Re: Updated JSON-LD draft - no more microsyntaxes

On 05/16/2011 03:21 AM, Thomas Steiner wrote:
> Finally had the time to work through the new spec draft. Thank you
> very much for your hard work, this looks like something I was hoping
> for, and that I can perfectly live with. For what it's worth, I add a
> "+1" from my side under this draft. 

That's great!

> What would be the concrete next
> steps in order to push this forward in the RDF(a) WGs? What can I do?
> Please do let me know how I can help. 

An implementation of a JSON-LD parser in JavaScript would be great.
Hooking it up to Node.js would be even better.

Demonstrating how JSON-LD can be used in MongoDB or CouchDB would be
awesome as well.

A few of us are currently thinking about a test suite - we may setup the
test suite like we did the RDFa test suite - which would allow you to
plug in different processors. The validation aspect could be done in
SPARQL. However, that's just a first cut at an idea for a test suite -
there may be a better way to do this. So, help with figuring out how we
could do a test suite would be great.

Also - evangelizing JSON-LD inside of Google would be incredibly
helpful. See if you can find others within the company that have a need
for JSON objects that are first-class citizens on the Web.

We're not going to get this stuff back into RDF WG until we show
traction in the real world. In fact, even if we don't get this back into
RDF WG - if we have excellent traction in the real-world, the standard
becomes less important. We could follow the path that JSON took -
real-world success first, then become a standard. Some would say that is
how all standards should be developed.

So, let's not worry about the RDF WG right now - let's try to get as
many people using something that we can all agree on.

> Not sure if you had more
> feedback from people (I know of Gregg Kellog's email). Somehow no one
> seemed to react on your email, at least not on the JSON list.

I have had several discussions off-list, all positive. Some have asked
me to not mention RDF until the end. Some have asked me to get rid of
RDF entirely and focus more on the Linked Data aspect. So, if anyone has
opinions on that stuff, it would be good to air them out in public. I've
asked those that have contacted me to do so, and some of them have - but
some of them can't because they work for large companies and some of
them can't because they don't want to give the impression that they're
insulting their colleagues.

So, I'm getting feedback - but I wish people would give more feedback in
public.

I'd also like to see more feedback on ideas like Glenn has proposed on
the list. Thoughts on directions other than JSON-LD that we should take.
This list wasn't meant to just discuss JSON-LD - it was meant to discuss
any implementable Linked Data concept that we could use JSON to express.

Our goal here is to get as many people publishing semantic web data as
possible. However, we should also make sure to focus on things that are
concrete and that we think we can deliver within the next 6-9 months.
Just my $0.02 - I know others think differently and we should discuss
all of this out in the open - all suggestions, plans, comments and
especially working code/implementations are welcome.

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny)
President/CEO - Digital Bazaar, Inc.
blog: PaySwarm Developer Tools and Demo Released
http://digitalbazaar.com/2011/05/05/payswarm-sandbox/

Received on Thursday, 19 May 2011 04:08:37 UTC