Question: harvesting managed vs. all resource records

[Aurelien Stebe]

Hi everyone,

I don't understand why we would need to know where a resource is coming 
from ?
or even when it was harvested.

As for the sets, I think it is a good idea to support them as described 
in the RI doc right now.
By using the "updated" attribute it should be easy to sort the resource 
modifications.
Except for cases when a resource is updated twice the same day.
For this, the "updated" attribute could be made a "datetime" instead of 
"date",
or we could force all registry to support the "datetime" granularity for 
the OAI interface.
or harvest the VO world as follows : harvest all resources from one full 
registry, just in case, and
then only harvest managed resources for all the other known registries. 
This way, a resource gets updated twice at the worse.
I think this last solution should be enough. No more than a few 
resources will change every day.

The other problem with sets is that we could harvest all SIA resources 
or all managed resources,
but not all managed SIA resources. Anyway, would it be useful ?

By the way, shouldn't the "search" interface return only "active" 
resources ?
There is not many "inactive" or "deleted" resources for the moment, but 
it might be a problem latter.

Cheers,

Aurelien STEBE

Ray Plante wrote:

> On Thu, 31 Mar 2005, Gretchen Greene wrote:
>
>>When you answered yes to the point of origin are you referring to the
>>'harvestedfrom' and if so is this now in the spec?  Also the
>>harvestedDate?
>>    
>>
>
>I wasn't, but for those curious as to what Gretchen is refering to, see 
>Question 4 from the discussion list at 
>http://www.ivoa.net/twiki/bin/view/IVOA/RegistryInterface.  
>
>cheers,
>Ray
>
>
>
>  
>