Final community group report

[rubensworks]

The RDF/JS specification may serve as starting points for proper W3C recommendations. This will require the setup of a new W3C working group, which will require the specs of our group here to be finalized and published as final reports.

During a CG call on April 13, we discussed the steps that are needed for publishing the final reports:

• Create final report for data-model, stream, and dataset (non-experimental part) specs (other specs are not stable enough)
• Data-model spec
  • WebIDL errors
• Streams spec
  • WebIDL errors
• Dataset spec
  • WebIDL errors
  • Remove interface marked experimental

Please vote on this plan for publishing final reports before May 3rd with 👍 (you agree) or 👎 (you don't agree). If you don't agree, please motivate your vote below.

If the majority agrees, we can start working on the final reports, after which we can start chartering the working group.

[awwright]

I'm indifferent; it seems like something needs to be published, and what's currently published doesn't have serious internal problems; but neither do they have the features that meet my needs.

As the biggest example, the specs don't have any way to represent triples, you can represent a quad with a "graph" component of "default", but these are not semantically the same. A Quad is making a statement that a given Triple exists in a given Graph, and right now we have no way to talk about a Triple without making this assertion.

[rubensworks]

It looks like the majority agrees with the proposed steps, so we can move forward with this.

Are there any WebIDL experts here that are willing to help out with fixing the WebIDL errors in the different specs?

[awwright]

@rubensworks I object. I thought we move things by consensus. A process where up to 50% of the members could be unhappy seems like a really bad way to decide on adopting a specification. (To be more specific, legislatures/Robert's Rules uses majority voting because any higher a threshold and it couldn't be later repealed; which is not something that applies here.)

[rubensworks]

I object

@awwright That's not what you said before. You said you were indifferent, and you did not cast a vote (👍 or 👎). Based on the votes, it seems to me like a consensus has been reached on this issue. I'll leave it up to one of the chairs (@bergos @RubenVerborgh) to decide whether or not we have indeed reached consensus on this issue.

In any case, I think once we move towards a proper WG, we can discuss https://github.com/rdfjs/data-model-spec/issues/144 in detail, because I don't expect much movement anymore on that issue within this CG. But before we can do that, we need to publish final reports.

[awwright]

@rubensworks Sorry, it's not so much the voting per se, I went too far on a tangent there. Rather I was hoping to get some sort of response to my earlier post before we announce any outcome. Since I'm identifying a significant issue, it seems to me the further the process goes, the less capability we will have to make changes.

But I don't know if this is a correct assessment or not. "Indifferent" may have been the wrong word; it's not that I'm abstaining. It's that the proposal is terse and depends on knowing W3C process; and I don't have the experience to evaluate the claim that publishing a final report is the best way to fix the outstanding issues.

[rubensworks]

Since I'm identifying a significant issue, it seems to me the further the process goes, the less capability we will have to make changes.

No need to worry about that. There are other significant issues that needs further discussion as well (such as https://github.com/rdfjs/stream-spec/issues/12). The goal of finalizing the specs here is just so that we have a sound starting ground for the WG.

Once the WG starts, it is not obligated to reuse the final reports from the CG as-is, they just act as input. This means that the WG can decide to completely change how the RDF/JS specs are defined (but I hope it doesn't 🙂). If you decide to become part of the WG, and you raise your issue(s) there, it will certainly be discussed, and the group will try to reach a consensus.

[awwright]

Alright that reassures me a little bit, thanks @rubensworks

[bergos]

Thanks for the votes and feedback. The next step would be fixing the listed issues. I assigned myself the WebIDL error issue in the data-model-spec and will try to work on it next week. Does anyone want to work on one of the other issues? The WebIDL error messages in the developer console look quite useful. I guess you don't need to be an expert in that field.

[rubensworks]

I can have a look at https://github.com/rdfjs/stream-spec/issues/20.

[rubensworks]

Anyone interesting in looking at the dataset spec issues? Perhaps @vhf or @blake-regalia?