>One is online here, the other is available anytime. You could save the web page or copy and paste in a text file for offline viewing. Is that what you mean?
>

Everywhere that I develop, the web is available, so they are both available all the time.

>Again, I was confused thinking each Call was/could be changed but even with the reference growing - adding new Calls or more comments to an existing one, I was wondering if XML was a preferred method of providing data for a living, growing and ongoing data source

I don't see how it makes a difference.

>Won't P/S payoff in more ways than downloading data to web sites as per the example at NOLA DevCon? Can't it serve for updating data sources for other things? I guess I got to that with the following question

Publish subscribe is for when an event happens that needs to trigger something to happen in another process or machine. I don't see the relevance here. An online repository can be accessed by a client at any time, why would the client need to subscribe to an event?

>>You could do that, and I have apps that do that now, but I don't see the use here. You write a client side app when a browser doesn't give you the functionality you need. Again, what's the problem with the browser interface?
>
>I prefer a desktop GUI over a browser one myself whether online or off. Do you perfer browser interfaces?

They each have their place. Simple online interfaces are much easier to develop, you don't have to worry about distribution/installation/updates/compatibility or replicated data. For an information repository, I think that a desktop app accessing a local database is about the least efficient way to do it.