to all WWW sites

ewan ewan
Thu Mar 23 05:34:16 EST 1995


[ Article crossposted from bionet.software ]
[ Posted on Wed, 22 Mar 1995 21:56:03 GMT ]


(I suspect that my original posting got lost somewhere during a
hiccup between EBI and the rest of the world... so I am reposting...)


There are numerous WWW/gopher sites for biologists out on the Net,
ranging from culture collections through to genome data and
sequence information. In some cases systems such as SRS are ideal
at interlinking data, but there are many databases (eg bionet
news articles) which are not as suitable for this sort of integrated
approach.

Despite this plethora of databases it is usually quite hard to 
know what sort of databases are there, where they are and
what sort of search terms they take. I had always liked the CUSI
interface to internet search engines made by Martijin Koster and
started to build (mainly for myself) a bioCUSI. I then advertised
it briefly and got some bad response from some people coupled
by alot of rather confused requests to put up server xxx - usually
accommpanied by some half formed HTML.

It looked like a pretty horrible task trying to figure all this out
(with not much hubris coming back at me...) - so I decided to make
a different system in which everyone could add their own search engine.

It was pretty clear that each search engine needed its own help
section. 'Cause of the wonders of the web, all you really
needed was a link... ie a short piece of valid HTML.

And so: bioCUSI(2) has been developed. It is at

http://www.ocms.ox.ac.uk/~birney/cusi/biocusi.html

and the form to add to it is at

http://www.ocms.ox.ac.uk/~birney/cusi/tish.html

[there is one - rather large - caveat in that this form
works with netscape but not mosaic. I am trying to figure out
which part of the CGI stuff I have missed out]

bioCUSI has a few advantages over the (far superior) search systems
such as SRS in that many different databases can be put into one
form and the little piece of perl can be run on anything which has
a server and a perl script (ie most UNIX boxes). SRS is a full
database system in which a whole series of interlinked databases
can be managed and queried: bioCUSI is a little form which redirects
your query to another database. They are not really trying to do
the same thing. 



What would be ideal would be if the manager of each site added
his or her URL. But... I guess we wont get everything, and
in about 2 weeks time I would like to invite anyone to add to
the system and I will try to organise the whole thing properely.

If all this is successful, then I will make a little tar file for
people to install the whole thing locally if they like. You can
always use the server we have here: it will just be (much) quicker 
to run your own.



ewan

birney at molbiol.ox.ac.uk



More information about the Bio-www mailing list