LINGUIST List 3.598

Tue 21 Jul 1992

Disc: Citing LINGUIST

Editor for this issue: <>

Directory

1. David Stampe, 3.594 Citing LINGUIST
2. "Richard L. Goerwitz", Re: 3.594 Citing LINGUIST
3. Swann Philip, 3.593 Citing LINGUIST

Re the form of citations, Stephen Spackman suggests:

 Christine Kamprath <LIFY460orange.cc.utexas.edu>. "Re: 3.562 Accents:
 LINGUIST in the news". In Linguist List 3.575 (July 1992)
 <linguisttamvm1.tamu.edu>. Archive ftp (ascii)
 anonymouslinguistics.archive.umich.edu:
 linguistics/linguist.list/volume.3/no.551-600 .

But most of the information here goes beyond traditional bibliographic
citations, and is actually quite ephemeral:

 * The author's address, for one, is not traditionally given, because
 it is likely to change. This is even more true of email addresses.
 The place for the address of an individual is a subscriber list,
 name server, etc. (Unfortunately some mailers still do not supply
 the author's name, and in that case, unless the posting is signed,
 the email address is the only name available. But in the case of a
 digest like LINGUIST, the editors should simply require signatures.)

 * Including the posting address for LINGUIST is like including the
 address for submitting manuscripts to a journal. Including the
 subscription address also seems to me superfluous. The question
 that the citation is trying to answer is where to find the posting
 (see the next item). Again, these addresses are ephemeral. They
 are best obtained from announcements, programs that list mailing
 lists, etc. (And on those, Bitnet as well as Internet addresses
 should be included.)

 * Listing an archive address is convenient but we don't typically
 include booksellers' ordering instructions. Archives come and go,
 means of access change, and certainly directory pathnames change.
 Many sites should have local archives. The place for instructions
 on where to get copies is (1) local archive directories, (2) files
 of instructions or faq (frequently-asked-questions) files posted by
 the editors, (3) archive information programs like archie. (The
 latter require files or parent directories to have a unique name.
 If one were settled on, *that* would be useful in a citation.)

This leaves the following information:

 Christine Kamprath. "Re: 3.562 Accents: LINGUIST in the news".
 In Linguist List 3.575 (July 1992).

This information is entirely available from the original posting. But
as noted above, ideally it should also include somewhere the unique
name that archives *should* use for the parent directory of archived
postings. If we followed the Michigan example, where the pathname is
linguistics/linguist.list/volume.3/no.551-600, then the relevant name
for programs like archie would be "linguist.list".

The weakest part of such citations is the "title", quoted from the
Subject line, which often provides no information whatever about the
actual content of the posting. But as a student recently pointed out
to me, this problem is not limited to titles of postings: he had found
a dissertation on linguistics entitled "How I spent my summer vacation".

How the information should be formatted depends on the citation style
of the paper or electronic journal in question. (I don't think the
LINGUIST editors have opted for any particular style.) I hope that
eventually we'll replace printed style citations with logical markup,
but just now even the best markup systems leave a lot to be desired in
clarity, portability, etc. (However, a discussion of whether the
Language style, with titles not internally capitalized, is likely to be
compatible with electronic systems of referencing, is long overdue!)

Re the "value" of postings to mailing-lists (or for that matter to
Usenet newsgroups) for any particular purpose, surely that must depend
on the intrinsic value of the posting, not the manner or place of its
publication. Judging books by their covers, or articles by their
journals, is as silly as judging grammars by counting features.

David Stampe <stampeuhunix.uhcc.hawaii.edu>, <stampeuhunix.bitnet>
Dept. of Linguistics, Univ. of Hawaii/Manoa, Honolulu HI 96822

[Moderators' Note: David Stampe is right that the citation-style
proposed earlier is inconsistent with the bibliographic style used
elsewhere. However, there are practical reasons to depart from
tradition with electronic media. The aim of a citation is to make
the source used by an author accessible to readers. A citation which
refers only to a list-name and the relevant issue fails in this, for
there are no standard ways of locating electronic sources. To locate a
book or a journal article we simply go to the library. In difficult cases,
we consult a librarian. But seekers of electronic sources are likely to
be searching on their own, from a personal computer. And there are
no standard reference works which will tell them where to look.
So it seems only reasonable to include the place where the archives
of a source are to be found.

One further point: the true "name" of LINGUIST is LINGUISTTAMVM1,
and all networked machines running Listserv software know it by
this name. This also happens to be its Bitnet address.
It's best to refer to it this way, not only for the convenience
of Listservs, but also because a list has no exclusive right to its
"first name." Another Internet list could call itself Linguist too.

Anthony & Helen]

Message 2: Re: 3.594 Citing LINGUIST

Stephen Spackman suggests the following citation format for our group:

 Christine Kamprath <LIFY460orange.cc.utexas.edu>. "Re: 3.562 Accents:
 LINGUIST in the news". In Linguist List 3.575 (July 1992)
 <linguisttamvm1.tamu.edu>. Archive ftp (ascii)
 anonymouslinguistics.archive.umich.edu:
 linguistics/linguist.list/volume.3/no.551-600 .

Two questions, Stephen. How can we work into this format a more explicit
indication of the source's nature (i.e. that it is an electronic discussion
forum)? Also, are e-mail addresses really useful information? Machines
die, and domain configurations change. Is the e-mail address really going
to be worth the space?

Other questions come to mind, now that I'm thinking about the matter. Does
anyone else find the above format to be verbose? Note: not everyone snarfs
files using a program called "ftp." And do we actually have assurances from
umich.edu that the archives will be online for at least, say, five years
or so? The archive site is not important unless we have a semi-permanent
address.

Nit: capitalize ASCII. If FTP is meant in its generic sense, then capital-
ize this as well.

Has anyone actually made use of references of this type? If so, what was
the context?

 -Richard L. Goerwitz goer%midwayuchicago.bitnet
 goermidway.uchicago.edu rutgers!oddjob!ellis!goer

Stefan Harnad runs a list called PSYCHOLOQUY (or some such spelling!).
This is rather more formal than LINGUIST and most of the issues
regarding citation and so on have been carefully resolved in a direction
that seems pretty good to me.