new take on resource registration best practice

[Ray Plante]

On Thu, 24 Oct 2013, Markus Demleitner wrote:
> [Ray said:]
> > Will we need to *require* a registry to implement the ingestion
> > behavior you described?  Is it better to get the original resource
> > record in a optimal state to begin with?
> 
> Since the effects of whether or not that  happens are visible to the
> registry client, I am completely convinced that yes, this must be
> mandated.  Maybe not in a REC initially, but clients must be able to
> rely on this when posting queries, or we'll see the hell of
> registries giving a different view of the VO...

I think it is better not to mandate this but leave it as a value-added 
feature.  (Of course, I was never bothered by the different answers 
issue, but I admit that people complained loudly about it.)

To explore this choice, we might consider what the world would 
subsequently look like.  We have a user that is looking for a catalog 
related to some science topic and servced by a TAP service.  
Implicitly assuming that the best practices are in play, she forms a 
simple query that says, 

   gimme resourses that mention my science terms and has the 
   standardID for TAP.

She goes to the VAO Registry to submit the query.  This registry 
follows the DataCollection2 best practice for its own resources, but 
it does not attempt to "fix" descriptions where there are separate 
collection and service resources.  Thus, the query only returns the 
DataCollection2 resources.  If she is savvy and realizes that not all 
publishers follow this convention and thus expects she may be missing 
relevent catalogs, she may retry her query putting in extra (complex) 
constraints to follow the relationships.  

Savvy or not, she tries now going to the VO-Paris registry and submits 
her simple query again.  This time she gets more resources back 
including those that followed the DataCollection2 best practice *and* 
those that didn't.  At this point says:

  A.  Wow--VO-Paris is awesome.  They returned better results than 
      VAO.  Or,
  B.  *^$%&(&%!  What the hell is going on here?  I can't get 
      consistant results from these registries!

(Of course, with or without this special feature, to get consistent 
results would also require a uniform way of handling keyword 
searches.)

A problem I see with what I've called the "value-added" feature in the 
RegTAP case is that one in effect has to change the representation of 
a resource upon ingestion ingo the RegTAP catalog.  With the 
"text-optimized" registry interface, it might be easier to blur the 
distinction a bit.  Maybe the answer is if we mandate this behavior, 
we do it only in the context of the "text-optimized" registry 
interface.  (Does this make sense?)

cheers,
Ray