[MOBY-dev] observations about registrations with RDF

[Mark Wilkinson]

Martin Senger wrote:

>1) Testing registry seems to have still older version - it still returns 
>RDF with CDATA. It would be good to have it restarted,
>  
>
Done

>2) The service description, as returned in the RDF, lost its CDATA 
>protection. Perhaps it is correct,
>

Um... well, I removed it at your request, so... :-)

> I do not know enough about RDF 
>documents. I have just observed that the 'strange characters', 
>such as '>', that I put into my service description, were returned 
>escaped: the character '>' was escaped as '>' - which does not seem 
>to be correct (but perhaps I am mistaken).
>  
>
AFAIK, anything that is not contained in a CDATA section will be 
escaped.  It is done at the library-level and I have no (?) control over it.

>Anyway, I wonder why to play with escaping at all, why not to keep there 
>CDATA.
>
Because you asked me to remove it, and your reasons were valid :-)  The 
RDF is a valid XML fragment, so there was really no good reason to put 
it in a CDATA section.

> We have CDATA in service description just because we did not want 
>to escape everything there - is that correct?
>  
>
Correct.

>3) I also wonder where are all our LSIDs? There is no single LSID in the 
>whole returned RDF. I thought that we expressed almost everything as LSIDs 
>in biomoby (at least internally).
>  
>
That's a very good point!  I agree with you 110%, and I can implement 
this fairly quickly at at the RDF-production end of things.  At the 
RDF-consumption end, I'll have to talk to Nina about it.  I'm fairly 
sure that we can work it out quickly.  The problem at the moment is that 
LSID resolution in MOBY is completely broken.  The latest version of the 
Perl LSID resolver stack (or at least, the latest the last time I 
looked) is built on top of alpha code libraries, and other libraries 
that are not yet in CPAN.  I was able to get it to compile (with a bit 
of fussing) on my Linux machine, but it throws errors up the ying-yang 
on mobycentral (Solaris).  As such, LSID resolution of MOBY LSID's will 
not be possible until someone has time to write an LSID resolver in 
Java, or the LSID code developers make changes to their stack  :-/   :-(


>4) I use this list to strongly express how I agree with the last Rebecca 
>email about being still able to deregister a service by calling a 
>deregister method even if the service has been registered with the 
>signatureURL.
>
>  
>
Well, I strongly disagree with you :-)   Gosh, THAT has never happened 
before!  ;-)

You have to remember what the driving force was to create the RDF-based 
deregistration!! The problem is that any kid with the ability to read an 
API could have come in and desroyed our registry by deregistering all of 
the services in it with a 5-line Perl script!  Authentication was just a 
pain in the arse and totally impractical, so the only remaining solution 
was to remove that functionality of MOBY Central and move the problem to 
the service providers machine.  I am loathe to move away from that model 
because in the last 18 months nobody has come up with a better solution!!

If a better solution exists, then we can certainly revisit the issue...

>5) Finally, just to make this list complete, I would like to remind the 
>others that there had been a discussion about possibility to get RDF from 
>the registry even for services that had been already registered. I would 
>like to have this feature soon - because it may need an addition tag in 
>the service object, so the sooner we know about it the better.
>
>  
>
Yes, I might actually try to get that done tonight in my room, and 
upload it to mobycentral tomorrow.  I'll let you know.

Cheers!

M

>  
>