<chaals> scribe: Travis

plh: will need to get pre-implementation info prior to May 21 (so that we can mark at risk)

bikeshed

arronei: Recent Bikeshed update now has an "index out of range" in a biblio file.
... only seems to happen in Travis.ci
... ??
... reached out to Tab. Still don't know what the problem is.
... (problem with Travis.ci?)
... Have changed to use live Bikeshed to workaround this issue.
... want to continue investigating how to use the cached copy.
... but want to fix the license issue and some links.

adrianba: I want to try to help fix this technical issue.
... we have seen that we can't rely on the live Bikeshed. It causes too many issues.

plh: we're only talking about the editor's draft, not the publishing process.

adrianba: We're protecting ourselves by using Travis.ci to ensure we build clean.
... so that folks' PRs can be checked and clean.
... when we started out using the live version it was chaos (Bikeshed was updated a lot and broke us.)
... we can do a temp workaround.
... add something to the repo readme to say that we're currently working on the live version; if [customer] has an issue, let us know.

arronei: hope to be able to fix in about a week..

Will be reviewing issue now.

WHATWG sync'up

<chaals> scribeNick: chaals

-> https://github.com/whatwg/html/commits/master?page=11 our starting point…

-> https://github.com/w3c/html/issues/364 Things we would like to take into 5.1

-> https://github.com/w3c/html/issues/368 Things we will look at after 5.1

-> https://github.com/w3c/html/issues/365 Things we don't expect to take into HTML

Normative references

PLH: CSS-backgrounds is OK

AE: We can reference CSS-color-3 not 4
... CSS generated content will not stabilise soon enough, ditto all the CSS references following that until CSS-style-attr.

PLH: CSS syntax should be OK.
... text is not stable, nor transitions. TL: Why isn't transitions done - it's old.

[but doesnt' work interoperably]

AE: UI3 is OK, 4 no. value is stable. Writing modes is stable but moves slowly.

PLH: Link to CSS 2.2 instead of 2.1?

Content is not stable, images is.

CSS-Ruby isn't stable.

<scribe> ACTION: PLH to publish a Working Draft of DOM serialisation. [recorded in http://www.w3.org/2016/05/11-html-editors-minutes.html#action01]

[editing] is unstable, encoding we should shift version, fetch is going to be interesting.]

PLH: WebAppSec convinced the director that they could point to fetch. We would need to examine where we reference fetch.
... An alternative is to say "we're just going to copy definitions into a spec appendix, so we don't have to worry about things shifting"

AD: There are only 4 places we refer to Fetch.

CMN: Can we just drop the dependencies

PLH: FileAPI was accepted for 5.0
... is fullscreen normative?

AE: I'll check to see if things are real normative references.
... think we use Geometry spec.

PLH: HR-Time 1 is good enough.

ARIA reference needs to be sorted, to point to the right one and make sure it advances in time.

scribe: why tdo we reference Japanese layout requirements.
... Why do we reference MathML 1.01?

Why is media-source normative?

- because it extends HTML.

scribe: MIME sniffing we need
... why do we depend normatively on OggSkeleton?

PLH: Selectors 4 - where is that?

AAE: We depend on that.

PLH: Service workers as a normative ref?

AE: We may not need to do that?

AD: Clean up SVG stuff
... no need for 1.2tiny or 1.0

PLH: Should we reference Pointer Events?
... URL is nasty so we should try to keep it as-is…

<scribe> ACTION: Chaals to figure out how we point to URL concepts. [recorded in http://www.w3.org/2016/05/11-html-editors-minutes.html#action02]

AD: We don't need TypedArray anywhere.

AE: WebIDL should point to /TR not haycam's repo. Maybe it's the reference text.
... webworkers and whatwgwiki aren't normative dependencies.
... no idea why we have worklets.
... XHR link is a bug.
... SO, Fetch, DOM, are the likely pain points.

Looking at diff

<plh> scribeNick: plh

Section 4.7.14.12.6 should be marked at risk

Section 4.7.14.12.9 should be marked at risk?

+ associated event handlers

datetime and datetime-local at risk?

inputmode?

Drop 4.10.19.3 Autofill

<scribe> ACTION: plh will look into section 4.11 and figure what we should drop or mark at risk and also for section 4.12 [recorded in http://www.w3.org/2016/05/11-html-editors-minutes.html#action03]

<chaals> Editing is a mess, and known to be a mess, but there isn't much of a solution available now :(