HTML editor's meting

02 Aug 2016

See also: IRC log


LJWatson, Steve



longdesc http://github.com/w3c/html/issues/507



<arronei> Travis and I are connecting now


Travis: Ade says he isn't a big fan of this integration
... it's important that we articulate what has been added to the spec
... making it clear what we have done
... and then he has a list of thigns
... 1) removing the attribute from the deprecated list

Steve: was in the obsolete section of the spec

Travis: (very reasonnable)
... we added examples: unnecessary in Ade opiion but editorial only
... added a reference, unnecessary but ok

<chaals> exactly what we added…

Travis: added some technical details, which he is not certain on that one
... but we didn't import the technical content which is already in the longdesc spec
... [that was Ade feedback]
... if we can outline these 4 points, he could clear the air
... ie clearing the misunderstanding

Leonie: the issue is still open but the conversation disgressed
... chaals believes it should be closed
... others would lime to keep it open
... Ade list is an accurate summary?

Chaals: not sure about the examples. attribute list yes. IDL attribute in.

Leonie: we have this concept of applicable specs
... an extension is not HTML5 conforming, ie HTML5 conformance doesn't depend on extensions

<chaals> [+1 to remove the IDL piece]

plh: we should remove the IDL definition from HTML 5.1. it's already defined in the longdesc spec
... unless we remove it, it will destroy our extension story
... and it duplicates the definition, which isn't a good thing

Leonie: what about the examples?

Steve: as long as the examples are non-normatives, we should be ok
... the attribute shoudln't be listed in the attribute list
... since it's defined through extensions

Leonie: so leave examples if those are non-normative but remove the attribute from the list
... ok?

Travis: I think it's appropriate to keep the examples in
... so we remove longdesc from the tables but keep it in examples since we want to show end to end scenario perhaps

Leonie: are we ok with that?

Chaals: so, assuming that we collect a set of specs that impact HTML, that seems ok to me
... if we have 14 specs, which some who have 0 meaning outside of html, it should be clear that it's applicable to HTML

Leonie: if we have such list, that would be useful indeed

[discussion on where to put the "registry" of html elements and attributes]

plh: how about a group note and link it from the spec?

Summary: remove IDL attribute, and from the attribute/index list, propose to the Group to publish a Note and link the note from the spec. keep the examples.

[no pushback]

Modularisation plan

Leonie: Mozilla wasn't/isn't happy with the lack of modularization
... we need to start looking seriously for 5.2
... and present a plan to the WG at TPAC
... Travis mega patch is probably related to this?

Travis: I suggested that one way of modularizing the spec is to extract the pieces related to scripting (setting up the env, execution, ...)
... pulling those into one document
... it's an incremental benefit, but the desire to continue to do easy ports of WHATWG contributions in this area
... a lot of really good changes are happening from Domenic
... and we'd like to take a lot of that content and adapt it
... a success here would be to modularize and make it easier to integrate
... ideally, we wouldn't have the duality here, but modularization seems against their principle
... an other case: web workers and web messaging
... they felt by the way side that bringing back and update

Chaals: +1 to Travis. splitting different from WHATWG will make hard for people working across the 2
... make it harder for folks to keep it in sync

Travis: Ade is working on a plan to align with WHATWG. could lead to a closer relations or could get the situation worst
... I'm interested in finding a way to do the modularization with the WHATWG as well
... perhaps we could start with html parsing
... we could make a module for that
... and it's straightforward

<chaals> [+1 for splitting out parsing]

Travis: like the DOM parsing spec

Leonie: we already spinned out aria btw

Travis: we should think about modularization the parts of html that are least likely to change
... would be easier to update in the future
... html parser has bug but not really an active part of the spec

Leonie: can you provide a list of stable areas of the spec?

Travis: sure

<scribe> ACTION: Travis to provide a list of stable parts of the spec as potential candidates for modules [recorded in http://www.w3.org/2016/08/02-html-editors-minutes.html#action01]

<LJWatson> `zakim, take up next item

5.1 implementation report

[Thanks to Chaals for producing an implementation report]

Chaals: "scooped us from WHATWG" isn't a useful comment
... so document your changes
... one thing which is not implemented
...: dir is only in Firefox
... and one thing that is only in Blink
... current proposal is to go to PR with warning for those
... so that we can start focusing on html 5.2
... it's possible that the implementation report is shaky

<Travis> [+1 to Chaals' plan to push forward with current implementation report]

<chaals> PLH: dir is implemented in Firefox, but isn't an HTML feature

<chaals> … for promises rejection handlers, blink has partial implementation, and that's it.

<chaals> … looking at that and probablySupportsContext(), compard to re-doing the CR process in toto to clean that up looks like a lot of work.

chaals: we forgot some bits related to multiple for input/range
... there is a PR to remove the rest

Leonie: so, we're good for a CfC?

Chaals: editors should add a note

Steve: I'll do it
... I'll do a PR for longdesc as well

WHATWG triage

Leonie: can we meet on August 16?

Steve: I'm away

Chaals: difficult for me

Leonie: August 9?

Steve: I'll be in a plane

[dates are flying]

August 30

Chaals: I can do 25 as well

August 25, time to be defined


[plh raised the MSE issue 550]

<LJWatson> Zakim: rrsagent, make minutes

Summary of Action Items

[NEW] ACTION: Travis to provide a list of stable parts of the spec as potential candidates for modules [recorded in http://www.w3.org/2016/08/02-html-editors-minutes.html#action01]

Summary of Resolutions

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.144 (CVS log)
$Date: 2016/08/02 21:08:18 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.144  of Date: 2015/11/17 08:39:34  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/without/with/
Succeeded: s/15/16/
Succeeded: s/ance/ane/
Succeeded: s/24 and//
No ScribeNick specified.  Guessing ScribeNick: plh
Inferring Scribes: plh
Present: LJWatson Steve

WARNING: Fewer than 3 people found for Present list!

Regrets: Alex

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Got date from IRC log name: 02 Aug 2016
Guessing minutes URL: http://www.w3.org/2016/08/02-html-editors-minutes.html
People with action items: travis

[End of scribe.perl diagnostic output]