... No. [...] ... I think your understanding is perfectly right: Well-designed URIs are great, but they don't influence whether something is RESTful or not. ...
Ben, In my HTTP i18N patterns article [1], I deal with this question as follows: "Health warning: URI structure is orthogonal to REST Before I go into more...
Hi, ... Our in-house architectural style (which I like to think of as more or less RESTful) mandates URLs to be hierarchical with authority boundaries in mind....
... I haven't thrown any babies out with any bathwater. ... Right. Technically. Or as I put it in my first reply, "strictly speaking." Then I went on to offer...
... Correct. ... No. The confusion you refer to isn’t new. It has existed for as long as the term “REST” has and is the mark of a certain incomplete...
On Mon, Feb 2, 2009 at 8:42 AM, Aristotle Pagaltzis <pagaltzis@...> ... (I know I'm going to regret asking this, but it's been bothering me for so long that...
... If a server returns to the client a URI template -- and assuming clients have a way to identify URI templates in the returned hypermedia e.g. the...
OpenSocial REST API - http://docs.google.com/View?docid=dcc2jvzt_37hdzwkmf8 If I understand it right, the above URL is not RESTful as it's not denoting a...
... If both of these URIs dereference to the same representation, everything else being equal, what difference does it matter if one of them uses a query...
Here is a way to look at this. Is the client required to understand that it needs to pass "/view" in the URI to "view" this document, and, perhaps, add "/edit"...
On Mon, Feb 2, 2009 at 10:32 AM, Ramamoorthy Subramanian ... No. You could build a perfectly RESTful app using the above URL design. It just retrieves a...
... It doesn't. It is however a sign that someone may have made some unRESTful decisions elsewhere that is reflected in their choice of the URI. In and of...
Nick, I see where you're getting at. Cool URIs don't change because of linking. But those URIs are opaque, even though they are entry points. The point of the...
... Given my argument, what's the difference between "bookmarking" a URI template for later use and bookmarking the set of "fully instantiated" URIs the...
... Because templates are not addressable, URIs are. If you want bookmarking of templates, as a specific jump through hypertext to a resource representation,...
... If you grab an HTML page with a form in it and save it for later, you might have just bookmarked a URI template for later. ... I do. Some sites support...
... template ... that ... Great point Assaf! I'd completely forgotten about OpenSearch. OpenSearch is effectively an architecture for "bookmarking" search URI...
... other ... between ... could ... discourages ... clients. ... URI ... I am ... I agree with this. Yes, the problems of URI design (opaque vs "hackable")...
... There is no opaque vs hackable. All URIs are opaque to all but a couple of processes. It's not that URIs arguably should be opaque. Every single one of...
... couple ... single ... If a URI scheme is human-readable, clearly documented, and guaranteed by the server to never change, how is that opaque? I'm not...
My answer to the 'REST-ful URI question is this: URIs are the *product* of a REST-ful implementation, not the initiator. IOW, URI design is that *last* step in...
... It’s hard to argue that it is incorrect per se. As long as state transitions are driven by links and forms, and you respect the uniform interface, then...
... Which is, of course, *the whole point*: the server can evolve without having to inform the clients – the resources created today can be put into a...
... The things reading it not being human, http://example.net/main_api/users/user23 is no less opaque than http://example.net/ohiuwftojiics09fsdao. The...
... I think maybe we're arguing semantics here, but your first example most certainly has the potential to be what I would consider transparent or "hackable"...
... Yes, but not unimportant semantics. It's an important principle of web architecture that URIs are opaque. Using "opaque" differently doesn't change that. ...
... or ... Yes, I agree that abuse can lead to brittleness. But I would apply that same statement against the semantics exposed by pretty much any API. In ...
... In my case too. Also in Roy Fielding’s case, which leaves little wiggle room for turning this into a matter of consensus. ... URI templates or forms...
... This is where I get a little bit stuck when I think about the hypermedia constraint of REST. At some point it seems that a transition needs to be made...
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.
