Here's an interesting tie-in to the simplicity debate - Four years after the hype, Google has quietly axed the SOAP Web service API to it's search interface. (full story here) And nobody cares. Why?
The SOAP concept grew out of a great simple idea. The Idea was this:
The first iteration of what would become SOAP was pretty straightforward - XMLRPC. Version 1.0 of SOAP built upon that, and was still human readable, and okay. Som
etime after that, the big boys like Microsoft, IBM and Sun jumped on board, and proceeded to complexify, 'standardize' and completely bugger the specification about until the final modern iteration of SOAP was so confusing, that only the most zeal-ridden platform zealots were singing it's praises - and even they didn't actually know how the thing worked anymore.
Marketing execs all over the world were so desperate for 'Web Services' that they confused them with 'Web Sites', 'Web Parts', and 'Windows Services'. The only thing spreading faster than misinformation was glossy brochures. The whole thing was deranged.
The original, simple idea was lost in a maelstrom of add-ons and transactional plug-ins and security goo and it all became horrible. Actively trying to build a solution architecture on top of this thing is something that no self-respecting architect would do. Instead, the developer world has slowly gotten over this mondo corporate brainwashing, turned back to the more sensible ideas - REST based XML web services - where you ask a question and get an answer using a URL (kind of like...the Internet!)
I think that today marks the beginning of the end of the end for SOAP, now that the Google SOAP API is finished (not that many folk were using it, anyway...) Whatever good ideas SOAP held at the beginning have been totally wrecked. I mean, the S in SOAP used to stand for SIMPLE. When they stopped using the acronym and tried to make out that it was 'just a name', alarm bells should have gone off all over the world...
Can we all just put it behind us and move on now?
The SOAP concept grew out of a great simple idea. The Idea was this:
"wouldn't it be nice to do RPC stuff via XML?
The first iteration of what would become SOAP was pretty straightforward - XMLRPC. Version 1.0 of SOAP built upon that, and was still human readable, and okay. Som
etime after that, the big boys like Microsoft, IBM and Sun jumped on board, and proceeded to complexify, 'standardize' and completely bugger the specification about until the final modern iteration of SOAP was so confusing, that only the most zeal-ridden platform zealots were singing it's praises - and even they didn't actually know how the thing worked anymore.Marketing execs all over the world were so desperate for 'Web Services' that they confused them with 'Web Sites', 'Web Parts', and 'Windows Services'. The only thing spreading faster than misinformation was glossy brochures. The whole thing was deranged.
The original, simple idea was lost in a maelstrom of add-ons and transactional plug-ins and security goo and it all became horrible. Actively trying to build a solution architecture on top of this thing is something that no self-respecting architect would do. Instead, the developer world has slowly gotten over this mondo corporate brainwashing, turned back to the more sensible ideas - REST based XML web services - where you ask a question and get an answer using a URL (kind of like...the Internet!)
I think that today marks the beginning of the end of the end for SOAP, now that the Google SOAP API is finished (not that many folk were using it, anyway...) Whatever good ideas SOAP held at the beginning have been totally wrecked. I mean, the S in SOAP used to stand for SIMPLE. When they stopped using the acronym and tried to make out that it was 'just a name', alarm bells should have gone off all over the world...
Can we all just put it behind us and move on now?
Comments
Post a Comment