OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help



   Re: [xml-dev] Can you stand yet another SOAP-RPC vs HTTP GET question?

[ Lists Home | Date Index | Thread Index ]

On Saturday 20 April 2002 04:08, Mike Champion wrote:
> I asked the Google Web API people why they didn't just offer a simple HTTP
> GET of an XML result.   The (FAQ, I think) response was: "We chose to
> deliver Google Web APIs via SOAP because we believe that the SOAP developer
> tools make Google Web APIs accessible to a broad developer community."  OK,
> fair enough ... if we were talking about some complicated interface.  But
> we're talking about generating very simple URIs for 99% of what people do
> with Google.
> Every language I use has a URL library and a string manipulation library,
> and it would be easy to take an arbitrary query string, generate such a
> URI, do a GET on it, and parse the XML response.  Wrap up this logic in a
> simple API! How much broader a developer community could they want?.

That rant (or semi-rant, depending on the tone) has been around the net [1] 
for a few days and imho isn't about to stop too soon as more widely visible 
web services appear. 

But I don't think that those, such as you and I, that would rather have an 
XML document rather than a more or less baroque SOAP API will see their 
voices heard.

The fact is that the programming world is driven by mediocrity. Genuine 
mediocrity, ie not actual poorness but simply code of low or moderate 
quality. As it is, most programmers are not library or architecture creators 
but, irrespective of their language of choice, scripters. They simply import 
whatever they need, create an object, call a few methods, and are done. 
Seeing the world through those eyes, one might easily conceive that it's 
actually easier to jump through SOAP hoops rather than having to throw in a 
trivial SAX handler to gather the data you need from an URL. It's more 
conform to habits.

I'm not denigrating anyone, and in fact this approach could be better in the 
very same way that some languages are sometimes considered better because 
they let you do less. We certainly wouldn't mind being able to escape such 
constraints when the need arises, no more than we'd mind being able to come 
up with our own simpler ways, but that won't sell to many "decision makers".

Of course there are other factors, such as hype (an XML-enabled app is 
nowhere near as hypable as a SOAP-enabled app these days), but I think that 
the fact that many would rather take the cosy API route rather than try to 
extract information from a document plays a role, and not a minor one.

[1] Quick example with comments amongst many others: 

Robin Berjon <robin@knowscape.com>

The more you run over a dog, the flatter it gets.


News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS