+1 I just tried to "refactor" this API (http://www.subbu.org/weblogs/main/2007/12/a_restful_versi.html ) to be resource oriented, and it is not that hard. ...
... I nominate it for the 2007 Restless awards, in the much contested category of "things that claim to be RESTful but do side effects in their GETs" along...
This is a variant of what I call as SOAPy REST (http://subbu.org/weblogs/main/2007/10/soapy_rest.html ). Whoever wrote this API had no idea of why they were...
Hi, As I just wrote about DBPedia in "James Gosling has a foaf name" [1] I thought this would be a good time to point people to that article which shows very...
... Amazon does the same thing with their ecommerce api, which this list has criticized long ago. Not that it did any good...I just checked, the latest...
Bob Haugen
bob.haugen@...
Dec 15, 2007 11:15 pm
9995
... I should probably not publicly admit that I looked at that for some time going "Why is 'GET ...
... I think it has to constrain state-transition paths, to some degree. That is, without intelligence, the possible state transitions (both sucessful/[23]xx ...
I agree with Chris. Symmetry is irrelevant. For some scenarios it makes sense for others it doesn't. POST/PUT can accept various kinds of media-types and GET...
... I don't think I understand what you're trying to say here. I didn't, or didn't mean to, suggest that there should be symmetry between media types supported...
... I can tell you I do that as a matter of practice, supporting multiple content-types on requests and responses, specifically for structured data. If the...
... I think there is a difference between offering multiple representations of a resource via GET and updating the same resource via a POST/PUT. With POST/PUT,...
... So once again, like clock-work my e-mail client decided to reply but not to the list, and I ended up in a private conversation about this with John. Too...
Julian: Yes, we are understanding each other better now. Extending Accepts is one possibility. Chris Dent mentions adding information to the body of a 415...
Chris: Thanks for the response. Your approach sounds quite similar to the one upon which I am depending. Esp. your description of how "conneg" is handled for ...
... Docs. Maybe not best, but easiest. When the client learns of the existence of your service, make the information available then. When you send a 415 to...
... OK, so this is about two different formats essentially containing the same information, not two variants (such as "en" and "fr" language versions that...
Julian: Thanks for helping me clarify. I may be making a lot out of nothing here. If I want to support PUT/POST to a resource URL in more than one media type...
... I think this must be fairly common, or if not it should be. It's kind of the point right? You have a resource or some collection of resources that you want...
... Content-Location. ... I think it is uncommon, at least for PUT. ... Upfront? I may be misunderstanding what you're trying to do. Is this about authoring...
Mike, I I understood you correctly, then #1 is extremely common. For #2, don't have a good answer. Maybe this could be done via an OPTION request on the...
Julian, I believe Mike was asking about supporting multiple content-types for a request, not negotiating the response content-type. - Steve ... Steve G. Bjorg ...
Julian: Maybe I am not reading your reply correctly, did you mean Content-Location or Content-Type? Again, my question is two-fold: - how common is it to...
... I would strongly discourage to use content-negotiation for authoring. Let the server return a Content-Location upon GET/HEAD; and use that URI for...
I'm strong believer in overloading by content-type. This is the direction we're heading with our wiki API as well. The interaction space is defined by URI x...
Having problems with message search?
Fill out this form to ensure your group is one of the first to be migrated to the new message search system.
