Re: [ALL] [RDF] Commons RDF approach

[Sergio Fernández <wi...@apache.org>]

Hi,

On 30/01/15 23:15, Reto Gmür wrote:
> On Wed, Jan 28, 2015 at 4:22 PM, Sergio Fernández <wi...@apache.org> wrote:
>> Summarizing, we understand that the Apache Commons project wants to keep
>> the communication rules as they currently are. Though we think that in this
>> phase of the Commons RDF project, which focuses on the API design and
>> actively involves the developers of existing toolkits, it is better to have
>> a more focused community and infrastructure. Therefore we have come to the
>> conclusion that incubation is probably the best path, and then gradually
>> prepare the Commons RDF community for working within the larger Apache
>> Commons community. So we invite everybody to join the project in case you
>> are interested, particularly the current module at sandbox.
>>
> Could you give a link to the discussion of the "we" that came to this
> conclusion?

Well, just read the threads. I'd prefer a "we" including you, but of 
course I can only talk from the Commons RDF at GitHib folks. If do not 
agree on that, please just put your arguments on the table, and we'll be 
willing to discuss them.

> In my understanding there is the possibility of using the commons sandbox.
> And we have two proposal for rdf-commons one coming from clerezza and the
> other from the community on github. I also see some interests on both sides
> two align the proposals. The proposal in the sandbox svn is a first attempt
> of such an alignment.

Of course both project could co-live. But then at some point the Commons 
PMC would have to take a decision between both components' proposals: 
the one coming out from Incubator and the one from sandbox.

I'm more of the opinion that collaboration is the key for achieving the 
challenge of having a Commons RDF component. But of course you're free 
of following your personal path.

> I've no fundamental objection as having a new project with a narrower focus
> than clerezza in the incubator rather than in the commons sandbox. The
> crucial is that your incubation proposal mentions exclusively the github
> code as starting point. At the clerezza project we have been following the
> goal of a common RDF API since incubation as the first goal.

Because the main goal is the API itself, and that's we want to initially 
contribute, having a clean codebase, both from the technical and IP 
point of view.

Personally I've always seen Jena and Sesame as the real implementations. 
But for sure the in-memory implementations is one of the points in the 
agenda for the project. But step by step, please.

> We have
> addressed issues that are not yet fully addressed in the github proposal
> (see the other discussion on blank nodes) and we have been working as an
> apache community for more than 5 years.

Please, take aside the blank nodes discussion for now. Such single 
technical detail is something would be solved later, as soon as the 
scope of the project becomes clearer. Giving such level of importance 
does not allow you to see the forest for the trees.

> I'm happy to participate in this project but I would appreciate a more
> neutral starting point.

I could say the same about neutrality... The current Commons RDF at 
GitHub is the result of several discussions with the tool major Java 
toolkits, while I can currently see at sandbox is just your personal 
design out of the Clerezza adapters.

Then, I have to ask, should I include you in the proposal? See the 
current draft at https://wiki.apache.org/incubator/CommonsRDF which it's 
a draft we can discuss together.

Best,

-- 
Sergio Fernández
Partner Technology Manager
Redlink GmbH
m: +43 660 2747 925
e: sergio.fernandez@redlink.co
w: http://redlink.co

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org

Re: [ALL] [RDF] Commons RDF approach

[Sergio Fernández <wi...@apache.org>]

Hi,

On 03/02/15 18:55, Reto Gmür wrote:
>> Personally I've always seen Jena and Sesame as the real implementations.
>
> They sure are two of the best and most popular triplestore arounds,
> nevertheless an API should map the standards to Java, not concrete
> implementations. That's what clerezza wants to do, and if that's your goal
> we should join our forces.

Well, for me the main motivation behind this simple library is revise an 
historical incompatibility issue in the Java world. In particular, 
commons RDF aims to provide a type-safe, non-general API that covers RDF 
1.1.

The simple implementation in-memory is something that has been 
discussed, and we already have a prototype. So I see Clerezza providing 
another implementation, just like Jena or Sesame will do, that would be 
great indeed.

>> Please, take aside the blank nodes discussion for now. Such single
>> technical detail is something would be solved later, as soon as the scope
>> of the project becomes clearer. Giving such level of importance does not
>> allow you to see the forest for the trees.
>
> I think it is the most important difference between the two proposed APIs.

Then it's a good point for further discussion, as many other aspects 
will be.

>> I could say the same about neutrality... The current Commons RDF at GitHub
>> is the result of several discussions with the two major Java toolkits.
>
> The API is the API in use in clerezza for >5 years, with some changes we
> discussed at clerezza to accommodate the changes that came with RDF 1.1 and
> with some renamings to make the names more intuitive.

Well, we cannot compare the usage of that API in comparison with Jena or 
Sesame, we have to be honest about that. We designed the API based on 
the experience and backgrounds from developers of both libraries (Andy 
and Peter, I'm just a 'facilitator'), but I'm sure yours is valuable too.

>> Then, I have to ask, should I include you in the proposal?
>
> Sure, add me to the list.

