Re: I-D ACTION:draft-ietf-atompub-protocol-14.txt

[Eric Scheid <eric.scheid@xxxxxxxxxxxxxxx>]

On 11/3/07 1:44 PM, "James M Snell" <jasnell@xxxxxxxxx> wrote:

> The critical issue behind the preservation of metadata is the
> communication of intent... e.g. what specific piece of the entry is the
> client intending to modify.  If the current entry has a foo:bar element,
> and the client PUTs an entry with no foo:bar element, does that mean
> that a) the client wanted to remove the foo:bar element or b) that the
> client ignored the presence of the foo:bar element and ended up dropping
> it on the floor accidentally.

... I follow so far.

> The bias in the spec as currently written
> is to say that only intentional modifications should be made.

this, I draw a blank.

What, according to the bias you read in the spec, should be the likely
outcome .. (a) foo:bar goes away, or (b) foo:bar remains?

Put yourself in the position of the client, talking to my server. I'm not
going to say what my server is going to do, other than "it's behaviour will
be as according to the spec".

or is there some <ext:intention/> element I'm unaware of ;-)

e.