LATEX-L Archives

Mailing list for the LaTeX3 project


Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Phillip Helbig <[log in to unmask]>
Reply To:
Mailing list for the LaTeX3 project <[log in to unmask]>
Fri, 31 Oct 1997 07:59:23 GMT
text/plain (109 lines)
>  > Although I guess there is no reason to REQUIRE a journal to support
>  > bibliographies generated by BibTeX, it seems a sensible thing to do,
> i don't quite understand this. how can a style know whether a
> bibliography was formatted by BibTeX or by hand?

What I meant was not requirement for support of bibliographies generated
by BibTeX which conform to some bad scheme favoured by the journal, but
support for doing things as they should be done, coding content and not
presentation.  BibTeX is the most sensible way to do this.  A standard
THE NATIVE NATBIB FORMAT TO BE USED HERE, and thus this should be done.
That is, if someone wants to explicitly code the bibliography and use
the standard cite commands, then that's fine (though I would not
recommend it) but, if one goes beyond this, then efforts should
concentrate on natbib for citing and the native natbib format for the
list of references, preferable via a .bst file generated with BibTeX.
That is, I don't want a journal to REQUIRE me or even SUGGEST TO me to
use stuff like \shortcite in the text and

   \bibitem[\protect\citename{Jones et al., }1990]{jon90}

which, as Daly points out in the natbib documentation, `is not a good

You can sometimes tell if a bibliography has been formatted by hand,
since mistakes, if they exist, tend to be inconsistent, while BibTeX, if
set up correctly, will get everything right with no effort.

Now to the following points.

> do you mean one or
> all of:
>  - journals must supply a .bst file on demand

A very good idea (assuming natbib can read it, as the USE of natbib
should be explicitly ALLOWED).

>  - journals should strongly recommend that authors use \cite


>  - authors should supply only a .bib file

Supplying a .bib file is a bad idea; either one has a separate one for
each paper, conflicting with the goal of BibTeX, or one sends in a lot
if irrelevant (and potentially embarrassing:) information to the

>  > itself not to be journal dependent, some sort of extended citing scheme,
>  > covering all the possibilities, should be at least allowed if not
>  > required.  BibTeX seems to cover enough ground, and the formats, fields

> i think you are confusing two issues here. layout of the contents of
> the \bibitem (by BibTeX) is one thing, the interaction between \*cite*
> and the label of the \bibitem is another. i dont think the internal
> formattng is much of an issue (unless someone proposes structured
> markup inside a \bibitem), and i would suggest that standardizing on
> the extended \cite markup of `natbib' would suit almost everyone.

They are two issues, but related.  Standardising on the extended \cite
markup of natbib is perhaps more important (since it can read \bibitems
genereated with a variety of .bst files (or hand-coded correspondingly).
Distributing standard .bst files with the .cls would make sure that the
reference list corresponds more closely to what the journal intends, and
the sooner everyone realises that natbib's native format is the best and
most flexible, the better.  It wouldn't be a good idea to require natbib
to support all kinds of bizarre \bibitem formats indefinitely.

> I have had to do styles and editorial support for quite a few journals
> here, and since i started telling people to use natbib, i have not had
> any more complaints of `but i cant express this bizarre citation
> situation'. really, anyone who has a scheme they cannot express should
> tell Patrick, and ask him to support it.

As far as citing goes, yes, but the reference lists should, as soon as
possible, use the native natbib format and not the other, inferior

> so far as bibliographies is concerned, do you mean that you want core
> LaTeX to define extended \cite commands? well, we know that won't
> happen for LaTeX2e, so our only recourse is a standardized package. i
> personally would like to see the problem solved by moving natbib into
> the macros/latex/packages area of CTAN, so as to give it the
> `preferred supplier' status.

My thoughts exactly.

>  > before that we must agree on what information is needed, how it is to be
>  > grouped, what the syntax should be and develop a sample (or real) .cls,
>  > .bst etc which shows that the scheme can actually fulfill its purpose.
> i'd suggest doing one bit at a time. get front matter right first, and
> produce a sample package that implements it. if you try and do the
> whole job in one go, it'll fall to pieces.

I am now convinced that citations and references have already been
solved by natbib and custom-bib.  The other non-front matter stuff is
relatively simple, though there really should be standard macros for
stuff which is often in different styles, such as \textit{et al.}.

OK, shift the emphasis back to front matter.

Phillip Helbig                          Email ... [log in to unmask]
Nuffield Radio Astronomy Laboratories   Tel. ..... +44 1477 571 321 (ext. 297)
Jodrell Bank                            Fax ................. +44 1477 571 618
Macclesfield                            Telex ................. 36149 JODREL G
UK-Cheshire SK11 9DL                    Web ....