Nice to have you on board! Just added your name to the current draft at 
Incubator's wiki: https://wiki.apache.org/incubator/CommonsRDF

Cheers,

-- 
Sergio Fernández
Partner Technology Manager
Redlink GmbH
m: +43 660 2747 925
e: sergio.fernandez@redlink.co
w: http://redlink.co

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org

Re: [ALL] [RDF] Commons RDF approach

[Reto Gmür <re...@apache.org>]

On Mon, Feb 2, 2015 at 8:20 AM, Sergio Fernández <wi...@apache.org> wrote:

> Hi,
>
> On 30/01/15 23:15, Reto Gmür wrote:
>
>> On Wed, Jan 28, 2015 at 4:22 PM, Sergio Fernández <wi...@apache.org>
>> wrote:
>>
>>> Summarizing, we understand that the Apache Commons project wants to keep
>>> the communication rules as they currently are. Though we think that in
>>> this
>>> phase of the Commons RDF project, which focuses on the API design and
>>> actively involves the developers of existing toolkits, it is better to
>>> have
>>> a more focused community and infrastructure. Therefore we have come to
>>> the
>>> conclusion that incubation is probably the best path, and then gradually
>>> prepare the Commons RDF community for working within the larger Apache
>>> Commons community. So we invite everybody to join the project in case you
>>> are interested, particularly the current module at sandbox.
>>>
>>>  Could you give a link to the discussion of the "we" that came to this
>> conclusion?
>>
>
> Well, just read the threads. I'd prefer a "we" including you, but of
> course I can only talk from the Commons RDF at GitHib folks. If do not
> agree on that, please just put your arguments on the table, and we'll be
> willing to discuss them.
>
>  In my understanding there is the possibility of using the commons sandbox.
>> And we have two proposal for rdf-commons one coming from clerezza and the
>> other from the community on github. I also see some interests on both
>> sides
>> two align the proposals. The proposal in the sandbox svn is a first
>> attempt
>> of such an alignment.
>>
>
> Of course both project could co-live. But then at some point the Commons
> PMC would have to take a decision between both components' proposals: the
> one coming out from Incubator and the one from sandbox.
>
> I'm more of the opinion that collaboration is the key for achieving the
> challenge of having a Commons RDF component. But of course you're free of
> following your personal path.
>
>  I've no fundamental objection as having a new project with a narrower
>> focus
>> than clerezza in the incubator rather than in the commons sandbox. The
>> crucial is that your incubation proposal mentions exclusively the github
>> code as starting point. At the clerezza project we have been following the
>> goal of a common RDF API since incubation as the first goal.
>>
>
> Because the main goal is the API itself, and that's we want to initially
> contribute, having a clean codebase, both from the technical and IP point
> of view.
>

> Personally I've always seen Jena and Sesame as the real implementations.


They sure are two of the best and most popular triplestore arounds,
nevertheless an API should map the standards to Java, not concrete
implementations. That's what clerezza wants to do, and if that's your goal
we should join our forces.


> But for sure the in-memory implementations is one of the points in the
> agenda for the project. But step by step, please.
>
Fully agree. Clerezza is NOT about providing inmemory imlementations.


>
>  We have
>> addressed issues that are not yet fully addressed in the github proposal
>> (see the other discussion on blank nodes) and we have been working as an
>> apache community for more than 5 years.
>>
>
> Please, take aside the blank nodes discussion for now. Such single
> technical detail is something would be solved later, as soon as the scope
> of the project becomes clearer. Giving such level of importance does not
> allow you to see the forest for the trees.
>

I think it is the most important difference between the two proposed APIs.
Or what else do you criticize in the clerezza core RDF API?




>
>  I'm happy to participate in this project but I would appreciate a more
>> neutral starting point.
>>
>
> I could say the same about neutrality... The current Commons RDF at GitHub
> is the result of several discussions with the tool major Java toolkits,
> while I can currently see at sandbox is just your personal design out of
> the Clerezza adapters.
>
Out of the clerezza adapters?

The API is the API in use in clerezza for >5 years, with some changes we
discussed at clerezza to accommodate the changes that came with RDF 1.1 and
with some renamings to make the names more intuitive. For the renaming
inspiration have also been taken from your proposal.

The fact that the clerezza project as well as third parties could provide
adapters for various triplestores exposing other APIs is an indication that
the API is suitable.



> Then, I have to ask, should I include you in the proposal? See the current
> draft at https://wiki.apache.org/incubator/CommonsRDF which it's a draft
> we can discuss together.
>

Sure, add me to the list. If there is a possibility we should avoid
duplication between clerezza and this project, I don't think the use cases
are so different that we need two APIs.


Cheers,
Reto

>
> Best,
>
>
> --
> Sergio Fernández
> Partner Technology Manager
> Redlink GmbH
> m: +43 660 2747 925
> e: sergio.fernandez@redlink.co
> w: http://redlink.co
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
> For additional commands, e-mail: dev-help@commons.apache.org
>
>