Re: Status of VOspace specification

From: Guy Rixon <gtr-at-ast.cam.ac.uk>
Date: Fri, 22 Jul 2005 07:17:05 +0100 (BST)


On Thu, 21 Jul 2005, Reagan Moore wrote:

> I have not seen a note or working draft for the VOSpace
> specification. Are there general guidelines for what VOSpace will
> provide? Are the following capabilities intended:
>
> - namespace into which images and catalogs may be registered

Yes, definitely.

> - descriptive attributes for each registered item

Yes, although the scope of the description is up for debate.

> - administrative attributes (such as location)

Yes.

> - authenticity and integrity attributes (desired for preservation)

Possibly; seems sensible. This hasn't been discussed on the AstroGrid side. You've summarized these requirements in various talks and papers. For the record, could you mail a reference to this list?

>
> I am interested in seeing if the VOSpace specification can encompass
> the preservation attributes required for long term storage of images.
> Can a VOSpace instance be used as the IVOA preservation environment?

The AstroGrid view is that VOSpace is a catalogue of data-items on VOStores. Most of the preservation requirements then devolve on the stores, and I expect that some of these will not be suitable; e.g. some may not back up their contents; e.g. some, like the ones that ingest tables into RDBMS, do not preserve the data bit-for-bit. Therefore, the preservation environment would have to be select, through VOSpace, stores with suitable characteristics.

We need to decide whether there are multiple VOSpaces or just one. I.e., is there one, global directory tree or is there a separate tree for each VOSpace service? If we support multiple spaces, then the preservation environment could be one such, in which the storage is internal.

Guy Rixon 				        gtr-at-ast.cam.ac.uk
Institute of Astronomy   	                Tel: +44-1223-337542
Madingley Road, Cambridge, UK, CB3 0HA		Fax: +44-1223-337523
Received on 2005-07-22Z08:17:23