GuusI've read up on the request, which in itself made sense to me
ralphmI'm +1 on this request
GuusI was wondering if that omission was there by design though
ralphmWe can change the design
Guusor if it is simply something that was missed
Guuswell, if it was there by design, then it must've had some reasoning behind it
ralphmWell, fortunately the Approving Body makes the final decision on taking this new transition, so I think that's good enough.
ralphmGuus: well, we've changed the transitions before. It is ok, really.
GuusI'm not saying we can't change it 🙂
ralphmIf you want additional blessing, bring it up when we talk to stpeter.
GuusI'm just wondering why it's missing in the first place 🙂
ralphmProbably because it didn't come up.
ralphm(I don't recall it coming up in my 8 terms)
ralphmnyco?
Guusok, I'll bring it up in the talk with Peter, for my ease of mind 🙂
ralphmGuus: are you voting anyway?
nycofollowing discussion, not disagreeing 😉
Guusno, I'll wait for Peter's feedback.
Guuswhich makes this a matter for next board... I guess?
ralphmOk. Let the records note that we carry this over to (a) next meeting of the Board.
waqashas joined
ralphm5. AOB?
nyconope
Guusyour quick reaction to my idea of launching some kind of IOS contest, perhaps?
Guusdid you guys catch that?
nycowhat?
ralphm?
Guusbasically: I'm toying with an idea to have the XSF organize some kind of XPrize-style prize for development of something that we're sorely missing today, an ios-client being an obvious, but perhaps not only, choice.
Guusthis was discussed in this MUC a couple of days ago, wondered if you guys caught it.
nycoI missed it
nycomakes sense
Guusthe idea is to stimulate development, to make the XMPP ecosystem more complete.
nycoI fear difficulty may appear on the money front, but to be proven
Guusat this stage, I primarily would like to know if the idea is worth exploring, according to you
nycoI strongly support an attempt
ralphmNot against the idea, but needs more detail.
Guusk, tx
ralphm6. Date of Next
nyconope
nyco😉
nycowe let "them" decide
ralphmAs a handover, I suggest holding a meeting a week from now again.
nycoah, ok, then I'm in, +1W
ralphmEven if I'd not be elected, I'd still be chair, so that's ok for me to schedule :-D
Guusone of the board candidates is from the US, which might not be up by then. 🙂
ralphmSo penciling it in.
ralphm7.
ralphmClose
nycobye, team 😉
ralphmI'd like to thank everyone on the 2018/2019 Board and Council for past year's effords. Curious about the election results.
ralphmGe0rG: It is highly unlikely that we will have Jabber™ hoodies made.
Guuscan we, license wise?
ralphmThere've also been requests last year to have other things than hoodies. Kev and dwd might have ideas.
ralphmGuus: yes
moparisthebestdid memberbot go down?
nycohas left
ralphmAlex ^
ralphm(it is up for me, I think)
Ge0rGmemberbot works for me as well
Guusmemberbot is up and responsive to me
Nekithas joined
jjrhhas left
moparisthebestyep working again, though I just got on gajim and it has an angry face and under construction emoji as a name has that always been there?
efrithas left
Yagizahas left
Guushaven't noticed, but that doesn't say much.
moparisthebestfor the second time it froze after I cast my first vote for board...
Ge0rGmaybe Gajim froze?
moparisthebestmaybe I'm breaking it somehow, first time I was using conversations, now gajim
Yagizahas joined
GuusI've had zero problems with Converse, and Conversations a few elections back.
moparisthebestneither have I
ralphmmoparisthebest: those are User Activity and User Mood icons.
moparisthebestdoes it normally have those? :)
ralphmWell, has had them for a long time
moparisthebesthas left
ralphmhas left
moparisthebest[10:15:55 AM] SleekXMPP got into trouble.
