Re: ldp-ISSUE-46 (services): services and LDP

El 24/01/13 20:23, Linked Data Platform (LDP) Working Group Issue 
Tracker escribió:
> ldp-ISSUE-46 (services): services and LDP
>
> http://www.w3.org/2012/ldp/track/issues/46
>
> Raised by: Roger Menday
> On product:
>
>
> As characterized by Arnaud:
>
> "In practice, I think there are two general categories of use cases. 1. generic/vanilla server  that simply stores triples and regurgitates them without doing anything special with them. 2. application specific server - this is a bug tracking system for instance - which translates the triples into an actual application specific object."
>
> It does seem that most people are looking at category 1 type applications.
>
> It seems that some of us are interested in category 2. We would like to have an issue so that we can monitor the parts needed for category 2, track what we won't cover in LDP, etc.
>
> Then how do we cover the missing parts ? Some options ::
>
> * wait and see what others in the community (or this community do with it)
> * encourage an existing group (i.e. Networked Data) to take this on
> * put some documentation on the wiki
> * have another top-level specification inside LDP
> * combination of above
> * ?
>
> [1] http://www.w3.org/community/networked-data/

Dear all,

In our case (the ALM iStack project), we are also more interested in the 
use of the LDP specification in applications than in the LDP protocol 
itself.

Hence, we prefer to differentiate between both things and use these 
definitions:

LDP server
.- Complies with the LDP protocol

Application
.- Has a concrete data model
.- Has a concrete business logic
.- Supports interaction

LDP enabled application
.- Is an application
.- Supports interaction using the LDP protocol (therefore, it is an LDP 
server).

Summarizing, the differences between an LDP server and an LDP enabled 
application are that the application has a concrete data model and 
business logic.

The LDP protocol should not take care of those things that are specific 
to LDP enabled applications (e.g., consistency checking, referential 
integrity). However, the protocol should not hinder them.

Anyway, we think that taking into account issues in the application 
level is out of the scope of the specification. Besides, it will only 
make discussions more confusing.

Kind regards,

--

Raúl, Nandana and Miguel

Received on Friday, 25 January 2013 13:25:11 UTC