Seairth, Why not use IDL or WSDL? Why use HTTP methods? -Alex- ___________________________________________________________________ S. Alexander Jacobson...
3942
Alex Jacobson
shop2it
Sep 5, 2003 12:38 am
Why not simply talk about document types? What exactly is a RESTful API aside from the verbs of HTTP? -Alex- ...
3943
Seairth Jacobs
seairthjacobs
Sep 5, 2003 1:11 am
Welcome back to this world. :) In answer to your question: simplicity. SID (which replaced SUDS) can be learned in a matter of minutes. No, it is not a...
3944
Phil Eskelin
philip_eskelin
Sep 5, 2003 1:20 pm
I don't really think of it as an API - the only API that exists in my opinion is the HTTP API. The framework is built upon this core API and provides you with...
3945
Alex Jacobson
shop2it
Sep 5, 2003 10:29 pm
I read Fielding's dissertation. If you read RFC2616, you can see why I ask this question. A framework compliant with RFC2616, would simply be a set of types...
3946
Seairth Jacobs
seairthjacobs
Sep 8, 2003 2:03 pm
Is an interface part of a resource? In other words, is the ability to make requests to and get responses from a resource part of the resource itself? Or is it...
3947
Walden Mathews
waldenmathews
Sep 9, 2003 12:08 am
Seairth, I don't think the question is clear the way you have worded it. In concrete terms, "when you GET a representation of a resource, should that...
3948
Seairth Jacobs
seairthjacobs
Sep 9, 2003 1:00 am
From: "Walden Mathews" <waldenm@...> ... That's not what I am asking. Rephrased: Is the *ability* to perform a GET against a resource part of that...
3949
Alex Jacobson
shop2it
Sep 9, 2003 1:44 am
Ok, but if you are interesting in doing something WSDL-like, why are you interested in ReST? My generic claim is that exposing an resource via HTTP is enough...
3950
Chuck Hinson
cmhinson@...
Sep 9, 2003 3:17 am
... Is that the same as 'Is the ability to respond to a GET request part of the resource?' ... Are the words that are used to describe an idea part of the...
3951
Mark Baker
gonga_thrash
Sep 10, 2003 4:48 am
I'd appreciate comments on this, which is basically a way of communicating what HTML forms does, with RDF, and with some RESTy self-description voodoo (that...
3952
jalgermissen@...
algermissen1971
Sep 10, 2003 6:42 pm
Mark-- thanks for this, it's very interesting. Some questions: The container provides the ability to specify which media type it accepts. If this is XML, then...
3953
Mark Baker
gonga_thrash
Sep 10, 2003 8:38 pm
... Good to hear, thanks. ... "acceptedNamespace" is just a hack to address the problem that people are using application/xml incorrectly, as if the...
3954
Hugh Winkler
hwinkler99
Sep 10, 2003 9:56 pm
... http://www.w3.org/DesignIssues/Fragment.html#Fragment2 argues that rdf namespaces must end in fragment identifiers....
3955
Alex Jacobson
shop2it
Sep 10, 2003 10:10 pm
Exactly what purpose does this spec serve that is not already served by HEAD, 405, 413,414, and 415? HEAD gives you the content-type which implies both the...
3956
Mark Baker
gonga_thrash
Sep 11, 2003 2:29 am
... Right. Tim's wrong. This is the rathole known as httpRange-14[1] in W3C TAGland. Tim believes that the range of http URIs is a "conceptual work",...
3957
Yannick Loiseau
pioupiougroups
Sep 11, 2003 6:51 am
... First, you need an uri on which to perform your HEAD. Therefore you need at least an uri list. ... Suppose that the default content-type is html, to be...
3958
Mark Baker
gonga_thrash
Sep 11, 2003 1:09 pm
... Why do we have phonebooks? I suppose one can just call all phone numbers and ask for who they're looking for... And why do we have garbage cans and...
3959
jalgermissen@...
algermissen1971
Sep 11, 2003 1:40 pm
-- Mark-- here are some ideas about describing existing POST containers: In general, there are millions of them on the Web in the form: <form...
3960
Tyler Close
tjclose
Sep 11, 2003 5:37 pm
Waterken Inc. recently presented the web-calculus to HP Labs in Palo Alto, California. The talk explains how the web-calculus is derived from existing REST...
3961
Ken MacLeod
poy_mp3
Sep 11, 2003 6:21 pm
HttpMethodsSupport[1] is copied from the Atom Wiki CarrotVsOrange page[2], summarizing the support of PUT and DELETE methods in various server and client...
3962
Yannick Loiseau
pioupiougroups
Sep 11, 2003 6:54 pm
Is their an consensus to deal with http header in url query string when it's impossible for the client to manage them (e.g. html forms). It could be useful if...
3963
Alex Jacobson
shop2it
Sep 11, 2003 7:46 pm
<polemics> Hmm, perhaps that is better than having to spider the entire internet to find out what the entry in the phonebook actually means.... </polemics> But...
3964
Alex Jacobson
shop2it
Sep 11, 2003 7:46 pm
--On Thursday, September 11, 2003 8:52 AM +0200 Yannick Loiseau ... Yes, you obviously need a URI. The question is whether it make sense to provide the...
3965
Yannick Loiseau
pioupiougroups
Sep 11, 2003 8:27 pm
Le jeu 11/09/2003 à 20:18, Alex Jacobson a écrit : [...] ... [...] I disagree. your example is simple enough to make it obvious, but it is not always the...
3966
Alex Jacobson
shop2it
Sep 11, 2003 8:56 pm
--On Thursday, September 11, 2003 10:28 PM +0200 Yannick Loiseau ... Ok. Give me a non-obvious case that is not an example of poor engineering and that cannot...
3967
Mark Baker
gonga_thrash
Sep 12, 2003 2:25 am
... There's a middle ground here. It's not perfectly RESTful, but is more so than the everything-with-POST approach. And what's making it unRESTful is HTTP's...
3968
Mark Baker
gonga_thrash
Sep 12, 2003 2:41 am
BTW, I should also mention that I think that Tim's solution is "good enough". Tunneling is bad, but tunneling uniform semantics ("delete") makes it bearable....
3969
Yannick Loiseau
pioupiougroups
Sep 12, 2003 6:47 am
... as the use of html forms imply the use of an html browser as a GUI, and not a custom client, the client can't deal with many http headers (e.g. changing...
3970
Bill de hÓra
bdehora
Sep 12, 2003 11:14 am
... Urrgghhh... Fwiw it's still a lot better that a recent and nasty hack I tried where DELETE and PUT were mapped onto URIs which is a complete mess. I know...
