Re: Proposal for work on an efficient, browser-friendly, HTTP-based communication protocol for fine-grained information exchange

On 12.08.2010 19:54, Jukka Zitting wrote:
> Hi,
>
> On Thu, Aug 12, 2010 at 10:36 AM, Julian Reschke<julian.reschke@gmx.de>  wrote:
>> Proposal for work on an efficient, browser-friendly, HTTP-based
>> communication protocol for fine-grained information exchange
>
> Sounds useful! See [1] for a few basic use cases and requirements I
> outlined for such a protocol last year. It seems like the proposed
> protocol could match these needs well, and I'd be eager to participate
> in ironing out the details. See below for some initial comments.
>
> [1] http://jukkaz.wordpress.com/2009/11/18/content-repository-over-http/

Yes, that's a good read. When we wrote the proposal I had forgotten 
about it already, otherwise I would have stolen more from it :-)

>> [... WebDAV / Atom(Pub) ...]
>> Both of those protocol specifications are not easily consumed by websites
>> and applications running current browsers and require a lot of client-sided
>> scripting to cover simple read and write use cases.
>
> I'd add that also on server-side processing JSON or HTML forms is
> usually easier than handling WebDAV or AtomPub.

I think that depends entirely on libraries. For WebDAV, there are good 
libraries out there (Jackrabbit and WebDAV for JAX-RS come to mind).

Actually, JAX-RS + the WebDAV extensions described in 
<http://weblogs.java.net/blog/mkarg/archive/2009/02/release_10_of_w_1.html> 
might be a nice prototyping environment for what we're doing.

>> # Data Model
>>
>> 1) Define a collection model (hierarchy, naming), and a representation
>> format.
>>
>> Can we re-use the WebDAV collection model here? Web application authors
>> probably would prefer a JSON representation, so can we simply define this as
>> an alternate representation of a DAV:multistatus description of a
>> collection?
>
> It would be good if anything that can be expressed by this protocol
> could be straightforwardly mapped to WebDAV and/or Atom. The reverse
> does not need to be true, I'd rather go for simplicity than strive for
> a full one-to-one mapping with another protocol.

Yes.

>> 4) Define a property model (something like the intersection between WebDAV
>> properties and Java Content Repository (JSR-283) properties?)
>
> As above, I'd focus on core property types shared by existing standards.

Indeed.

We'll need to think about types (none, some, many?), cardinality, and 
naming.

>> # URIs for collection browsing
>>
>> Assign either hardwired or discoverable URIs for inspecting collections (URI
>> templates?). Or maybe link relations for collection navigation (similar work
>> for versioning: RFC 5829).
>
> I'd rather avoid hardwiring URIs.

+1; but it will make things slightly more complicated.

> ...
> Thanks! This was my first post here, so let me briefly introduce
> myself: I've been working with open source content management since
> -97 and for the past few years I've been focusing on Apache Jackrabbit
> and other related Apache projects. I work for Day Software and have
> participated in the JCR standardization effort in the JCP. I'm from
> Finland, but currently based in Basel, Switzerland.
> ...

Welcome to the IETF (in case we decide to make this an IETF activity)! 
Otherwise, just welcome to this mailing list... :-)

Best regards, Julian

Received on Sunday, 15 August 2010 11:06:20 UTC