Re: Creating JSON from RDF

On Monday 14. December 2009 11:17:37 Dan Brickley wrote:
> I wish that kind of funding was easy to come
> by, but it's not. A lot of the work we need to get done around here to
> speed up progress is pretty boring stuff. It's not cutting edge
> research, nor the core of a world-changing startup, nor a good topic
> for a phd.

I can attest to that. I've been searching for this kind of funding for half 
a year now, and I've failed. The RDF toolchain is in a terrible state, but 
I do not think the focus should be on parser, and that basic stuff. Even 
though native toolchains are missing for some languages, the things that 
are there, notably Redland with all its language bindings, are still good 
enough.

What's missing are the tools that are needed on the top of that to make the 
developers efficient, not just in publishing data, but more importantly 
using the data to provide groundbreaking services that are truly useful to 
a large number of people. I've been around for 11 years, albeit more to and 
from than danbri, but I have yet to see one semweb application that does 
this. Some come close, like DBPedia Mobile, it had a lot of potential, but 
it needed polish that it hasn't got so far.

One problem, is that there are some open questions, like how to use an 
ever-changing model in an existing Model View Controller framework. This 
was something I tried to address, as it should be the strength of RDF and 
something to distinguish us from the SQL world, but the business 
perspective says that this is research and the researchers say it is mere 
stamp collecting. I've been involved in research projects that have had it 
as their main goal to provide semweb frameworks for general consumption, 
that has spent huge amounts of money on research, but has totally failed to 
achieve their goal because they didn't acknowledge that this problem, 
however trivial it may seem, has not been solved, or at least, the solution 
isn't on PEAR, CPAN or in Debian.

That brings me to another, but smaller problem, where the tools exist, with 
the exception of Redland, they are not available, and/or have not gone 
through the QA process of for example Debian. It is extremely important for 
a developer to have quick access to the toolchain. It is therefore 
critical, I believe, that e.g. Virtuoso and 4store goes into Debian.

Due to my failure to find funding, I'm changing strategy, I'm leaving the 
semweb world, hopefully temporarily, to learn the tools that good 
developers are using. I believe that we're not going forward unless we can 
provide developer tools, and they are not there. 

Cheers,

Kjetil
-- 
Kjetil Kjernsmo
kjetil@kjernsmo.net
http://www.kjetil.kjernsmo.net/

Received on Monday, 14 December 2009 19:14:46 UTC