Next message: Geoffrey M. Clemm: "Re: workspaces as collections"
Date: Wed, 31 May 2000 17:28:31 -0400 (EDT)
Message-Id: <200005312128.RAA25764@tantalum.atria.com>
From: "Geoffrey M. Clemm" <geoffrey.clemm@rational.com>
To: ietf-dav-versioning@w3.org
Subject: Re: A Plea for the workspace header (Was: Why do we need working resource ids ?)
From: Edgar@EdgarSchwarz.de
I would like to keep the workspace header. Workspaces are a fundamental
concept for DISTRIBUTED versioning so they earn their own header.
When starting to implement workspaces I at once ran into the authorization
problem. I decided to try controlling write access on a workspace scope.
So I wouldn't like to extract a workspace from a request URL.
It's much clearer if I can check write permissions by scanning
a Workspace and an Authorization header. I don't have to check whether
a prefix is a workspace or not.
Ah yes, of course. I withdraw my offer to remove the Workspace header.
For a server that does not support nested workspaces (and most probably
will not), this could be a significant optimization.
In this context I also have another question (somebody already raised it
some days ago). When authorization fails, is there some appropriate
HTTP code to send back to the client ?
401: Unauthorized.
> appropriate revision (probably the commonest case), or (since we are
> using the LABEL marshalling) via the revision URL.
I also prefer LABEL. It's a concept which is easy to explain to users.
OK, I think that's a pretty convincing majority in favor of LABEL.
Remember the KISS priciple :-)
Yeah, now if only we could all agree on what is "simple" (:-).
Cheers,
Geoff