finding of a suitable method for its description becomes
essential. One possible future would see, for instance, a standard convention
of keeping a continually updated standard reference for each of linguistic
theories brands X, Y and Z on the main ftp servers of their proper
universities....

As to email addresses for authors, this clearly IS a departure from standard
practise, but in my experience a very useful one: in computer science, many
authors list their email addresses on papers that they write, and it has proven
well worth the effort of transcribing this information into my private
bibliography file. One might disagree, but it seems to me better to pass the
information along than not, once it is handy.

And while it is true that personal electronic mail addresses are notoriously
subject to change *this tendency needs to be fought*. Many institutions
consider it reasonable to continue to forward physical mail for many years
after their employees move on; since electronic mail is generally more
important than physical mail and is certainly easier to forward, I feel it
should simply be expected that our institutions extend the same service into
the electronic domain. Better yet, of course, would be if societies could
provide us with centralised stable addresses - "stephen.spackmanacm.soc" and
"stephen.spackmanlsa.soc" are addresses that, if only they existed, could last
me the rest of my life!

Finally, yes, it is true that the correct spellings of ftp (not as a programme,
but as the File Transfer Protocol itself) and ascii transfer mode (after the
American Standard Code for Information Interchange - whose "real" name is now
something more like ISO 8859-2, not that I'm remembering the number correctly)
are FTP and ASCII. My error.

Richard Goerwitz asks if such citations are actually useful in practise. I can
only reply that while we are again here talking primarily of computer science
rather than linguistics, during my stay at the University of Chicago I
retrieved at least many dozens and perhaps as many as a couple of hundred
papers and technical reports by ftp or mailservers, and a large proportion of
these were indeed located through citations that gave ftp server addresses. The
quality of information located this way tends to be comparatively low - this is
a very informal, low-overhead publication technique, after all - but the amount
of detail in online documents is usually high in comparison to more
conventionally published documents, and when evaluating research results or
combing for technical methodology this is often exactly what is wanted.