This is needed, in case of controlleed server error, need to notify the user with the message that comes in the body of the request, and currently there is no way to get it.

Refers to ticket: 1031709

Thanks

Telerik already makes it super simple to integrate (data bound) controls with databases.

I would like to see the same easy-of-use when it comes to Web-Services.

While I think the RADClientDataSource is exactly the component I would like to see for that, I would like Telerik to extend it by adding Wizards so that it is possible to do a GUI and integrate it with Web-Services without coding.

Long story:

Basically, many web-based applications use tools like Microsoft BizTalk to implement processes and workflows, talk to databases, implement business logic etc. 
Most such applications also need user input, so development teams often split two ways:
1. the guys who do the backend 
2. the guys who do the frontend

In such setting, it would be perfect if the backend guys could provide a bunch of Web-Services that provide the data for the GUI and that also receive the data from the GUI and can do something with it.

THat means that the GUI guys will not communicate with databases as much but use the Services instead - and should be able to integrate their GUI just as swiftly and ideally without coding as it can be done with the data-bound controls. 
After all, while the database schema describes the structure of the data a GUI can pull and the structure into which the GUI may write data, XSD's/WSDL's do the same for Services. So if these XSD's/WSDL's are the "interface" between the GUI guys and the backend guys, a code-free integration at least for common use cases should be easily possible.

And I think extending the RADClientDataSource with Wizards etc. would be the perfect way to do it. 

What do you guys think?

The RadClientDataSource is a wonderful control to greatly simplify binding controls to data for client side operations. 

I'm finding though that I don't have many scenario's where I can get the data in one swoop. An example is an event log viewer with 100000 logs that I don't want to transfer to the client in one transfer. Rather, with a custom date range that can be set via date pickers, I'd like to be able to tell the data source to go back and get fresh data for the new filter parameters. 

Other scenario's can be for contributions where in our app, people are part of teams and they can be part of multiple teams. Upon switching contexts, from one team to another, the data needs to be refreshed because there's many thousands of records so I cannot load everything at once on the first go, besides, that kills the network pipeline.

I guess what I'm getting at is that I find there is the need to have the ability to filter the data by passing parameters from controls on the client to filter data returned from a stored procedure expecting said parameters. It's really the only way to go when you have screens that are generic that deal with large datasets because any alternative will kill performance. 

Robin