[IPOL discuss] Ipol articles class + an example

[rafael grompone von gioi]

Hi,

I prepared a new version of the IPOL LaTeX class, incorporating
most of Nicolas' suggestions. The class files and examples are here:

  http://dev.ipol.im/~jirafa/ipol-latex-format/

The main changes are the addition of the IPOL logo and
reference at the beginning, and the links color. There are
also three variants to cope with the problem of the number
of characters per line:

> In "The Elements of Typographic Style", Bringhurst writes: "Anything
> from 45 to 75 characters is widely-regarded as a satisfactory length
> of line for a single-column page set in a serifed text face in a text
> size. The 66-character line (counting both letters and spaces) is
> widely regarded as ideal." I think 66 characters per line may be too
> short and the current 100cpl too long, maybe 80 cpl is possible with
> more margins?

I am well aware of this rule, but the compromise in our case is not easy.
I think we should rule out the use of two columns, a classic solution,
because it is not comfortable on the web. Then, we can change
the margins and the paper size. The three variants proposed are:

1) The same margins as before in A4 paper. The full page is used and
this reduce the space wasted on the web page. The problem is that
there are about 100 cpl and it is not nice to read. The line length rule
is for continuous long reading, and IPOL articles are not to be
read as a novel. The reading quality of the LaTeX articles should be
far better than the current HTML version. So, the first option is to pay in
long lines this compromise.

2) The second option is to reduce the margins so as to get about 80 cpl.
The result is intermediate. The margins are better for a printed article,
but worse for a web page, and the lines (even if still long) are far better
than in option 1.

3) The last option is to use A5 paper. The length of the lines is optimal,
about 67 cpl. The page size may be even better than A4 for web reading.
The compromise is not so clear for printing. Printing two pages per paper
sheet produce a reasonable printed size without wasting paper. I think the
result is even better than the previous two options. However, it is an
uncommon paper size and it can confuse some people.

What do you think?

Please note that the commands defined by the LaTeX class changed
from the previous version. There is no longer a command to produce
the title; it is always generated after the IPOL logo and reference.
The user must only define the title and authors with a command.

> * What should we write in the "ipolAbstract" ? Why do we need an
>  "ipolCode" and "ipolSupp"? Can't this information be written in the
>  article body, like it is currently in the wiki/HTML articles?

This was a suggestion by Jean-Michel Morel to give to the code
and supplementary material a especial status, suggesting they
are not optional but obligatory parts, particular to IPOL.

> I am quite reluctant to publish PDFs provided by the authors, because
> it feels like publishing the compiled programs without having the
> source. We would be stuck to a format with very limited options for
> future evolution.

I think it is now clear that the initial submission should be on any
PDF format, as long as the editors and reviewers can read it.

I agree too that the final PDF should be generated in
IPOL server, so we have the sources, we can generate
different quality/file size versions, and eventually we could
generate new versions on the future. But there are two
important points to discuss:

1) The procedure.

I am quite sure that in many cases, going from the initial
PDF version to the one that can be compiled on IPOL servers
will not be straight forward. People sometimes uses particular
packages, and adapting the figures and tables to IPOL format
can take considerable time. I think this work must be done by
the authors. First using our LaTeX package and guidelines in their
own machines and then in our server. So we should provide an easy
way in which the authors can upload their zip/tgz to test as many
times as they need. Then, once they think the generated file is OK,
the editor should only check that the format is satisfied. But this
only implies checking the resulting PDF. Otherwise, it would be
too much work.

2) Future compilation of LaTeX files

I agree that it would be nice to have well behaved LaTeX
sources to be able to be re-compiled to new formats
in the future. Unfortunately, I don't think this is possible.
We know already how much work it is needed to produce
C programs that satisfy a standard and that probably
we will be able to compile in the future. The problem is
worst for LaTeX: there is no real standard and we need
to use many packages developed by many different
people and different quality level. Moreover, we can ask
the authors to do the standardization effort on the code
because it is the essence of IPOL, but there is no point
in asking the same effort for the LaTeX files. The PDF
version of the articles are the standard version we will
surely have in the future.

I don't think we can rely on the possibility of re-compiling
the articles in the future. As in any other journal, as format
evolve, old and new articles will co-exist having different formats.

Comments and suggestions are very appreciated.

Best regards,
rafael