Re: Review of draft-ietf-webdav-quota-02.txt

Just to restate, here are other issues that make the spec more 
complicated than it needs to be and/or incomplete:

- The spec tries to make handling of quotas simple by just reporting a 
single quota, selected by the server (just like in NFS). This is fine as 
long as the information is only used for display in the UI or for 
diagnostics. However, it fails completely if you want to make the quota 
authorable, because in this case you have no control whatsoever about 
*what* quota you're trying to change. I think this is the cause for most 
of the confusion about the "third" property. The simplest way to fix 
this is to remove support for authoring quota values from the spec.

- Disk limits are not quotas. Treating disk limits the same way causes 
funny effects. In addition, method failures due to physical storage 
limits should be treated differently (for instance, when mapping to a 
Unix FS like in the MacOS X driver, you may want to map to two separate 
error codes). IMHO the spec should clearly state that disk limits are 
*not* quotas. If it adds live properties and precondition/postcondition 
names to marshall to disk limit information *separately*, that'll be 
fine as well.

Regards, Julian

-- 
<green/>bytes GmbH -- http://www.greenbytes.de -- tel:+492512807760

Received on Saturday, 13 December 2003 08:12:05 UTC