moparisthebestthere finally got an error to show up in gajim
moparisthebestthat's after I cast my first vote for board
ralphmmoparisthebest: the User Mood is 'serious' and the Activity is 'working'
moparisthebestif it resets and responds to me again I can try to capture some XML, but uh, who can check up on it? :/
nycohas left
jjrhhas left
ralphmAlex
Nekithas left
Nekithas joined
jjrhhas left
j.rhas joined
lhas left
lhas left
lnjhas joined
lovetoxhas left
ThibGhas joined
blablahas joined
Guushas left
efrithas joined
lhas left
guusdkhas left
guusdkhas joined
waqashas left
waqashas joined
guusdkhas left
nycohas left
Holgerhas left
efrithas left
vanitasvitaehas left
guusdkhas left
guusdkhas joined
Marandahas joined
jjrhhas left
genofirehas left
!xsf_martinhas joined
jjrhhas left
moparisthebesthas joined
!xsf_martinhas left
labdsfhas left
!xsf_martinhas joined
jonas’moparisthebest, you can also vote in the meeting later at 19:00Z
jonas’so if memberbot hates you for some reason...
Ge0rGI always missed the live meetings.
blablahas joined
vanitasvitaehas left
j.rhas joined
Steve Killehas left
Steve Killehas left
jonas’anyone from jabber.de who sent me a subscritpion request?
!xsf_martinhas left
!xsf_martinhas joined
genofirehas left
Alexhas left
jjrhhas left
blablahas joined
moparisthebestjonas’: Thanksgiving in US today so will be eating then :)
jjrhhas left
jonas’moparisthebest, hah, right
jonas’moparisthebest, I pinged him out-of-band, maybe that helps.
moparisthebestI emailed my votes to him and he responded so I'm fine
jonas’ah ok
moparisthebestMemberbot crashing is another story :)
jonas’woah
Alexhas left
jonas’git merge silently puts a revision block added in one branch of the xeps repo below revision blocks which were added later on
jonas’that’s... something to be aware of
APachhas left
vanitasvitaehas joined
labdsfhas joined
lskdjfhas left
Steve Killehas joined
guusdkhas left
guusdkhas joined
matlaghas left
Zashhas left
matlaghas left
jjrhhas left
marchas joined
!xsf_martinhas left
tuxhas left
lskdjfhas left
j.rhas left
Steve Killehas left
lskdjfhas joined
tahas left
tahas joined
j.rhas joined
lskdjfhas joined
lskdjfhas left
lskdjfhas joined
ThibGhas joined
ralphmWhat does that mean?
jonas’practically, it means that editors must be extremely cautious when merging multiple branches affecting the same XEP
jonas’this should’ve been a conflict, because the original commit (which was being merged here, see <https://github.com/xsf/xeps/commit/cd8d05a6605ed742419a036c185ada9d174ba5fe>) added the 1.15.3 block to the top of the list (where other blocks have been added in the meantime)
jonas’I don’t see how this is *not* a conflict
jonas’but instead, git silently merged that by ordering the "new" revision block below the existing ones
ralphmWell, fuzziness is always tricky. In this case it just looked at the bottom of the patch chunk
ralphmFound a match
jonas’I don’t think that’s a valid thing to do in the general case
ralphmAnd cleanly applied
jonas’ignore the top part entirely
labdsfhas left
labdsfhas joined
lskdjfhas joined
jonas’it’s clear what techincally happened, but I don’t think this is correct behaviour
ralphmSo I guess there could be a script to test for such inconsistencies
jonas’yes, I think we might need that
ralphmWhich would let you know before the merge
labdsfhas left
labdsfhas joined
lovetoxhas joined
lskdjfhas joined
Alexhas joined
404.cityhas left
404.cityhas joined
marchas left
lumihas joined
moparisthebesthas left
moparisthebesthas joined
marchas joined
lskdjfhas joined
vanitasvitaehas left
vanitasvitaehas left
Nekithas left
Nekithas joined
guusdkhas left
Alexhe guys, I think its member meeting time ;-)
jonas’is here
guusdkhas joined
ralphmwaves
Zash\o/
Alexokay, lets kick it off
ralphmbangs the gavel for Alex
Alex(Y)
Alexhere is our Agenda for today:
https://wiki.xmpp.org/web/Meeting-Minutes-2018-11-22
Alex1) Call for Quorum
jonas’wait, top four candidates for board?
jonas’why not five?
Alexas you can see 35 members vote via proxy, memberbot and email
jonas’> The top 4 candidates will comprise the Board for 2017-2018. There is a tie for position #5 which will be resolvde after the meeting using random selection as defined n RFC 3797.
