Standards (chemical/* MIME)

peter Murray-rust (p.murray-rust@mail.cryst.bbk.ac.uk)
Tue, 30 Jan 1996 13:08:09 +0000 (GMT)

On Tue, 30 Jan 1996, Iddo Friedberg wrote:

>
> Probably MIME. Rasmol/PDB files are MIMEd differently from different
> servers (that's the nice thing about standards - there are so many of
> them :-).

This isn't quite accurate :-). When Henry Rzepa and I put forward the
chemical/* MIME standard we proposed a single type for PDB files:
chemical/pdb. There was no toplevel type chemical/* so we wrote it as an
INternet Draft, hoping that it would be constructively criticised and we
could then modify and resubmit.

Two things happened:
(a) the IETF (Internet Engineering Task Force) had many different
views on what sort of new toplevel MIME types should be admitted - the
common result was that chemical/ did not gain general approval. It is
generally agreed that the MIME standard does not have enough flexibility
built-in. There is some light at the end - see the discussion group
chemime@ic.ac.uk. It is probable that there *will* be an approved
namespace for chemical/, but we can't say what at present.
(b) the molecular community started implementing chemical/ MIME
immediately. There were a number of variants with differing numbers of
"x-" in (anything from x-chemical/x-pdb to chemical/pdb). This
illustrates the problem of standards - they are critical to efficient
working. many people regard them as boring, but they are an integral
part of bioinformatics which is why some of us are trying to get file
formats standardised, etc .
We effectively have a 'de facto' standard as chemical/x-pdb for
PDB files and if there are servers which diospense something else, it may
be worth suggesting that they change. The definitive opinions can be
found on the chemime list and the server set up by Scott Nelson at
Lawrence LIvermore to serve definitive MIME types. If your browser won't
deal with Scott's files, you may need to reconfigure it.

>
> Use two MIME types:
> application/x-rasmol

This is different and should not be used for the transfer of PDB files.
It represents a non-standard application (anyone can use x- types) and is
currently used for passing rasmol scripts (unless things have changed).

> chemical/x-pdb
>
> Some old PDB files have the ".ent" suffix. Add that to your configuration
> as well.

Suffices on remote servers are irrelevant to the client. If the remote
server stamps them as chemical/x-pdb it doesn't matter what the suffix
is. (It matters to the *server* maintainer who stamps files according to
suffix).

If you have a LOCAL file which you view with a WWW browser it looks in
you LOCAL .mime.types file (UNIX - there are PC/Mac equivalents) to
determine what the MIME type is. If you have LOCAL files called *.ent, I
suggest you cahnge them to *.pdb if possible.

It's great to see this sort of discussion :-)

PeterMR

Peter Murray-Rust, Glaxo Research & Dev. (pmr1716@ggr.co.uk); (BioMOO: PeterMR)
Birkbeck College, ubcg09q@cryst.bbk.ac.uk, CBMT/Daresbury mbglx@seqnet.dl.ac.uk
http://www.cryst.bbk.ac.uk/PPS/index.html, http://www.dl.ac.uk/CBMT/HOME.html