Re: A use case for @vocab?

François, I think you make good points. I re-opened the issue at [1]. We can continue the conversation on that issue.

Gregg

[1] https://github.com/json-ld/json-ld.org/issues/26

On May 24, 2012, at 10:11 AM, François Daoust wrote:

> Hi Markus, Josh, Gregg,
> 
> On Thu, May 24, 2012 at 6:38 PM, Markus Lanthaler
> <markus.lanthaler@gmx.net> wrote:
>>> I think François is advocating for `@vocab` for the same
>>> reason we have prefixes:  to avoid having to explicitly define
>>> *every* predicate with the same prefix. (It's not "built-in,"
>>> it's established in a context.)  The special thing about
>>> `@vocab` is to establish an *empty* prefix, with no colon.
>>> And this might sound trivial (?), but colons make property
>>> names really annoying to work with in JavaScript.
>>> 
>>> E.g. with @vocab the SMART context could be ~1k instead of 16k
>> 
>> Oh sorry.. then I completely misunderstood his mail.
> 
> Dammit. And I thought I was clear ;)
> Thanks for the clarification Josh, that is indeed what I meant. See my
> previous email.
> 
> 
>>>    "Alert": {
>>>         "@id": "http://smartplatforms.org/terms#Alert"
>> 
>> You know that you can use prefixes in the context as well!? Something like
>> 
>> {
>>  "@context": {
>>    "spt": "http://smartplatforms.org/terms#",
>>    "Alert": "spt:Alert",
>>    ...
>>  },
>>  .. data (not using prefix) ..
>> }
>> 
>> 
>> I'm still not convinced that @vocab would be a good idea.. there are many
>> ways to solve this in an application but of course it depends on the
>> application's specific requirements.
> 
> Hmm, I can only think of 2 main ways:
> 1. list properties in the @context.
> 2. parse the created JSON-LD before running the framing algorithm to
> create the @context on-the-fly; or use a specific implementation
> framing algorithm that makes certain assumptions on the @context
> 
> That's why I propose @vocab as a third possible way. Now I agree that
> introducing new ways of doing something is not necessarily a good
> thing.
> 
> Best,
> Francois.
> 
> 
>> 
>> 
>> --
>> Markus Lanthaler
>> @markuslanthaler
>> 
>

Received on Thursday, 24 May 2012 20:19:58 UTC