I think you are referring to the issue of certain protein structures (in
the BNL PDB mirror of the course) not invoking Rasmol when they are
clicked on.
As far as I'm aware, the WWW server at Brookhaven PDB does not stamp PDB
files with the chemical MIME type, which makes sense because in most
cases, people presumably access the PDB files at BNL to download and
store them, not to look at them with Rasmol. In the course material, when
we point to a URL of an actual protein structure, we give the absolute
URL of the copy of the file in Birkbeck Crystallography Dept's PDB mirror.
Our WWW server delivers .pdb files as chemical MIME.
But there are a few cases of .pdb files which are not in the PDB (e.g.
example alpha helices and beta strands in the Protein Geometry section),
which are therefore stored as local copies in the course hypertree- and
relative URLs were used to reference them, so that they are served by
whatever server you are accessing the course pages from- so if this is
BNL PDB, they dont invoke Rasmol.
There are a number of solutions-
1) Change these relative URLs to absolute. However, this means that the
pdb file must be delivered from Birkbeck, which is not ideal if accessing
the pages from outside Europe. This problem also applies to the various
PDB structures pointed to in the rest of the material, some of which are
large files of course.
2) Have Brookhaven PDB server stamp .pdb files with the chemical MIME
type (and then use relative URLs). I was thinking that this is probably
not a reasonable request, ie having their WWW congifuration determined by
the PPS course.
3) Use a different extension for the PDB files, e.g. '.ent' , '.pps' etc,
which is not currently present on any files on the Brookhaven Web server,
and ask Dave Stampf to configure the server to deliver these with the
chemical MIME type (and use relative URLs). But my gut feeling is that
filename extensions are quite important, and preferably, the '.pdb' extension
should be used for PDB files.
Thanks for raising this issue- I would be interested to hear people's
opinions on this.
John Walshaw