Posted by Jonathan Altman
Wed, 02 Jul 2008 06:07:00 GMT
Paul Graham first wrote about a strawman hypothetical programming language blub that examined the constraints that people who choose to stay firmly embedded in only one language seem to impose upon themselves and their programming capabilities and creativity.
Steve Vinoski, who has been writing a bunch about some of the failures of RPC-style distributed systems technologies (and should know, having been involved in CORBA), I think just extended the theme to distributed systems programming here”.
Posted in technical | Tags advocacy, blub, RPC, services, web, webservices | no comments
Posted by Jonathan Altman
Wed, 02 Jul 2008 06:07:00 GMT
Paul Graham first wrote about a strawman hypothetical programming language blub that examined the constraints that people who choose to stay firmly embedded in only one language seem to impose upon themselves and their programming capabilities and creativity.
Steve Vinoski, who has been writing a bunch about some of the failures of RPC-style distributed systems technologies (and should know, having been involved in CORBA), I think just extended the theme to distributed systems programming here”.
Posted in technical | Tags advocacy, blub, RPC, services, web, webservices | no comments
Posted by Jonathan Altman
Wed, 02 Jul 2008 06:07:00 GMT
Paul Graham first wrote about a strawman hypothetical programming language blub that examined the constraints that people who choose to stay firmly embedded in only one language seem to impose upon themselves and their programming capabilities and creativity.
Steve Vinoski, who has been writing a bunch about some of the failures of RPC-style distributed systems technologies (and should know, having been involved in CORBA), I think just extended the theme to distributed systems programming here
Posted in technical | Tags advocacy, blub, RPC, services, web, webservices | no comments
Posted by Jonathan Altman
Mon, 02 Oct 2006 02:17:00 GMT
I have posted several times about it, made a bunch of different arguments, and anybody who has talked to me about web services has heard me try and make the argument not to force a web service to be the serialized transfer of object artifacts. But here’s the summary in a nice single sentence from the post CommonRESTquestions:
REST can be seen a documented-oriented subset of Object-Orientation. It deliberately reduces the expressiveness of Objects down to the capabilities of resources to ensure compatability and interoperability between components of the architecture.
The rest of that paragraph goes on to say
Object-Orientation allows too great a scope of variation for internet-scale software systems such as the world-wide-web to develop, and doesn’t evolve well as demands on the feature set change. REST is Object-Orientation that works between agencies, between opposing interests. For that you need to make compromises rather than doing things your own way.
So there you go, someone said in a paragraph exactly what it has taken me 2 years to try and say.
Posted in technical | Tags advocacy, REST, services, SOAP, web, webservices | no comments
Posted by Jonathan Altman
Fri, 09 Dec 2005 19:47:00 GMT
Earlier, I posted about how not forcing all access to web services to go through objects that were serialized into and back out of XML but instead were XML documents that were designed to stand on their own made it easier to implement both web services and web services clients.
Elliotte Rusty Harold sets out a nice short example of the nature of the problem. He summarizes the problem nicely in this quote
“don’t “help” users out by changing XML into something else, especially not objects. Don’t assume you know what they’re going to want to do with the data. Give them the XML and let them use the classes and objects that fit their needs, not the ones that fit your needs. XML is exchangeable. Objects are not.”
Again, for the record: if I work with your web service, you do not know what data, data structures, or code artifacts I will have in place to access your web service. Please don’t try to guess by forcing me through your view of how the software artifacts should look. Let me figure out how to make the raw message, and how to interpret what you send back. SOAP toolkits that closely map XML to particular language-specific data structure constructs and back make this hard.
Posted in technical | Tags advocacy, document-passing, POX, REST, services, SOA, SOAP, web | no comments
