Hi all,

I've read this document and have delightfully little to say about it. It's a pretty straightforward enhancement filling a long-standing void, which is good.

The only real comment I might offer is that this is in last call jointly with a document defining 64-bit time types. Seciton 5.1 does mention future work for improving date representations (now concurrent work), but it doesn't seem that there is compelling reason to reopen this document to use the 64-bit time representation.

It's unclear whether we want to suggest that the values of the GetSizeV2 flags match the values of the DumpV2 flags, since they need not have the same flags. Given that there's only one flag at present and adding more would require some though, I don't think it matters either way.

It also struck me in parts of Section 8 that we sometimes request things of the registry and at other times suggest actions for them. I don't know how important the distinction is; we could probably just request things always.

Anyway, this document is fine and I would be happy if it was published as-is. I do not think any changes are necessary.

-Ben
_______________________________________________
AFS3-standardization mailing list
[email protected]
http://lists.openafs.org/mailman/listinfo/afs3-standardization

Reply via email to