Comments on: REST, WADL, forest, trees http://www.markbaker.ca/blog/2007/05/rest-wadl-forest-trees/ Celebrating the power of the Web Fri, 26 Aug 2011 20:00:51 +0000 hourly 1 http://wordpress.org/?v= By: KevBurnsJr – Blog » A description language for RESTful web services http://www.markbaker.ca/blog/2007/05/rest-wadl-forest-trees/comment-page-1/#comment-1408 KevBurnsJr – Blog » A description language for RESTful web services Thu, 03 Sep 2009 04:17:15 +0000 http://www.markbaker.ca/blog/2007/05/29/rest-wadl-forest-trees/#comment-1408 [...] REST, WADL, forest, trees So if you’re writing (or generating) contract/interface-level code which can’t late-bind to all resources, everywhere, you’re not doing REST [...] [...] [...] REST, WADL, forest, trees So if you’re writing (or generating) contract/interface-level code which can’t late-bind to all resources, everywhere, you’re not doing REST [...] [...]

]]> By: Mark Baker http://www.markbaker.ca/blog/2007/05/rest-wadl-forest-trees/comment-page-1/#comment-253 Mark Baker Mon, 04 Jun 2007 14:07:48 +0000 http://www.markbaker.ca/blog/2007/05/29/rest-wadl-forest-trees/#comment-253 Sure, Andrew. RDF is one attempt at that, offering a standardized data model and a handful of standardized serializations thereof (RDF/XML, Turtle). I don't think that's holding up REST's success at all though. Sure, Andrew. RDF is one attempt at that, offering a standardized data model and a handful of standardized serializations thereof (RDF/XML, Turtle). I don’t think that’s holding up REST’s success at all though.

]]> By: Andrew Wahbe http://www.markbaker.ca/blog/2007/05/rest-wadl-forest-trees/comment-page-1/#comment-247 Andrew Wahbe Sun, 03 Jun 2007 05:16:23 +0000 http://www.markbaker.ca/blog/2007/05/29/rest-wadl-forest-trees/#comment-247 Right... but there is a big fat contract in the way of a hypermedia format specification that is key to the REST interaction. The contract is negotiated in the HTTP interaction by way of the accept header. So while all services expose the same interface, all clients do not expose the same interface. We get away with it in UI-focused services by using standard formats like HTML, SVG, VoiceXML, etc. But for machine to machine interaction, its not clear what one should use, or if a "standard" markup is even possible. Saying "use XLink" doesn't cut it. You need to be able to express all forms of HTTP interactions (all verbs and headers). I think this is the biggest roadblock to RESTs success right now. Don't we need to get people to move their focus from WADL to a standard M2M markup (or at least a framework for building markup focused on a specific market)? Right… but there is a big fat contract in the way of a hypermedia format specification that is key to the REST interaction. The contract is negotiated in the HTTP interaction by way of the accept header. So while all services expose the same interface, all clients do not expose the same interface. We get away with it in UI-focused services by using standard formats like HTML, SVG, VoiceXML, etc. But for machine to machine interaction, its not clear what one should use, or if a “standard” markup is even possible. Saying “use XLink” doesn’t cut it. You need to be able to express all forms of HTTP interactions (all verbs and headers). I think this is the biggest roadblock to RESTs success right now. Don’t we need to get people to move their focus from WADL to a standard M2M markup (or at least a framework for building markup focused on a specific market)?

]]> By: John D. Heintz http://www.markbaker.ca/blog/2007/05/rest-wadl-forest-trees/comment-page-1/#comment-252 John D. Heintz Fri, 01 Jun 2007 21:04:45 +0000 http://www.markbaker.ca/blog/2007/05/29/rest-wadl-forest-trees/#comment-252 I agree that a description language for REST is a bad idea. I tried to describe a declarative data (microformat?) approach to solving the same type of problem. I've called the idea signposts to indicate that the data provides direction that is also machine processable. See http://johnheintz.blogspot.com/2007/05/does-rest-need-dl.html John I agree that a description language for REST is a bad idea.

I tried to describe a declarative data (microformat?) approach to solving the same type of problem. I’ve called the idea signposts to indicate that the data provides direction that is also machine processable.

See http://johnheintz.blogspot.com/2007/05/does-rest-need-dl.html

John

]]> By: Patrick Mueller http://www.markbaker.ca/blog/2007/05/rest-wadl-forest-trees/comment-page-1/#comment-251 Patrick Mueller Wed, 30 May 2007 17:07:00 +0000 http://www.markbaker.ca/blog/2007/05/29/rest-wadl-forest-trees/#comment-251 I posted a response on my blog: http://tinyurl.com/2f9xeq I posted a response on my blog: http://tinyurl.com/2f9xeq

]]> By: hughw http://www.markbaker.ca/blog/2007/05/rest-wadl-forest-trees/comment-page-1/#comment-249 hughw Wed, 30 May 2007 16:16:47 +0000 http://www.markbaker.ca/blog/2007/05/29/rest-wadl-forest-trees/#comment-249 Pete -- "Hypermedia as the engine of application state. (We need a shorter way to say that.)" HATE OAS? Pete — “Hypermedia as the engine of application state. (We need a shorter way to say that.)”

HATE OAS?

]]> By: Mark Baker http://www.markbaker.ca/blog/2007/05/rest-wadl-forest-trees/comment-page-1/#comment-250 Mark Baker Wed, 30 May 2007 02:44:41 +0000 http://www.markbaker.ca/blog/2007/05/29/rest-wadl-forest-trees/#comment-250 Bingo! Bingo!

]]> By: Pete Lacey http://www.markbaker.ca/blog/2007/05/rest-wadl-forest-trees/comment-page-1/#comment-248 Pete Lacey Tue, 29 May 2007 23:08:03 +0000 http://www.markbaker.ca/blog/2007/05/29/rest-wadl-forest-trees/#comment-248 "10 kudos to whomever identifies the specific constraint being violated)" Hypermedia as the engine of application state. (We need a shorter way to say that.) “10 kudos to whomever identifies the specific constraint being violated)”

Hypermedia as the engine of application state. (We need a shorter way to say that.)

]]>