[Table of Contents]


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Bibliography



Richard Gwyn Thomas wrote:
>
> This joke is getting a bit tiresome and I doubt if most Ozcons
> subscribers
> understand it.

Oh well, as one of those guilty, please allow me to explain...

LaTeX is a document preparation system with very strong support for
mathematics. It's popular in the academic world in areas where severe
maths is likely: mathematics, physics, computer science, etc.

LaTeX is a "mark up" system, not a word processor. What you actually
give it is lumps of writing with information about how that writing is
to be presented; this is a new section, lay this out as a quote, some
poetry goes here, etc. LaTeX lays it out for you in a manner (almost)
guaranteed to be pleasing to the eye. You can adjust style parameters to
lay the result out in whatever format the journal you are submitting to
requires[1]. There are all kinds of extensions to LaTeX/TeX for anything
from typesetting music to drawing Feynman diagrams.

BibTeX is the bibliographic database format used by LaTeX. It follows
the LaTeX philosophy in that what you give it is detailed information
about the things you want to cite and a style guide and then it inserts
citations into the text and builds a bibliography in the end, laying out
the bibliography in the required style.

For more on BibTeX, point your browser to
http://www.ecst.csuchico.edu/~jacobsd/bib/formats/bibtex.html or
http://hypatia.dcs.qmw.ac.uk/html/bibliography.html

(The URL http://www.ecst.csuchico.edu/~jacobsd/bib/formats contains
quite an extensive list of bibliographic formats, if anyone is
interested.)

Unicode is a semi-proposed/semi-standard replacement for the standard
7-bit ASCII character set; the one that this message was sent in. ASCII
is fine as long as you speak English and only English. Accented roman
characters and other alphabets do not get a look in. Unicode is
cunningly designed to fit the worlds major alphabets (and quite a lot of
the minor ones) into a 16-bit wide character. The Java applets you see
while surfing the net use Unicode.

For more on Unicode, try http://www.unicode.org/

"Why is all this relevant?" I hear you ask/scream. Admittedly, this is a
computer scientist inflicting his stuff on innocent conservators, but
what you are talking about here is data manipulation, and I have a few
opinions ...

Cross Platform Availability. Most of the conservators that I know of use
Windows PCs. However, there are probably many Macintosh users and, also,
reprobates like DH who use Unix. Trying to agree on a standard program
for bibliographies is probably doomed to failure, as it will not be
available on all platforms. IMHO, all that can be reasonably asked for
is a common data format.

Plain text/Binary. Data can be stored in binary form or as formatted
text. Binary storage is often more efficient in terms of space and
speed, but text data can be read by both humans and a wide range of
programs. Text processing programs are not a big feature of
Windows/Macs, but Unix users have a huge range of programs that allow
them to manipulate data in this format and do all kinds of funky things
with it. These abilities are put to good use by Web CGI servers. The fly
in the ointment here is that text, at the moment, means 7-bit ASCII;
Unicode looks like binary data to most text processing programs and
systems like LaTeX perform all kinds of gymnastics to represent accented
characters.

Flexibility. Ordinary database systems, and the export format of
EndNote, tend to be pretty inflexible. In contrast, BibTeX or the
Uniform Resource Citation system tend to allow extension as new kinds of
data appear. Eg. I doubt whether a bibliographic database designed in
1985 would have thought to include a field for the URL of a Web
document. Along another axis, formats such as the USMARC library system
are designed to handle a lot more than just books and papers.

Searchability. Most bibliographic programs provide some sort of search
facility. If a common bibliographic format is agreed upon, searchable
indexes can be pooled. BibTeX format is not readily searchable, although
there are plenty of programs which maintain databases that can produce
data in BibTeX format. Neither is the EndNote export format, although I
would imagine that EndNote must be itself.

Databases. EndNote advertises the ability to search the Web for
citations. I don't know how it does this successfully, but I wonder
whether it produces a decent signal to noise ratio. What David seems to
be implicitly proposing is the creation of a bibliographic database for
conservation information, which people can contribute to. This database
could be maintained as a centralised Web server[2] or as an exchange
protocol by which people exchange new bibliographic information. The
exchange protocol is more democratic and gives access to people without
Web access, but it is fraught with data pollution problems. The
centralised database tends not to have data quality problems, but it
requires a sponsor and administrator.

[1] And if you are rejected, you can just adjust the style parameters
and submit it to another journal :-).

[2] And no, I am not volunteering to set this up.
--
Doug Palmer   Applied Financial Services
WWW: http://users.orac.net.au/~doug
mail: doug@afs.net.au, dougal@acm.org






[Subject index] [Index for current month] [Table of Contents]