Doug,

While a web server is assembling content from potentially multiple distributed web services, it could ask for just the data (sent as XML) or an HTML representation of the data (which is itself a form of data) -- it all depends on what the web service offers.

Many of us doing web development now are working with middle-tier classes that can query a data source or data service according to parameters sent in with the call, and then return the data as an ADO recordset, an XML stream, or maybe an HTML table -- depending on how we want to use it.

Web Services could be constructed to do exactly the same thing, but with the call to the service done with SOAP (XML over HTTP).

>The conceptual thing I was trying to get my head around was the separation of data content from data transportation . That's all, really. Having said that I happen to think this is a very big deal, as the three tier model just seems so robust.
>
>Being able to combine data from multiple web sites is IMO a direct result of being able to go get just the data.
>
>My guess is that one could still display a web page as the author(s) intended as well.
>
>We sure do live in a fast-changing development world... <g>
>
>Best,
>
>DD
>
>>Doug,
>>
>>>Web Services extends the concept of the separation of data from user interface to the Internet. Now, instead of web servers providing both content and formating with Web Services the web server may provide only content, in a universal format (XML) thus allowing the recipient of this data to format it as they please.
>>
>>And your web pages may be pulling that content from multiple web services, or from a web server that itself is pulling various parts in from other servers via web services.
>>
>>In that regard, our web pages may be constructed more along the lines of portal pages, which typically pull news, weather, etc from various places.