Hi all,

I attended the PREMIS-Tutorial at the IPRES this week, but unfortunately 
the really important questions only pop into my head, when I'm already 
lying in bed, so I ask it here:

What is the best practice to be able to search quickly to a huge amount of 
PREMIS-Metadata

For example:

I migrate files from one format into another with a certain tool and 
document this process in the PREMIS-Metadata. Two years later I realize, 
that the tool I used, was not behaving as expected. Therefore I want to 
find all objects in my archive, that have been migrated with this 
particular tool and I have to search through all the PREMIS-Metadata of all 
of my archived objects. Do you store the relevant information only in xmls 
and then you parse each of this xmls? Or do you store it in the xmls and 
additionally index it in a database, which you can query? Or do you have an 
other solution for this kind of problem?

Thanks for any help
Chris