(also the year number is off; sorry to interrupt you)
labdsfhas left
Alexsorry, copy and paste error ;-)
Alexits corrected now
Marandahas joined
Alexdon't know the results yet ;-)
Alex2) Items Subject to a Vote
!xsf_martinhas joined
Alexboard and council elections, you can see all the applicants here:
https://wiki.xmpp.org/web/Board_and_Council_Elections_2018
Alex3) Opportunity for XSF members to Vote in the Meeting
Alexanyone here who has not vited yet and wants to do so now?
Alexotherwise I shutdown memberbot
Alex3
Alex2
Alex1
Alexnobody
Alexshutting down the bot and start counting
ZashBegin drumroll!
ralphmDrrrrrrrrrrrrrrrrrrrrr
GuusRollllll
ralphmrrrrrrrrrrrrrrrrrrrrrrrrrrrr
ralphmrrrrrrrrrrrrrrrrrrrrrrrrrrrr
ralphmrrrrrrrrrrrrrrrrrrrrrrrrrrrr
waqasumm
GuusNow you're just purring...
ralphmIt's a roll people.
labdsfhas joined
ralphmrrrrrrrrrrrrrrrrrrrrrrrrrrrr
Ge0rGA people roll?
Ge0rGI'm so excited!
Alex4) Announcement of Voting Results
Alexwhen you reload the page you can see the results:
Ge0rGralphm: so you are making a *badumtss* on every bad joke?
ralphmYes
Ge0rGLike when somebody mentions my full name? 😢
Zashhas left
Yagizahas left
Yagizahas joined
guusdkhas left
guusdkhas joined
Alexhas left
Alexhas joined
Alexhas left
ralphmNah
rionhas joined
vanitasvitaehas left
vanitasvitaehas joined
thorstenhas joined
Neustradamushas left
Alexhas joined
waqashas left
guusdkhas left
guusdkhas joined
guusdkhas left
guusdkhas joined
pep.has left
!xsf_martinhas left
Alexhas left
Alexhas joined
goffihas joined
Alexhas left
lovetoxhas left
lumihas left
vanitasvitaehas left
vanitasvitaehas joined
guusdkhas left
Holgerhas left
ThibGhas left
404.cityhas left
404.cityhas joined
ThibGhas joined
goffihas joined
rionhas left
vinx55has joined
vanitasvitaehas left
Yagizahas left
vanitasvitaehas joined
404.cityhas left
lhas joined
lskdjfhas joined
!xsf_martinhas joined
marchas left
matlaghas left
goffihas joined
Nekithas joined
lnjhas left
vinx55has left
ralphmhas joined
genofirehas joined
lhas joined
lskdjfhas joined
rionhas joined
krauqhas left
alexdehas left
tuxhas left
danielI’m starting the thing that an xml notation for message styling / markup wouldn’t be the worst idea. but i don’t want the full feature set of xhtml-im. It can borrow html5 vocabulary but i want the feature set to be limited to something that message styling and message markup offer. so probably something like <strong>, <em>, <pre>
daniel*think
jonas’this is going to re-invent the XHTML-IM wheel.
danielmaybe. but if it is limited to 5 elements w/o any attributes people might not attempt to reuse browser renderers
jonas’people will always attempt stupid things
danielxhml is just soo complicated. i don’t think anyone but Link Mauve ever wrote a renderer for it
danieland that one probably isn’t even complete
jonas’how about re-defining the XHTML subset used by XHTML-IM?
ZashDidn't it have profiles already?
jonas’it used XHTML profiles I think
danieli mean it has freaking css in it
Link Mauvedaniel, uh, a lot of people have.
Link MauveThere are in all of the 2000s clients.
jonas’libpurple probably has one with an RCE even.
