Roadmap for draft v5?

[erosb]

Hello,

as an implementor of the json schema spec I would like to know when will the final version of draft v5 be available and where will it be published. What is the best source of information to keep an eye on this?

Thanks.

[danielpeintner]

Hi,

I do have a similar question.

It seems that there is now interest in using JSON Schema.

• The EXI working group looks at JSON schema to improve the efficient encoding of JSON (see https://www.w3.org/TR/exi-for-json/)
• the W3C WoT interest group considers to use JSON Schema to define the types used in thing descriptions for the Web of Things (see http://w3c.github.io/wot/current-practices/wot-practices.html#type-system).

These specifications require other specifications to refer to. JSON schema is not yet an RFC or a specification.

May I ask what prevents the group doing so? Is it the actual work that needs to be done or is there any other blocker.

Thanks!

[handrews]

See also:
#16 Wiki is not migrated
#27 What does it take to get the JSON-Schema specification adopted by an IETF working group?
#33 Update json-schema/json-schema to redirect people here!

Note that as shown in comments to #33, there is a lack of will to get the old JSON Schema repository on GitHub closed down as much as possible, and a disagreement about whether or not that benefits the project. While we cannot close the repo (the owner is AWOL and unresponsive), I believe that closing all of the issues (either migrating or resolving), dismantling the wiki (except maybe pointers to new homes), and tagging the repo (to preserver current state) followed by deleting everything but the README on master (to make it very clear that the repo is dead) would dramatically decrease the confusion surrounding the project. I am trying to build support for this position, in case that's not obvious :-)

[brutzman]

Sounds like a plan! Once there is a draft JSON schema v5, the Extensible 3D (X3D) Graphics working group is interested in testing it - pretty thoroughly - as part of validation for the X3D JSON Encoding. Current work includes evolving from a manually written schema to an autogenerated X3D JSON schema.
http://www.web3d.org/wiki/index.php/X3D_JSON_Encoding

[erosb]

Who can make these steps happen?

2016-10-07 21:24 GMT+02:00 brutzman notifications@github.com:

Sounds like a plan! Once there is a draft JSON schema v5, the Extensible
3D (X3D) Graphics working group is interested in testing it - pretty
thoroughly - as part of validation for the X3D JSON Encoding. Current work
includes evolving from a manually written schema to an autogenerated X3D
JSON schema.
http://www.web3d.org/wiki/index.php/X3D_JSON_Encoding

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
#71 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/AAQrBoyjDNHFMqHABsyFPXfCzGlm6HNjks5qxpxtgaJpZM4KOrVS
.

Bence Erősme@github https://github.com/erosb

[brutzman]

When there is a stable json-schema v5 that the community expects will be proceeding towards IETF, the X3D working group will add such testing to our ongoing efforts.

The heavy lifting for json-schema itself is up to this community.

[handrews]

@awwright is doing a lot of work right now to make a stable v5, and has produced a much improved web site to host the new work. Then we'll look at the v6 proposals. Based on the proposals so far, here's what I see as likely v5 to v6 changes (this is purely my opinion/speculation):

core: There's some argument over id, including whether it should still exist at all, whether it should be $id instead, or if some other change should be made. As it is, I believe @awwright has significantly improved the language for v5 without otherwise changing the keyword, but it might get bigger changes in v6.

validation: It seems likely that validation will remain backwards compatible, or very close to it. Additional features are likely, although how many of the proposals will be accepted is not yet clear.

hyper-schema: I expect/hope to see something between major and drastic changes here. There is a bug open asking for any example uses in the wild, and it's been notably quiet. I have a slew of proposals to file around v6 hyper-schema based on a prior project where we felt compelled to develop an alternative due to problems involved in using hyper-schema (I do not plan to propose the whole alternative, but rather pick elements and see how they can best fit into the project here).

UI: Not currently part of JSON Schema, there is a bug proposing an extension project. Some proposals around validation keywords may belong here instead.

[awwright]

@erosb I don't anticipate any major changes to functionality, just clarifying bits and pieces here and there. @handrews described most of the work.

I plan on having an I-D submitted within a week or so.

If you wanna take a look right now, check out the 'master' branch, review the current documents, and let me know if it all makes sense!

[Relequestual]

I'm excited that there is traction!! Super great to see a number of people wanting to make stuff happen. A few new issues to review! Will be glad to do what I can, which may be limited, true.

[Relequestual]

It's worth noting, probably in another issue I expect, that some of the people who worked on draft 4 (I forget who), felt like splitting core and validation was a mistake. Re-converging the two is something the community should consider.

[Relequestual]

@erosb Do you feel this issue can be closed?
I feel that issues which are included in the draft-5 milestone are sufficient.

[erosb]

yes, thanks!