link soup

[Mark Smith]

@ 01:10 on 22.8.05, Yuri T. wrote:


>Proper bibliographic references are more than footnotes.


Yes.


>First, there would need to be a way to support several styles of

>references. 


There is scope for this. Its not perfect. I don't think it ever will be
in a simple writing syntax like Markdown.


>Second, for each style there are several separate cases depending on

>how the reference is used in the text:


Yes. Its not possible to replicate every cite style. Conceded. This is
not the point at which I give up and say "it won't work" though. Online
papers *must* (or at least should) be different from printed papers in
some respects. The bracketed, comma separated list of refs is not a good
model IMO for online papers.


>Note that while the numbered style can be roughly approximated by

>footnotes, Harvard looks altogether different.  


Yes, but you could style this, if you wanted to.



>And it's really not >the case that the numbered style is becoming more

popular - it depends >on the field, and I don't know of any that
recently moved from Harvard >to numbered.

There does not require to be any journals going from a non-numbered
style to numbered styles. New journals adopting numbered will increase
the popularity of numbered alone. But this is a side issue.


>Finally, with a proper bibliography >solution one should be able to use

a bibliographic database a la >bibtex.

I agree that something like this would be nice. BibTeX (bibtex is the
processor), though popular and available, would not seem to make the
besy sense. Something built on the Xbiblio project could be
advantageous.


>This doesn't make much sense for footnotes.


One of us is not thinking this through. (It could be me.) It seems to
me, that even with a nice 3rd party solution for managing your refs and
building your reference list, you are going to need a Markdown syntax. I
suppose it possible that a bibliography app could "do it all", but it
seems extremely complicated and prone to error. This is why I think
using footnote syntax is likely to be the way to go.

This is *not* a solution for producing article for submitting to
traditional journals. (They don't accept Markdown input of (usually)
(X)HTML. There is a subset that accepts LaTeX, of the remainder, the
vast majority want Word. I am not thinking about using Markdown (much)
for writing papers for submission to the dinosaurs.

Markdown *could* be a solution for instant independent academic
publishing. This seems to me to be the inevitable future.


>I think what's causing some confusion is that there are fields like

>law that actually use footnotes for bibliographic references and that

>many of the computer science publications use numbered references that

>do _look_ like footnotes.


Not in my case. I'm looking at bio/medical/engineering sciences. I
appreciate the conventional distinctions and acknowledge your points. I
just think we should drop most of the distinctions for web output - some
because we will have to, some because it makes sense.

OK. I've used up a lot of bandwidth (sorry) and I think I've made all my
points. So, I'll bow out now, unless somebody wants more info.

mark.