Web Services - slower adoption
Mark Baker was right about Web Services last year:
I predicted that XMethods would list less than 400 services by today, and lo-and-behold, that's the case; they list 366.
He also has a challenge for web services promoters
So here's a challenge to Web services promoters; make a prediction about the number of Web services available on the Internet by the end of 2004 and/or 2005. If they're so great, then surely, at some point, there's going to be thousands of them, right? When will that be? At this rate, they won't get to 1000 until 2010 ... assuming the hype - which is the only thing keeping them even linear, IMO - lasts that long.
Here's the thing - Web Services are simply the latest incarnation of distributed services - we've had various proprietary RPC services, DCA, CORBA, DCOM, and now Web Services. Notice how all of them have been promoted (in their time) by the large consulting firms and the trade rags, and how all of them have been used by a relatively small number of projects. The plain truth is, not that many projects really need complex distribution mechanisms. Most apps look a lot like this:
- Read data from a store (db, files, etc)
- Modify data
- Dump data back to the store
Where in that mix do most people need complex distribution? That's right, nowhere...
Comments
XMethods poliics
[Mark Derricutt] December 31, 2003 16:56:01.286
I'm not sure how accurate this is these days, but several months ago when we tried to get our SOAP service listed on XMethods they refused to add us - simply because they already have a web-service listed that does the same thing ( send SMS messages ). Seems they had a policy to only list ONE web-service in a given area - which well, kinda sucks. Judging how many people make use of web-services just by XMethods, or other "open for the public consumption" probably isn't of much use. I know of quite a few people using web-services in their back-end systems for inter-process communication. One instance there the back-end code is all in PERL, and the front end all PHP. Others java backends, with PHP/JSP/ASP front ends. If you -are- dealing with multiple languages in your deployments ( for whatever reason ), having language neutral transports with a consistent API is handy. Esp. when using a lighter-weight XML-RPC servces that any man-and-his-dog can just throw at a server with a bit of string manipulation.
Re: Web Services - slower adoption
[Michael Lucas-Smith] December 31, 2003 17:21:44.851
Comment on Web Services - slower adoption by Michael Lucas-Smith
We're starting to use web services a great deal more now - more specifically SOAP a great deal more now. Web services, eg: WSDL - is the biggest over engineering I've seen in ages - and that's where people are coming unstuck. XML is best when left unstructured and XML-Schema is like strapping all the typed languages on your code at once. I very much doubt Wizard will ever publish to something like XMethods. We'll publish to our clients and thats it.
XMethods
[Mark Baker] December 31, 2003 21:06:38.095
Mark, thanks for the info, I didn't know that's how XMethods operates, but should have. I'll talk to the Hongs about it.
XMethods
[Mark Baker] January 7, 2004 16:44:30.030
Tony Hong says the only time they've ever said no to a service was either if the WSDL is invalid, or if it's an "echo service". So I'm not sure what your issue might have been, but they certainly allow duplicates; in fact, they already have lot of dups.