BitWorking

Web Services vs. Internet Services

I think a new name is in order, "Internet Services" to distinguish them from "Web Services".

First let's discuss Web Services. The first person who comes to mind for this definition is Jon Udell, who has been covering the idea of hooking up various services on the web for a long time, most recently covering how to transform an RSS feed into HTML fit for a Nokia which was done by Robert Ivanc, and how his bank could have used Web Services to fix his online account.

But since 1994 or so, things have been different. Integrating two Web-based systems, if only by brute force, is not only necessary but possible. If I can log in to both systems and drive them interactively, I can write a script to join them programmatically. Every Web application can be tortured into behaving like a Web service, even when its creator never intended it to. [Jon Udell]

To me that sums up Web Services, as a way to streamline such natural information flows.

Now, in contrast, I have been lurking on the W3C Web Services and the W3C Web Services Architecture Working Group mailing lists. As a side note, I think Mark Baker, no matter what you think of his arguments, definitely deserves a medal for persistence, as he tries to promote the REST architecture to the Web Services Architecture Working Group. Anyway, medals aside, I have been following the arguments and think that the term Web Services, and the actual goals of the majority of the people in the Working Group have little, or nothing, to do with the ideas Jon Udell is talking about. The group seems pretty centered on using SOAP via HTTP to connect big business databases together. As far as I can tell, this group would be much better off if they dropped HTTP and just started creating a BEEP profile that incorporated SOAP, which they could call "Internet Services" to differentiate themselves from "Web Services".

Now, back to Web Services, brings us to Robert Scoble, latest borg victim, is trying to answer the question "Is IE Dead" while not trying to break an NDA ends up asking what people would like to see in a new version of IE. (See the comments to this post.) Which got me to thinking, what is the least amount of innovation that would be necessary to get Jon Udell's concept of Web Services up and running. That is, what's the minimum amount of tweaking to the browser that could be made so that services could be designed to be used from a browser, or programmatically. Here's the short list of changes:

  1. Add the ability for HTML forms to submit their data as "text/xml" in addition to "application/x-www-form-urlencoded". This would be a half-way step between current HTML forms and XForms, which is only supposed to make an appearance in XHTML 2.0.
  2. Also, have the optional ability for form values to be populated via XML. Now it looks like what I am asking for is XForms here, but I've looked at XForms and it is way too complex and too radical a departure from current HTML forms that it won't be adopted anytime soon. There was no effort to leverage current developer skills, nor any effort to minimize breakage when migrating from HTML forms to XForms.

That is the minimal list, if I were to go for the whole enchilada, I would also want:

  1. A native form element for editing HTML. A long time request of Dave Winer.
  2. The ability for HTML forms to use the "PUT" and "DELETE" methods in addition to the current "POST" and "GET" methods.

That's it, those are all the changes that would be needed to support Jon Udell's concept of Web Services. These changes would make it dramatically easier to create Web Services that could be used either by a browser or programmatically. No SOAP, no WS-Security, no InfoPath, no XForms. Unfortunately, this has a zero percent chance of being implemented. In a way I wish we we're back in the days of the browser wars, at least back then there's a chance at least one of the competitors would implement these ideas just on the hope it would give them an edge. Now Microsoft dominates the browser scene and is unlikely to try anything innovative, and the W3C has moved well beyond this level of tinkering and is dealing with bigger and better things like RDF, XForms and XHTML 2.0.

Posted by DJ on 2003-05-28

(Sorry, finger/keyboard trouble on previous attempt). Some of your thoughts on 'web' vs 'internet' reflect very similar thoughts I had too. Interestingly, it was on the cusp of me discovering REST which brought renewed focus (for me) on HTTP.

Posted by DJ on 2003-05-28

2003-05-28