W3C

– DRAFT –
(MEETING TITLE)

13 August 2021

Attendees

Present
Bo, comandeer, GameMaker, johanneswilm, snianu, Travis, whsieh
Regrets
-
Chair
-
Scribe
Travis

Meeting minutes

<whsieh> is the link https://meet.google.com/wdp-xkma-taf?

no...

https://meet.google.com/pdx-dnmm-cen

Anyone want to scribe today?

<johanneswilm> Travis introduces himself. He is to replace Grisha for the time being.

agenda

Any other agenda items from folks here in IRC?

<johanneswilm> Bo: will get a PR ready for what part sof Input Events level 2 have to be marked optional

<johanneswilm> Wenson: we are on board with svg as safe clipboard mimetype

<johanneswilm> ... but we might want some sanitization that we can discuss here

<johanneswilm> Bo: virtual keyboard API - we have worked on it a while. have internal and external users. Is the group willing to move it to be a WD?

Note: not hearing opposition here on the call.

Action: BoCupp to send mail asking the same.

a?

https://github.com/w3c/editing/labels/Agenda%2B

Will be looking over these from top to bottom.

First issue: https://github.com/w3c/editing/issues/327

Anyone have a preferred day/time to meet at [virtual] TPAC?

Dates are 25 - 29 Oct.

Suggest that we keep same relative time and day?

Any other day we might also like to meet (for schedule flexibility)?

johanneswilm: think asking for 2-3 hours might be good?

https://github.com/w3c/editing/pull/265

johanneswilm: first we need an update from Bo/Grisha...
… they worked on a very long PR. Can we get this merged?
… there was a question in the issue asking about moving into the HTML spec.
… then Doemenic had a discussion about using the terminology of execCommand.
… I personally don't think we need to do it.
… if HTML thinks it's central and should use that naming scheme..
… would be good to have common understanding of execCommand.

<johanneswilm> Bo: it stalled because noone is going to jump on delivering this

BoCupp: to complete the PR... I think it stalled because may not be likely that anyone is going to be working on this?

<johanneswilm> it became deprioritized

<johanneswilm> I would rather wok on those things we want to ship

<johanneswilm> like virtualkeyboard api

<johanneswilm> wenson: we will nor prioritize

johanneswilm: this started as a request on the JS side... editors unhappy showing actions that won't be taken.
… whenever someone wants to know the current state, folks have to read through the spec + PR. Can we merge as an intermediate step.
… would it be easier to merge and then still not worry about putting further time into it.

I think that's a sensible request.

johanneswilm: there may be a few issues...

BoCupp: looks like there's a reduction in scope... dom token list was dropped.

johanneswilm: it was asked to include, but not done. But.. not a problem. We can file an issue to do it.

Action: johanneswilm to file issues that remain from this PR, and then merge the commit as is.

https://github.com/w3c/editing/pull/260

johanneswilm: question to whsieh? CAn we move undo manager api?

whsieh: Will talk to tess.

is it in scope of our charter?

johanneswilm: I think so, but whereever it ends up, we probably shouldn't link to a person account.

wondering if webkit/explainers would work?

BoCupp: is undo manager a concept you'd like to invest to move forward?

whsieh: momentum has stopped, but we still see value in it.
… good use cases :) I'd say "far on the horizon"

johanneswilm: there are buttons for undo/redo, and authors want to disable them. authors are trying to figure out how to integrate their own undo/redo.
… so how do we do this? Add events? Have an Undo Manager? What's the larger direction?

whsieh: I don't think having undo events is mutually exclusive to undo manager.

Action: whsieh to move undomanager to webkit/explainers, then update the issue so we can get a better link.

https://github.com/w3c/editing/pull/244

johanneswilm: people are still interested in fixing/changing execcommand.
… there's a new command "useCSS"
… implemented in Safari 10 years ago.
… not present in all the browsers.
… so technically shouldn't be deprecated.
… spec should say "we recommend "styleCSS" but "useCSS" shouldn't be marked as deprecated.

any objections to undeprecating?

