Ge0rGLet's see if we have anybody else, just to help the chair
ZashHere, but so tired
Ge0rG1) Roll Call
susmit88has left
Ge0rGLooks like there is no dwd...
Ge0rGdaniel?
jonas’here
jonas’thank you, Ge0rG
Ge0rGhi jonas’, looks like we have a quorum, if Zash is more than half awake at least.
jonas’let’s assume that he is
jonas’2) Agenda Bashing
jonas’any modifications?
Ge0rGlgtm
Zashsame
jonas’3) Editor’s update
jonas’congrats MattJ for getting 0.7.0 of XEP-0313 through the pipes
jonas’4) Items for Voting
jonas’4a) Revisit PR#963: XEP-0178: Clarify SASL-EXTERNAL specification when s2s
auth fails
URL: https://github.com/xsf/xeps/pull/963
Abstract: A while back it was discussed that XEP-0178 (SASL-EXTERNAL) for s2s
was kinda misleading - it says that server should close connection if
authentication fails but it seems that "everyone" (at least Prosody[0] and
ejabberd) actually fallbacks to dialback in that case.✎
jonas’4a) Revisit PR#963: XEP-0178: Clarify SASL-EXTERNAL specification when s2s auth fails
URL: https://github.com/xsf/xeps/pull/963
Abstract: A while back it was discussed that XEP-0178 (SASL-EXTERNAL) for s2s was kinda misleading - it says that server should close connection if authentication fails but it seems that "everyone" (at least Prosody[0] and ejabberd) actually fallbacks to dialback in that case. ✏
jonas’so there was some list discussion on this and we should cast a vote
jonas’I’m on-list though
dwdhas joined
Ge0rGon-list is a sane default.
Ge0rGhi dwd
dwdSorry, server issues. :-(
jonas’hi dwd
jonas’I’d be curious about your takeaway from the list discussion on PR#963 (see above)
dwdStill +1 for (4a).
dwdThe attack was predicated on a compromised key.
jonas’in which case SASL EXTERNAL itself would already be an issue, right?✎
jonas’in which case SASL EXTERNAL based on TLS PKIitself would already be an issue, right? ✏
jonas’in which case SASL EXTERNAL based on TLS PKI itself would already be an issue, right? ✏
dwdYes.
jonas’I see
jonas’Zash, vote?
jonas’dwd, thanks
ZashLet's say +1
jonas’alright
jonas’5) Pending Votes
jonas’we have a bunch of pending votes on various things
Ge0rGYes, I'd like to cast some
jonas’Ge0rG, go ahead!
Ge0rG-1 to https://github.com/xsf/xeps/pull/972
Ge0rGthat was easy.
Ge0rGoh, right. The rationale: would change normative language of a Final XEP without a damn good reason
Ge0rGre #971 I'm not even sure how it is a clarification, as it's actually *removing* text from the XEP
Ge0rG+1 to https://github.com/xsf/xeps/pull/969
Ge0rGWas there any list discussion of #971 yet? Everybody is on-list on it
jonas’no
paulhas joined
jonas’I considered it editorial actually
jonas’but I decided to look closely again after last week’s discussion, didn’t have the time tho
Ge0rGroger.
jonas’I think the PR removes noise, which makes things easier to read
Ge0rGI think I'll on-list it as well, with a gut feeling of +0
dwdI have lots of pending votes - sorry, work has been spiking again. I'll set aside some time now to go through them.
jonas’dwd, thanks
jonas’sorry for the text screenshot, but here’s a rendered word-diff:
jonas’looking closely, I noticed that removes scarcely-defined ("MAY") behaviour for type="cancel"
jonas’other than that, I think it makes the text easier to read
jonas’since the duplication of MAY and OPTIONAL is folded into a single OPTIONAL
jonas’dwd, now as in "right now"?
dwdIt looks - well, it looks confusing anyway, but it looks as if 971 restricts the OPTIONAL to only submit/result, whereas prior to that type is always a MAY, but particularly OPTIONAL for submit.
dwdjonas’, Not in the meeting. I'll go through afterward on list.
jonas’dwd, thanks!
jonas’OPTIONAL == MAY
jonas’ftr
jonas’so the only change is for type="cancel", which was in the previous text covered by "fields provided in the context of other forms types MAY possess a 'type' attribute as well.", which is now gone
jonas’how about I make the editors add the "cancel" to the enumeration in the last sentence?
jonas’(or flow)
ZashHmmm
dwdYeah, so OPTIONAL == MAY (and maybe === depending on your language of choice), but it does read as though submit is more optional, for types, than others.
jonas’which is where I think the clarification actually comes in handy
jonas’either way, I’m under a bit of time pressure, so let’s take this to the list or to the next meeting
dwdThe problem is, sort of not really. The text says "type" is optional, and if it's not present it means "text-single". But you might not get it on submit because you know the type is.
