Re: RDF-ISSUE-5 (Graph Literals): Should we define Graph Literal datatypes? [RDF Graphs]

On 2011-03-05, at 16:18, Nathan wrote:

> Richard Cyganiak wrote:
>> On 4 Mar 2011, at 23:45, Nathan wrote:
>>>> RDF-ISSUE-5 (Graph Literals): Should we define Graph Literal datatypes? [RDF Graphs]
>>> yes, a main reason being, if we don't, somebody else will, and possibly a few different people, which would lead to interop problems and require standardization in the future. So may as well get there first.
>> I'm sorry Nathan. This has to be the worst advice I've ever heard regarding standardization.
> 
> surely it can't be the /worst/ advice, I've definitely heard worse :p
> 
>> If there's no interoperability problem yet, then there is nothing to standardize, and a standardization group has no business concerning itself with the matter. What you're talking about is R&D, and standardization groups are the worst possible place for that.
> 
> normally I'd agree, however RDF is seen as being in the domain/control of a standardization body, people haven't added graph literals to turtle, because that wouldn't be turtle, wouldn't be handled by RDF, and thus is seen that they "can't" do it, not that they "don't want to" do it.

Note that Turtle didn't start out as a W3C standard, it was just an ad hoc agreement around a type of format that a bunch of us wanted, and Dave wrote a grammar for. Since then it's been written up a semi formal note, and this group will bless some variant of it, but it evolved a fair bit first.  

> If there was a way to easily put a chunk of RDF in to a graph and talk about it, people would do it - likewise chunks of HTML and chunks of JSON. People get data from the web and people want to be able to wrap that data up and strap meta data to it, without the indirection names provide.

Implementors have added things that are properly hard / semantically nasty as extensions to RDF, e.g. property functions. I don't think graphs-as-literals would cause anyone to lose sleep — if there was demand for them.

- Steve

-- 
Steve Harris, CTO, Garlik Limited
1-3 Halford Road, Richmond, TW10 6AW, UK
+44 20 8439 8203  http://www.garlik.com/
Registered in England and Wales 535 7233 VAT # 849 0517 11
Registered office: Thames House, Portsmouth Road, Esher, Surrey, KT10 9AD

Received on Saturday, 5 March 2011 23:15:43 UTC