<johanneswilm> bo: don't feel know enough to comment

BoCupp: don't know enough to comment--but seems reasonable if everyone has implemented it.

johanneswilm: there is a full-replacement for it "styleCSS".

not hearing objections to this undeprecation

Action: Ask @saschanaz to files bugs on implementations that may be firing deprecation warnings, and merge the PR.

https://github.com/w3c/input-events/labels/Agenda

https://github.com/w3c/input-events/pull/118

johanneswilm: We need to figure out if we can merge these specs.

johanneswilm: we could just auto-publish both and then think about merging later?

johanneswilm: we can defer this for another month.

https://github.com/w3c/input-events/issues/117

johanneswilm: We decided that we don't need a new input type, but we should agree on which type to use.
… then we got a question back: should there be a paste event?

Does anyone implement this with a paste event?

johanneswilm: new spec, implementers may do this if we spec it.

whsieh: similar to emojis.
… we treat it like a paste, and our implementation does fire 'paste'
… it's for compatibility.

johanneswilm: any argument against having a paste event?

BoCupp: is there a spec change here?
… where would this be documented?
… or is this just advice from the WG on how to maximize compatibility?

johanneswilm: we could be a non-normative note on how to handle paste.
… in the note we could explain to fire it... you'd write the event order.
… e.g., when does input event fire in relation to the paste event?

The firing of 'paste' is written somewhere, right?

BoCupp: I think it would be a good addition.
… also fine with a note.

Action: include a note in the spec (non-normative) about firing paste event as well in this scenario.

https://github.com/w3c/input-events/issues/115

johanneswilm: let's just remove the agenda label.

BoCupp: anupam is working on pickling API.
… questions on security aspects, sanitization.
… our plan is to make updates to clipboard-api spec (before shipping anything)
… we're working on it.
… :)

johanneswilm: can we update the editor (it's grisha)

BoCupp: who are we asking? I nominate Anupam!

Action: we'll take the nomination for Anupam back and try to make the switch.

whsieh: was a blog post recently about adding image/SVG+XML... to the web-safe list.
… one of the challenges is dealing with sanitization.
… SVG may use XML parser (vs. HTML parser)
… sanitization may involve more than adding this to the list of types.
… if we're not careful could lead to XSS..?

johanneswilm: does the sanitization need to be sanitized?

<johanneswilm> * standardized

whsieh: in blog post (link) there was a note about not wanting to standardize the sanitization algo.

<whsieh> link: https://docs.google.com/document/d/1Rx7gi01avpRRNYKSpp3U4WQdjery0H0IkX2XxDtfZ8I/edit

BoCupp: what is the outcome of "adding to web-safe list" does it join a list of things UAs must support?
… we had feedback from Moz in anotehr issue. Putting a mime-type on a list isn't sufficient...
… need to know the format when "html is on the clipboard"
… there are some interop issues.
… Moz feedback was to add that level of detail (to get interop)
… there are variations by platform, yet we'd like a platform neutral.
… I'm very supportive of providing the details.
… a little concerned about adding another one with no good docs.

whsieh: good poitn.

BoCupp: proposal: maybe have whsieh become a co-editor of clipboard-api to help iron out some of the details?
… (also helps improve clipboard-api spec).

RRSAgent: please make the minutes

Summary of action items

  1. BoCupp to send mail asking the same.
  2. johanneswilm to file issues that remain from this PR, and then merge the commit as is.
  3. whsieh to move undomanager to webkit/explainers, then update the issue so we can get a better link.
  4. Ask @saschanaz to files bugs on implementations that may be firing deprecation warnings, and merge the PR.
  5. include a note in the spec (non-normative) about firing paste event as well in this scenario.
  6. we'll take the nomination for Anupam back and try to make the switch.
Minutes manually created (not a transcript), formatted by scribe.perl version 136 (Thu May 27 13:50:24 2021 UTC).

Diagnostics

No scribenick or scribe found. Guessed: Travis

Maybe present: BoCupp, Note, RRSAgent