Re: AW: Show me the money - (was Subjects as Literals) +1 for Frank's comments

On 3 Jul 2010, at 09:40, Chris Bizer wrote:

> I cannot agree more with what Frank has said. 
> 
> I think that this working group can be very dangerous and destructive if it
> is developing into a talk shop for ivory tower people.

I don't agree that this is necessarily ivory tower people, btw. And I am certainly
not one of the ivory tower ones, as you know.

But I don't think one should underestimate the importance of conceptual beauty. You
may say, that is difficult to judge, but one of the authors of the spec, and Tim 
Berners Lee with N3, and a few others have found things like literals in subject
position to be useful and beautiful. So those are not words to put aside lightly. 
And in matters intellectual we are not in a democracy I fear - that is if democracy
is understood as the rule of the majority. I am pretty sure you subscribe to this view.

Making this change could end up simplifying the work at all other levels, by reducing
complexity. 

If a great Go master, say Go Sei Gen a Chinese expatriate who ruled Go in Japan for most 
of the post war years, said a move was beautiful, this was not something to be 
swept aside by his students as a quaint aesthetic statement, but is meant to attract the deepest
respect. 

This is the weekend so let us mediate a bit please. On this thread I found
that many of the reasons given against the change were mistaken. Perhaps the
one reason of concern may be how long it would take to do such a change.

> 
> Thus I think it is very critical to keep the duration and scope of the
> working group as small as possible.

Perhaps it would be good then to ask how long such a change would take,
and if the repercussions of it, could make other changes a lot easier too.
But also I don't see what the hurry is. RDF/XML works well as it is now, so 
why does this working group have to do things in 1 year?  What would it help
to do these small changes? 

> 
> Concerning scope, I think that it is critical that the working group does
> not try to develop/pioneer any new technologies, as recent W3C working
> groups tried with mixed results. Thus I think the working group should only
> do two things:
> 
> 1. tidy up the specs and delete all stuff that has not been widely adopted
> by the community over the last years
> 2. give an official blessing to widely used RDF data model extensions

What are you thinking of here? Named Graphs?

> and widely used alternative syntaxes.

So you mean turtle I suppose.

> Everything else might produce more harm than blessing.

Certainly if done in one year. On the other hand I think it would
clearly be beneficial to look at standardising N3, and this will require
allowing literals as subjects. Also it was pointed out that SPARQL could
benefit from such a cleanup, and that rules languages and OWL too.


> Concerning the duration of the working group, I think that the group should
> only work for 1 year. With any longer duration, the harm done by the
> unstableness resulting from the working group will be bigger than the
> benefits of having tidied-up specs. 


> 
> Also having such a short duration reduces the potential scope of the working
> group and helps to stick to the things that really matter.
> 
> Just my 2 cents,
> 
> Chris
> 
> 
> 
> -----Ursprüngliche Nachricht-----
> Von: semantic-web-request@w3.org [mailto:semantic-web-request@w3.org] Im
> Auftrag von Frank van Harmelen
> Gesendet: Freitag, 2. Juli 2010 22:32
> An: semantic-web@w3.org
> Betreff: Re: Show me the money - (was Subjects as Literals)
> 
> As someone who wasn't at the workshop, but who has been following it
> closely, I'm amazed by the lack of social intelligence in the debate.
> 
> Not all the worlds' problems can be solved by writing more specifications,
> and getting Linked Open Data widely adopted is an example.
> 
> Yes, there are some useful additions & changes to be made to RDF that have
> real use-cases screaming for them (and people already implementing because 
> they need them).
> The top 7 at [1] is a good list of these,
> and for all the other items on that list (including "literals as subjects", 
> c'mon!), social intelligence should prevail over technical arguments, no 
> matter how correct they are.
> 
> I'm in full agreement with Richard Cyganiak, Dan Brickley, Ian Davis, 
> Benjamin Nowack and others, summed up by the following quotes from 
> different messages in this thread:
> 
> Benjamin Nowack <bnowack@semsol.com>
>>> Our problem is not lack of features (native literal subjects? c'mon!). 
>>> It is identifying the individual user stories in our broad community and
>>> marketing respective solution bundles. The RDFa and LOD folks have
>>> demonstrated that this is possible.
> 
> Richard Cyganiak <richard@cyganiak.de>
>> Quite right.
>> 
>> But telling those user stories and marketing the solution bundles is  
>> not something that can realistically be done via the medium of *specs*.
> 
> Benjamin Nowack <bnowack@semsol.com>
>> We suffer from spec obesity, badly.
> ..
>> RDF "Next Steps" should be all about scoped learning material and 
>> deployment. There were several workshop submissions (e.g. by Jeremy, 
>> Lee, and Richard) that mentioned this issue, but the workshop outcome 
>> seems to be purely technical. Too bad.
> 
> Dan Brickley <danbri@danbri.org>
>> Spend the money on a W3C-license javascript SPARQL engine,
>> or on fixing and documenting and test suiting what's out there
>> already. And whatever's left on rewriting it in Ruby, Scale, Lua ...
> 
> Ian Davis <lists@iandavis.com>
>> In my opinion the time for this kind of low level
>> change was back in 2000/2001 not after ten years of investment and
>> deployment. Right now the focus is rightly on adoption and fiddling
>> with the fundamentals will scare off the early majority for another 5
>> years. 
> 
> As much as I admire Pat <phayes@ihmc.us> I couldn't disagree more with his:
> 
>> But after reading the results of the straw poll [1], part of me
>> wants to completely forget about RDF,  never think about an ontology or a
>> logic ever again.
> 
> Pat, you may be technically correct, but I think you are socially 
> completely wrong on this one. You/we have to choose between an imperfect 
> spec that's on its way to being widely used, or one that shines in splendid 
> isolation.
> 
> Frank.
>    ----
> 
> [1] http://www.w3.org/2010/06/rdf-work-items/table
> 
> 
> -- 
> Frank.van.Harmelen@cs.vu.nl		http://www.cs.vu.nl/~frankh
> Working on the Large Knowledge Collider	http://www.LarKC.eu
> 
>

Received on Saturday, 3 July 2010 08:08:55 UTC