Re: [Myexperiment-discuss] Re: googling an exp-object

[Stian Soiland]

On 1 Jun 2007, at 08:21, Steve Pettifer wrote:

> 1) Use a mime/helper application setup. This requires some co- 
> operation from the server side in terms of generating content that  
> can be recognised by the client end as being targeted at a  
> particular helper application. In the case of myExperiment this may  
> be ok (e.g. a bit of scufle, or some other thing that we have  
> control over?) -- for UTOPIA which has to deal with much more  
> heterogeneous resources that aren't under our control, this isn't  
> so good. Advantage: easy to implement, fits neatly with browsers'  
> helper apps (especially if they are java)

I would recommend this, as it's really the cleanest, and requires the  
least effort. I have chosen these mime-types for use by the long  
running workflow service:

application/vnd.taverna.scufl+xml  (workflows)
application/vnd.taverna.baclava+xml  (input/output data documents)

(The use of Baclava documents is transitional only, though, as it's a  
very heavy XML serialisation that can't handle large data properly. I  
am thinking of a simpler structure where input/outputs are just  
folders described with text/uri-list - and fetching the actual data  
elements will give you the raw binary of the claimed mime-type -  
unless requested with an Accept of text/rdf.)


I have not made up any mime-types for the progress report or the  
provenance log, although the ladder would presumably be available as  
an RDF serialisation. I'm not so certain if using an RDF mime type  
would be very helpful here, as it is a bit too general, and the  
browser might pop up Address Book to try importing the FoF data  
instead of popping up a provenance browser. Maybe application/ 
vnd.taverna.logbook+xml or application/vnd.taverna.logbook +rdf+n3 or  
some weird variation.


(The REST-part of the workflow service uses XML documents of  
application/vnd.taverna.rest+xml)

I have not yet registered these with IANA, though, but I don't see a  
reason why we can't. (They are quite relaxed about vnd registrations)


However, I don't see why a 'google' for "experiment objects" (?)  
would have to jump straight to the XScufl or similar raw data. Why  
can't we show a HTML-page that describes the main features of  
whatever it is we are talking about? That way people without the  
necessary local software will be able to see what it's all about  
before starting to install lots of stuff. This is the Web 2.0 way.

--
Stian Soiland, myGrid team
School of Computer Science
The University of Manchester
http://www.cs.man.ac.uk/~ssoiland/