On 3/22/06, Mark Baker <[EMAIL PROTECTED]> wrote:
> On 3/21/06, Mark Nottingham <[EMAIL PROTECTED]> wrote:
> >
> > On 2006/03/20, at 12:20 PM, Ryan King wrote:
> >
> > > Also, html forms could be useful for documenting the parameters
> > > available on rest resources. Mixed with microformats, as above,
> > > this could be a very useful way to document a rest api. It would
> > > cover much the same territory as WADL, but in a web-native way
> > > which can be very useful to developers, as the forms are a
> > > functional sandbox.
> >
> > Is that really the case? A form seems almost like a choreography to
> > me (although I'm fairly ignorant of that world); i.e., it tells you
> > how to do one task, rather than document the whole interface.
> >
> > E.g., if I have a resource that accepts GET, POST, PUT and DELETE,
> > the form is probably only going to tickle one of those methods... You
> > could put together a set of forms that covered all of the ground, of
> > course, but for some tasks, that's pretty clunky.
>
> +1

Mark, I notice that form-based language in nature does not describe
the output/response details, contrasting with several others service
description language. I wonder what's the rational behind its design.

--
Teo HuiMing (toydi)
teohuiming.work at gmail dot com
_______________________________________________
microformats-rest mailing list
[email protected]
http://microformats.org/mailman/listinfo/microformats-rest

Reply via email to