Re: Ed's / Ian's View

Jonathan Rees wrote:
> On Wed, Apr 6, 2011 at 9:09 AM, David Booth <david@dbooth.org> wrote:
>> On Tue, 2011-04-05 at 22:20 +0100, Nathan wrote:
> 
>>> The day you can say the above for *all* applications (given this
>>> scenario of some name referring to two things) rather than some, is the
>>> day when it'll be an acceptable solution, until then it, like most of
>>> the other proposals, only addresses *some* situations. We need a for all
>>> here.
>> Agreed.  Once a pool has been contaminated it's very hard to clean it
>> up.  This is why in the draft of section 5.5 that I sent I included
>> other scenarios that discuss how to *prevent* that contamination, even
>> when you do want to merge graphs from different sources.
> 
> I think you're both missing the point of the issue-57 document. The
> purpose is to *start* a conversation. The point is neither to present
> a complete, sound version of the proposals (that burden would be on
> the people who like them), nor to provide exhaustive criticism. It
> only needs to say: Someone has proposed to do X. Here are the some of
> the consequences of X, secondary problems that would need to be
> addressed somehow (later). Then those reading the document can either
> decide for themselves, or engage in further discussion. There will be
> a followon document that describes the consensus, and all the holes
> will have to be plugged before anything gets into it. But we don't
> need to plug them now, OR prove them unpluggable.

quick confirm: I do see the purpose of the issue-57 document just as you 
described, but since I'm also one of the ones looking for the solution 
and the conversation that will follow, some of that conversation spills 
on to this list as a bi-product of discussing the document.

and on that note, I received another use-case yesterday from Melvin 
Carvalho:

[[
   Joe has used:

   creator <http://graph.facebook.com/1234>

Nathan, would love to know your thoughts on this.  I think the above
is a document and not a person.  But do you think can we allow some
leeway here and introduce some fault tolerance?
]]

in regards to http://data.fm/ (a RWW datawiki) which is made by Joe 
(Presbrey), from MIT/DIG. You can click the graph.facebook uri to see 
the data at the other side. This is/will be a common style of usecase, 
"didn't know" - and one that increasingly leads me to believe the Ed use 
case of knowing that we're talking about a Person(<u>) from the creator 
property, is one we may end up having to focus on (as a community) - so 
def needs properly noted in the document, and not tied in with a 
particular solution or approach with certain restrictions such that 
people don't see it as being a viable option.

The clearer we say each problem and the more general the proposed 
solution we document, the better - imho. (i.e. I'd like section 5.5/5.6 
to be left general, as it's rather important, and not get in to a 
heavily technical proposal there).

Best,

Nathan

Received on Saturday, 9 April 2011 22:28:50 UTC