[Standards] Re: Feedback on I/O Object Forms

aliban at gmx.net aliban at gmx.net
Tue Mar 27 19:44:29 UTC 2007


Peter Saint-Andre schrieb:
> Alexander Gnauck wrote:
>> Alexander Gnauck schrieb:
>>> my understanding of the XEP is that they also want to route existing 
>>> XML data structures without converting them back and forth to/from 
>>> xdata.
>>> They also have some complex data structures which will not fit into 
>>> xdata forms. I think it would be easier if there is a real world 
>>> example which we can understand without having a diploma in medicine 
>>> or biology.
>>
>> i think also Object Forms is a confusing name for this XEP.
>> The authors work here more with very complex Objects which are not 
>> 'flat' like x-data forms.
>
> Agreed.
>
> I also agree with Alex that it would be good for this spec to include 
> more examples, especially examples from a domain other than 
> bioinformatics.
>
> Remember, we put the example in example.com :)
>
> Peter
>

Don't worry, we will generalize the examples.

The XEP will describe a generalized way to (1) discover and (2) execute 
a remote function in a way similar to the principles of SOAP based Web 
Services. As a special feature it will also describe a way for the 
server side (the Web Service) to (3) notify the client about completion 
of the job. In contrast to current http SOAP based Web Services the 
server side (the Web Service) can actively push data (the notification 
about completion or error) to the client. In addition to that it 
describes a way for the client to (4) keep track of the job and the 
possibility to (5) stop/cancel a currently executing job.

As Alex said, a flexible "Web Service over XMPP" requires a not 'flat' 
data object.

Edrin



More information about the Standards mailing list