Re: CML RSS

At 13:21 24/05/2007, Daniel Zaharevitz wrote:

> On May 24, 2007, at 7:38 AM, Rzepa, Henry wrote:

Had a chance to play with the latest Bioclipse and although there 
were bugs (which we may sort out tomorrow) the OPML tool is able to 
feed CrystalEye stuff to Bioclipse and display it. So I favour the 
traditonal idea of a feed with a fairly short list of back items - 
e.g. last 30 entries. If you miss them you have to look elsewhere for 
the archive. So the RSS feed should be a FIFO.

I am convinced that we should promote Bioclipse as the rich client 
and I think that it could work well for NCI. Will require quite a lot 
of work in quality assurance, etc. - testing different platforms etc. 
but then the CMLRSS could be very exciting. Bioclipse has "features" 
(bad name - really pluginSets) which can be customised to provide 
task-oriented collections of plugins. Thus NCI could have a 
"submisson" feature, or a "latest compounds" CMLRSS feature.

P.

> > At 13:32 +0200 24/5/07, Egon Willighagen wrote:
> > > > I think this way will be easier for clients and servers alike.
> > >
> > > It would not be too difficult to support this in Bioclipse.
> > >
> > > > If the link resolves to an RDF description or a GRDDL-able HTML page you
> > > > could retrieve additional metadata as well.
> > > >
> > > > Any thoughts?
> > >
> > > Yeah, I have seen CMLRSS used at more occassions to publish a full
> > > database. I do not feel this appropriate use of CMLRSS. Maybe it is, but
> > > running it on a large database is going to give problems.

Agreed. I am getting keen on using RDF as a medium-size database 
management strategy. It could manage, say, 10K compounds quite 
easily. Early days


> > See my previous post.  It would be silly to use  CMLRSS to express
> > a full database
>
> I agree.
>
> > > I my view, the area of application of RSS and CMLRSS is news, either about
> > > publishing new blog items or news headlines or new additions to
> > > molecular/crystal databases.
> >
> >
> > Agreed!   Or, as  I wrote earlier, if the  CMLRSS is generated from
> > a selective query (which could be, show me the latest  15 additions
> > for molecules with this sub structure).
>
> I also agree. We haven't done anything in a while mainly because we 
> haven't completely worked out our update procedures so the update 
> schedule has been inconsistent, but I always thought the main use 
> for CMLRSS would be to get notification of new data of interest, 
> where the interest is some defined by some reasonably restricted 
> search. I also echo Henry's misgivings about moving the actual 
> chemical information out of the main feed. The most important 
> summary IS the chemical structure (in the way I think chemists would 
> be interested in using it) and while there is no big problem with 
> having a display of structures with a "click here for additional 
> data", I think it would be pretty useless to have a list of "new 
> compound", "new compound", etc. with a "click here to see structure".

Agreed. We may have to rethink details when performance hits us. I 
think a real win would be a page layout of - say - 48 molecules and 
this could be easily produced from the feed. Perhaps with selection 
buttons so you could choose which to store locally. And maybe - as 
NickDay has done - from a avriety of feeds. So we have a steroid 
feed, a glycoside feed, etc. They can obviously overlapr and be 
merged or intersected. RDF will be great for the filters.

P.



> DanZ
>
> /********************************************
>  * Daniel Zaharevitz
>  * Chief, Information Technology Branch
>  * Developmental Therapeutics Program
>  * National Cancer Institute
>  * 
> <mailto:zaharevitz-KvOs56zQUu5Dzj7Auiz87e1ftBKYq+Ku@xxxxxxxxxxxxxxxx>zaharevitz-KvOs56zQUu5Dzj7Auiz87e1ftBKYq+Ku@xxxxxxxxxxxxxxxx
>  ********************************************/
>
>
>
> _______________________________________________
> Blue-obelisk mailing list
> Blue-obelisk-1MXgZGmSEouYropViTRrB8edHCjMHISts0AfqQuZ5sE@xxxxxxxxxxxxxxxx
> http://hardly.cubic.uni-koeln.de/mailman/listinfo/blue-obelisk

Peter Murray-Rust
Unilever Centre for Molecular Sciences Informatics
University of Cambridge,
Lensfield Road,  Cambridge CB2 1EW, UK
+44-1223-763069