W3C home > Mailing lists > Public > public-rdf-wg@w3.org > April 2011

Re: Problems with the RDF Semantics document

From: Nathan <nathan@webr3.org>
Date: Tue, 19 Apr 2011 01:25:32 +0100
Message-ID: <4DACD67C.9070104@webr3.org>
To: RDF Working Group WG <public-rdf-wg@w3.org>
CC: Richard Cyganiak <richard@cyganiak.de>, Ivan Herman <ivan@w3.org>, Dan Brickley <danbri@danbri.org>, Pat Hayes <phayes@ihmc.us>, David Wood <david.wood@talis.com>, Tim Berners-Lee <timbl@w3.org>, Manu Sporny <msporny@digitalbazaar.com>, Sandro Hawke <sandro@w3.org>
Richard Cyganiak wrote:
> On 19 Apr 2011, at 00:24, Nathan wrote:
>> just when are all these valid points and concerns and lessons learned going to be merged together in to an "RDF 2" like set of specifications
> 
> Well Nathan, if W3C is dropping the ball and not listening to the user community, then we can always fork and call it RDF5 ...

LOL, seriously that *cannot* happen again, does nobody any good at all. 
But, it could easily be close, I can't be the only one feelign these 
frustrations. Pretty much every member of this WG who I've spoken to, 
read blogs and mails from, or seen commenting on things seems to want 
quite a bit more than we're chartered to do. But, like myself will also 
just stick to the charter and do what is deemed to be needed.

In the interests of transparency though, here's a portion of a mail I 
sent to TimBL yesterday asking for advice on this very subject.

[[
I've spent pretty much all my time from the last couple of years, on sem 
web, web and w3 stuff, in order to get as full an understanding as I can 
and to help out wherever I can - much of my time is now spent between 
numerous WGs [1], as I say in my profile now "I'm a developer who sits 
at the intersection of the Semantic Web, WebArch, HTML, WebApps, 
REST/HTTP, JS, Auth/Security and Web Development communities, my primary 
focus is converging the various web technologies in order to realize the 
read write web of linked data, intelligent agents, and of course to aid 
the development of the social side of the web".

I'm itching to get on with pretty much everything below "deployment in 
library management" in the sem-seed diagram 
http://www.w3.org/2010/roadmap/sem-seed.svg and have done as much 
thinking, planning and background research as I can on the topics - I 
very much see the sem web as an ever growing tree with many branches 
that are both brilliant and useful - however, now that much of my time 
is spent down at the trunk, for instance in certain WGs, I'm very much 
feeling like this http://s3.amazonaws.com/image-sharer.com/images/EAYyV.jpg

To the problem, basically I feel unproductive and like the sem web and 
linked data isn't being brought to the masses, and that essentially a 
set of specs for something like N3 + LBase + RIF at the core needs done, 
that a chunk of owl 2 and rdfs need merged in to a single spec, and then 
perhaps more importantly that a subset of that core, namely ground 
graphs w/ no b-nodes, needs defined, as linked data (so that diff, 
merge, union, set operations, signing and encryption etc can all be done 
v easily), an API provided for the subset, design patterns for doing 
read write web over http with it, a JSON serialization for it, and a 
nice merge of JSON Schema with the OWL2+RDFs merge. The latter of which 
could be used for linked data validation and for which mappings could be 
provided to common rdbms, nosql stores and other output formats (csv etc).

So the problem exactly is:
  - I don't think the w3c WGs that are existing will go anywhere near 
any of this, as in, a fresh start is needed?
  - I know there are others who feel the same about key parts and who 
may help, but will they?
  - I know the web is an open platform, and we can build whatever we 
want on top of it, but it feels like it should be done within w3 space / 
lists / community group or something, as it really ties in with other 
groups (WebID, federated social web, rdf web applications, linked data).

Now I'm at a point where I feel like I just need to do this, that the 
time it is needed is now, and want to do it, I don't know where to start 
(not spec wise, place wise, on my own? in a group? get a new mailing 
list?) - I'm very aware that even any minor feature of these specs that 
could be discussed on any existing lists would lead to hysteria and mass 
panic that a change to the locked in stone RDF model may be being 
discussed, and tbh don't feel that would do anybody any good.

Sure you get the picture by now, I'll stop rambling, any comments or 
advice as to what I should do next, how I should proceed, or whether 
it's a stupid idea, greatly welcomed.

]]

So back to the subject, I could quite easily pick out almost every 
member of this WG and point to what they want from RDF 2 and what they'd 
like to see done (you've all written in blog posts, mails, irc logs etc 
in papers for the next steps meeting last year), likewise members of the 
last WG (it's all there in the history, same things for nigh on 10 
years), and likewise the broader communities and people on the edge of 
this group, especially around linked data and owl (just look at the 
common themes that come up and where the many different topics intersect).

I must stress, I'm not saying this on behalf of myself, and these aren't 
my ideas, they're just what you guys in the semantic web have shown me 
over the years and said, what you've discussed with each other, I'm just 
relaying them all back to you in one. So, with the last shred of 
wet-behind-the-ears-ness and new-ness I write this mail to you all (and 
please don't hold it against me), because it feels to me at least that 
the iron is hot and ready to be struck, goodness knows the extended 
communities we're all familiar with (general web, web apps, social web, 
web of trust, poi and loads more) all need it.

So, yes it appears that RDF 1.0 needs to be cleaned up to RDF 1.1 and 
this work must stay the course and keep full bc. But out-with that 
there's a job that needs to be done, and the people here together with 
there extended associates, friends and communities, are most definitely 
the people to do it - the quoted mail I sent earlier to Tim is a pipe 
dream, I'm one person and can't do all that, and if (nay when) I try 
I'll only be pestering you all any way) - so what happens now? Do we 
jump on this, do I get outcast, do I move to bad standing within the 
community and just stay silent offering what bits of help I can, or?

That's a direct question for you all - this is a publicly logged list, 
you're all people that are part of the core semantic web community, and 
I need to know whether I'm just a crazy fool on the outskirts with a 
pipe dream or whether I'm reading this right and there's a genuine urge 
to just say "let's do it" and move this semantic web forwards a few years.

Think I'll run and hide for a couple of days now, with my heart on my 
sleeve, Best,

Nathan
Received on Tuesday, 19 April 2011 00:26:18 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:25:41 GMT