Actually, it might be easier to just write some little code that can
automatically create the abbreviations when given a full title? Am not
very good at that sort of thing …
Bruce
Actually, it might be easier to just write some little code that can
automatically create the abbreviations when given a full title? Am not
very good at that sort of thing …
I did some googling on this subject, and the results weren’t pretty:
The appearance of the ISO standard for abbreviating journal titles was not
greeted by universal acceptance. It would appear that even today, some major
journal indexes/databases do not follow this ISO standard. For example,
although Medline (Index Medicus) and INSPEC (Science Abstracts) have
established a high degree of conformance to this ISO standard, other major
tools like SciFinder Scholar (Chemical Abstracts) or Biosis (Biological
Abstracts) have not.
There’s a list of ISO abbreviations (not in XML) at
ftp://nlmpubs.nlm.nih.gov/online/journals/lsiweb.pdf. I’d expect with some
searching around on the NLM site you might be able to find some XML. A
regular expression might also work. The NLM site notes at
http://www.nlm.nih.gov/pubs/factsheets/constructitle.html that:
NLM establishes title abbreviations based on the rules outlined in Information
and Documentation, Rules for the Abbreviation of Title Words and Titles of
Publications (ISO 4, 3rd ed. 1997) and American National Standard for
Information Sciences, Abbreviation of Titles for Publications (ANSI
Z39.5–1985). Where the ISO standard instructions conflict with the ANSI
standard, NLM generally prefers the ANSI standard, despite the fact that this
is no longer a current U.S. standard. In some instances, NLM’s rules for
title abbreviations differ from both existing standards.
The list of title abbreviations they use comes from the ISSN International
Centre, and would apparently cost 111 euros for those of us residing in
industrialized nations. I’m not having much luck finding a copy of either
the ISO or ANSI standards online. While it would certainly be nice to handle
journal abbreviations automatically, we may need to rely on user input here.
Simon
Yeah. I suppose one way is to do it in code, but then allow a user to
confirm or override. I just pinged Alf Eaton about the NLM stuff (he’s
the author of HubMed).
A problem is that most users (me included) really have no clue how to
abbreviate this stuff!
Am asking about this now because I’m trying to work out the conventions
for the id an link values on the CSL files. Right now, they are the
same, using an abbreviated title for the name.
The user issue matters when creating styles, BTW. Would like a little
script or other interface where I can do:
createstyle
… be prompted for:
full title
based on (APA, etc.)
… and then it would automatically create the file, with proper id,
link, and file name of course.
Bruce
FYI from Alf …Begin forwarded message:
Hi,
The list of title abbreviations they use comes from the ISSN International
Centre, and would apparently cost 111 euros for those of us residing in
industrialized nations. I’m not having much luck finding a copy of either
the ISO or ANSI standards online.
The “List of Title Word Abbreviations” (http://www.issn.org/en/node/88)
seems to be available online in human-readable form:
http://www.issn.org/en/node/344
As far as I can see, the machine-readable form is not available, though.
With a bit of work, source data could be extracted from the
human-readable HTML files but, obviously, it would be better if this
could be avoided…
Regarding simple online lists of journal abbreviations, I’m aware of the
following ones:
Bioscience - Journal Name Abbreviation, ISSN Number and Coverage:
http://www.bioscience.org/atlases/jourabbr/list.htm
ISI Journal Abbreviations Index:
http://library.caltech.edu/reference/abbreviations/
Thomson Scientific Journal List:
http://www.isinet.com/cgi-bin/jrnlst/jlresults.cgi?PC=D
Matthias
It might not be a bad idea to try to build up a free, machine-readable,
list? It’d probably be smarter to standardize the logic than the list
of journal abbreviations per se.
I just put the list I started with in a YAML file …
american: Am
annals: An
association: Assoc
british: Br
geographers: Geog
geography: Geog
historical: Hist
institute: Inst
journal: J
political: Pol
review: Rev
sociology: Soc
transactions: Trans
… and then the script loads it:
require ‘yaml’
class String
IGNORE = [“of”, “the”]
TERMS = YAML::load(File.open(“abbrev.yaml”))
def abbreviate(form=“display”)
words = self.split(/\W+/)
abbreviated_words = []
words.each do |word|
if IGNORE.find{|w| w.downcase == word.downcase} then
nil
elsif TERMS[word.downcase] then
abbreviated_words << TERMS[word.downcase]
else
abbreviated_words << word
end
end
if form == “filename” then
abbreviated_words.join("_").downcase + “.csl"
else
abbreviated_words.join(” ")
end
end
end
Note: ideally the file of abbreviations would have roots as keys, and
the script could therefore know that both “geography” and "geographer"
become “geog” … but I don’t know how to do that.
Bruce
Is this just for the names of the csl files? Or are you trying to
build this logic into bibliographic output. Personally I find the
abbreviations incredibly obtuse, and have trouble seeing them as
anything other than a hangover from earlier, less memory profligate,
days 
Would you mind reminding me of the purpose of this? Sorry for
falling behind!
–J
Is this just for the names of the csl files? Or are you trying to
build this logic into bibliographic output. Personally I find the
abbreviations incredibly obtuse, and have trouble seeing them as
anything other than a hangover from earlier, less memory profligate,
days
Agreed.
Would you mind reminding me of the purpose of this?
The immediate goals was to generate smart CSL file names and IDs/URIs.
But software often has to this in some form too, so was just wondering
if we could kill a couple birds with one stone.
Of course, right now it’s no real worry; can just manually generate the
IDs/file names.
Bruce