On 10/16/13 2:11 PM, Rushforth, Peter wrote:
> > And with that the ability to debug in the small, to have edges that
> are very strict but middle pieces that are very flexable ... but still
> defined interfaces.
>
> I've said this before here, so I don't want to go too deep with this
> now, but REST standardizes the interface. If XML defined common access
> to the interface (aka hypermedia affordances)
>
> (across *all* of XML, not just XML with namespaces), the "small pieces,
> loosely joined" could be XML.
I reflected a little bit on REST in the piece:
"I’ve also pondered what role REST may play in these conversations. The
PUT and DELETE methods are, of course, all about side-effects, and not
so much about flow. I don’t yet have a good answer here."
My general sense is that XML+REST is wonderful stuff, but more likely an
entry (or exit) point than the communications method used throughout
this kind of flow. I would be happy if REST - especially with a
hypermedia approach - would be sufficient, but it doesn't quite feel
like a neat fit with a pipeline-centric model.
I'm sure it could be made to work, but I'm not sure it should be made to
work.
I also have mixed feelings about hypermedia affordances *in* XML. Mostly
I'm wary because too many hypermedia conversations seem to reach a point
where everyone thinks they agree about what things mean and what things
do, and then realize that the agreement isn't that robust.
Roughly, I like the idea, but don't trust myself for liking it.
(I should note that hypermedia conversations were another piece
contributing to my sense that this model is coming together for real
this time, just not in as direct a way.)
Thanks,
--
Simon St.Laurent
http://simonstl.com/
