Re: New Version Notification for draft-snell-search-method-02.txt

[Julian Reschke <[e-mail]>]

Am 03.09.2020 um 10:57 schrieb Roberto Polli:
> Il giorno gio 3 set 2020 alle ore 09:11 Julian Reschke
> <[e-mail]> ha scritto:
>>> ... draft-snell-search-method-02.txt ..
>>> What do people think about it -- is it worthwhile? Are there any problems? Consider this a prelude to a CfA...
>> The draft currently is minimal in that it re-uses an existing method,
>> making it more generic (which seems to be good in general).
> 1- I like the general idea;
> 2- such a feature is useful;
> 3- implying that an xml payload must return a webdav response can
> limit its adoption or foster non-compliant implementation (Eg. xml
> requests returning non-webdav responses). This will happen unless  we
> revise https://tools.ietf.org/html/rfc5323, which is widely deployed.
> While I'm no fan of xml, this is quite a limitation.
>
> ```
> for backwards compatibility with existing WebDAV
>     implementations, SEARCH requests that use the text/xml or
>     application/xml content types MUST be processed per the requirements
>     established by [RFC5323].
> ```

That's a detail we can work on; I believe this can be relaxed: we can
make it depend on the namespace of the root element (or even the
concrete element name; see
<https://greenbytes.de/tech/webdav/rfc5323.html#rfc.section.2.2.2.p.2>).

> If somebody prefer to use an alternative method name, we could pick
> QUERY or FIND https://www.w3.org/2012/ldp/wiki/Proposal_for_HTTP_QUERY_Verb
> Once we agree on the semantics, the name can be anything.

Once again, the benefit of using an existing name is that there's
already code out there which knows about the method being safe...

Best regards, Julian