On Mon, 31 Aug 1998 07:51:03 -0400, Clinton Jones <clinton_at_TTALK.COM> wrote:
>I think that some journals,
>particularly 'reprinted' ones will always have to be provided in a PDF
>format, primarily because redoing them in HTML will be too costly and too
>time consuming. I am afraid that PDF's will be around for a long time to
>come but I agree that all new documents should be provided in a HTML
>format and not relatively inflexible PDF.
This posting reminds me that subscribers to this list might wish to
read another article in the current American Scientist: Brian Hayes's
Computing Science column, titled "Bit Rot"
(online at
http://www.amsci.org/amsci/issues/Comsci98/compsci1998-09.html,
and also as PDF and Postscript files).
The column points out how poorly formats such as PDF preserve knowledge for
the future. Electronic presentation of publications is now in a stage where
one must make hard choices between display/print quality and broad access
(offered by PDF), author-interface advantages and ease of self-publishing
(as with TEX's handling of mathematics) and archiving/user considerations
(optimized by markup languages such as SGML, HTML and XML). The archiving
questions are the stickiest: How do you make a document so that it will be
useful a decade from now?
As Tom Walker and Clinton Jones point out, PDF is what works most easily
today if you sidestep archiving concerns. I'd say further that it's the
obvious short-term (cheap!) solution in all disciplines where authors are
unlikely to be able to converge on a common software for article submission.
Rosalind Reid
Editor, American Scientist
Received on Tue Aug 25 1998 - 19:17:43 BST