- From: Ivan Herman <ivan@w3.org>
- Date: Mon, 30 Sep 2019 17:52:01 +0200
- To: Gregg Kellogg <gregg@greggkellogg.net>
- Cc: Benjamin Young <byoung@bigbluehat.com>, W3C JSON-LD Working Group <public-json-ld-wg@w3.org>, Matt Garrish <matt.garrish@gmail.com>
- Message-Id: <8F851CCE-5346-4B9F-93EA-19D5ECCCCB20@w3.org>
> On 30 Sep 2019, at 17:48, Gregg Kellogg <gregg@greggkellogg.net> wrote:
>
>> On Sep 30, 2019, at 8:37 AM, Ivan Herman <ivan@w3.org <mailto:ivan@w3.org>> wrote:
>>
>> Yep. What this means is that if we use @direction in the Publication Manifest, then we do have to add a note that this is a 1.1 feature, and if an application uses a JSON-LD processor _and_ @direction is used, then it is necessary to use 1.1
>
> I believe the feeling of the group was that you should normatively reference JSON-LD 1.1, not 1.0. You may note the specific 1.1 features you’re limited to, but the future is longer than the past, and I would expect that the distinction would fade pretty rapidly.
Yes, we would definitely refer to 1.1. The only thing I think should be done is to add a note in the document warning that @direction is a genuinly 1.1 feature.
That being said… on your advise the publication manifest's context does contain a bunch of @language:null statements (e.g., when the expected string value is a URL). I wonder whether it would be necessary to add a @direction:null to those, in case the author adds a global @direction. Probably yes, but that would create problems with 1.0 processors, too
Ivan
>
> Gregg
>
>> We can mitigate the effect if we do _not_ add a "direction":"@direction" alias into the publication manifest context.
>>
>> I realize the Publication manifest is not the JSON-LD WG's problem, but it is an obvious use case for the transition…
>>
>> Thx
>>
>> Ivan
>>
>>> On 30 Sep 2019, at 16:56, Gregg Kellogg <gregg@greggkellogg.net <mailto:gregg@greggkellogg.net>> wrote:
>>>
>>> We discussed this quite a bit at the F2F. IIRC, the feeling was that if schema.org <http://schema.org/> decide to use this, the impact would be limited, but would require the use of a 1.1 processor. And, not properly ignoring non-keywords that look like keywords could be considered a 1.0 bug, as the syntax spec says they should be ignored.
>>>
>>> Gregg Kellogg
>>>
>>> Sent from my iPad
>>>
>>> On Sep 30, 2019, at 6:59 AM, Benjamin Young <byoung@bigbluehat.com <mailto:byoung@bigbluehat.com>> wrote:
>>>
>>>> That's because of the schema.org <http://schema.org/> context using `@vocab`. If they didn't, then a JSON-LD 1.0 processor wouldn't choke. For example: https://tinyurl.com/y3vzjegw <https://tinyurl.com/y3vzjegw> (maps `name` directly in `@context`).
>>>>
>>>> Sadly, the most widely used context files (presumably Schema.org <http://schema.org/> and ActivityStreams) use `@vocab`.
>>>>
>>>> We did discuss this for a long while at TPAC...but I don't think we ever found a solution for old processors + `@vocab` containing context files...
>>>>
>>>> It's best expressed in this issue:
>>>> https://github.com/w3c/json-ld-syntax/issues/259 <https://github.com/w3c/json-ld-syntax/issues/259>
>>>>
>>>> Which resulted in reopening this one:
>>>> https://github.com/w3c/json-ld-syntax/issues/16 <https://github.com/w3c/json-ld-syntax/issues/16>
>>>>
>>>> We can make sure getting us all in sync around this status is on the agenda for this Friday.
>>>>
>>>> Cheers,
>>>> Benjamin
>>>> --
>>>> http://bigbluehat.com/ <http://bigbluehat.com/>
>>>> http://linkedin.com/in/benjaminyoung <http://linkedin.com/in/benjaminyoung>
>>>> From: Ivan Herman <ivan@w3.org <mailto:ivan@w3.org>>
>>>> Sent: Monday, September 30, 2019 4:29 AM
>>>> To: W3C JSON-LD Working Group <public-json-ld-wg@w3.org <mailto:public-json-ld-wg@w3.org>>
>>>> Subject: @direction in json ld 1.0
>>>>
>>>> Maybe I did not remember right, but we thought that "@direction" should be absorbed by a JSON-LD 1.0 processor, ie, we can deploy it safely. However, the construction:
>>>>
>>>>
>>>> {
>>>> "@context":"https://schema.org <https://schema.org/>",
>>>> "name" : {
>>>> "@value": "na mi van",
>>>> "@language": "hu",
>>>> "@direction": "ltr"
>>>> }
>>>> }
>>>>
>>>> will not work either; there will be an error message due to the presence of value and direction in one object (see [1]).
>>>>
>>>> I guess we cannot do anything about it, but it is a bit of a pain:-)
>>>>
>>>> [1] https://tinyurl.com/y6goh76p <https://tinyurl.com/y6goh76p>
>>>>
>>>>
>>>> ----
>>>> Ivan Herman, W3C
>>>> Home: http://www.w3.org/People/Ivan/ <http://www.w3.org/People/Ivan/>
>>>> mobile: +31-641044153
>>>> ORCID ID: https://orcid.org/0000-0003-0782-2704 <https://orcid.org/0000-0003-0782-2704>
>>>>
>>>>
>>
>>
>> ----
>> Ivan Herman, W3C
>> Home: http://www.w3.org/People/Ivan/ <http://www.w3.org/People/Ivan/>
>> mobile: +31-641044153
>> ORCID ID: https://orcid.org/0000-0003-0782-2704 <https://orcid.org/0000-0003-0782-2704>
>>
>
----
Ivan Herman, W3C
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
ORCID ID: https://orcid.org/0000-0003-0782-2704
Received on Monday, 30 September 2019 15:52:07 UTC