Re: Issue-34 Back_to_Basics proposal

Hi,

On 08/02/13 16:17, John Arwe wrote:
>> 3. One class of resource which contains either members or links to
>> members. When a container is
>> deleted all its contents are deleted. You use links to get aggregator
>> behavior.
>>
>> You are arguing for 1. correct? I thought the WG was moving towards 3.
>>> Not sure if we all agree, but at least IMO it is the most convenient
>>> solution to ISSUE-34.
>>
>> I mean, I vote for option 3.
>
> Putting aside for the moment that the only fleshed-out proposal that might
> correspond to 3 (I think) is Henry's... but from Ashok's email alone I
> cannot be sure if that was *his* intent, so I cannot be 100% sure what
> you're voting for with any specificity,

True, Ashok summarized the options, but actually he has not opted for any.

> when you say
>>> ... it is the most convenient
>
> that statement implies that you have some utility function as input to the
> optimization problem.  Could you share your utility function's inputs
> and/or construction?  I.e. along the lines of EricP's suggestion about
> IIRC issue 37, what criteria are you using to evaluate its convenience?

My criteria is that I think is much more convenient to have a default 
model, in this case a compositional model, that could be use by default. 
In addition, when necessary, have some aggregated members is fine since 
it does not mean to change the default behaviour. IMO, opting for 1 or 2 
would mean an overhead on the protocol for LDPC creation (which btw is 
something missing in the current spec).

> We might not all equally weight convenience for spec editors, web clients,
> pick your favorite audience(s).  Indeed, the audiences with non-zero
> weight are implicitly part of the question.

Maybe that's the key of the lack of consensus: each of us we approach 
LDP from our scenario with its specific needs. And of course that would 
need to be considered on the argumentation for each discussion.

Best,

-- 
Sergio Fernández
Salzburg Research
+43 662 2288 318
Jakob-Haringer Strasse 5/II
A-5020 Salzburg (Austria)
http://www.salzburgresearch.at

Received on Monday, 11 February 2013 08:18:57 UTC