Link MauveAnd the CSS parts it has are very small, nothing like CSS level 2.
danielputting css in there breaks with the whole let's use xml idea
jonas’ah, yes
Link MauveIt’s also optional.
danielit doesn’t make sense to use css
Zashstrip the CSS then
jonas’that’s a good point
jonas’more like 90ies HTML with stuff like <font color="xyz"/>
ZashThe mod_xhtml I made for Prosody can do that
Zashjonas’: Like how SVG works now? :)
Link Mauve“23:02:20 jonas’> it used XHTML profiles I think”, no, it defines profiles of XHTML, which is very different.
jonas’Link Mauve, ah right
danielalso if it is called xhtml-*IM* why does it have shit like headline
jonas’Zash, pretty much
danieli understand the use for blogs and stuff
danielbut IM?
jonas’daniel, rich text bots
Zashdaniel: Throw it out!
danielwho’s going to put headlines in messages
Zashµblogging
jonas’daniel, one of the reasons the company I work for swtiched away from XMPP was that they wanted fancy alert messages
Zashdaniel: People write forum posts in messages sometimes
jonas’... but clients lacked proper support for that
jonas’or maybe I should’ve tried more
Link Mauvedaniel, again, nothing forces you to implement the entire list of profiles defined in XHTML-IM.
jonas’whatever
Link MauveNothing will break if you don’t render a <h1/> a bit bigger than normal text.
lnjhas left
ralphmI think the biggest concern with XHTML-IM is that people just throw it at their favorite HTML renderer, opening up all kinds of security issues. No matter what subset.
Zashjonas’: I'm not sure fancy notifications should require all of XHTML. Eg Slack doesn't do that .. does it?
ZashFancier than plain text messages, sure
Link Mauveralphm, people will do that, no matter what the input format.
Link MauveI’ve seen more RCE using User Nickname in the wild.
Link MauveSimply because “it’s a nickname, so plain text, what could possibly happen if I use innerHTML?”
jonas’Zash, I’ve seen headings and other markdown-like formatting
jonas’ralphm, we had that discussion for a month or so a year back
ralphmAs for what Slack does, I think that's mostly XEP-393.
ralphmI know this
ralphmI didn't restart it today
Zashjonas’: `<subject>` :)
jonas’Zash, subheadings
Zashjonas’: nyeh
Link MauveHeadings used to make more sense with the email-like UI of normal messages in the likes of Psi and Gajim.
danielwell i like 0393 as an input format and i think it is enough for a feature set
Link MauveWhere you may have wanted full formatting.
jonas’daniel, I agree with you for non-automated beings :)
jonas’daniel, I agree with you for non-automated beings in the IM use-case :)
Link Mauvedaniel, yup, no issue with that.
ralphmdaniel: indeed
danielbut I think it might make sense to transmit the feature set of 393 as xml
ralphmWhy?
Link Mauveralphm, interoperability with other input markups.
jonas’daniel, tbh, I’d prefer if we could re-vive XHTML-IM instead of inventing our own vocabulary.
danielralphm, by transfering the input format you can never really remove the formatting because the client doesn’t know if it is formatting
jonas’(and define a subset which is equivalent to 393)
Marandahas left
Alexhas joined
danielparsing the input format to xml and using that could solve that
jonas’also, this is what I told you folks a year ago already >.>
jonas’if we’re going to define our own vocabulary, people will still come up with unsafe hacks to put it into an XHTML renderer directly
danieljonas’, well the vocabulary i suggested was html5 which for those cases should be the same as xhtml
jonas’indeed
ZashThe un-safety was always in having a renderer that also executes arbitrary code, not the wire format.
danielwell i can get on board with an xhtml-im-light that is limited to what 0393 can input
ralphmSo what Slack *actually* does, I think, is parsing the plaintext server side as per 393, and then send a marked up message back, including references.
Link Mauvedaniel, XHTML-IM is already that, tbh.
danielif you want to do that in the same xep or not i don’t care
Link MauveExcept it doesn’t have disco#info features telling others what you actually do support.
ralphmThat's why Flow came up with https://xmpp.org/extensions/inbox/bmh.html, no?
Link Mauvehttps://xmpp.org/extensions/xep-0071.html#profile-summary the current recommended set is a bit bigger than what you suggest, but I don’t see much to remove from it.
jonas’Link Mauve, for starters, @style should go out of the window.
Link MauveIt mostly just adds lists, images, links and CSS.
Link MauveOh, and quotes.
jonas’(393 has quotes)
jonas’that profile lacks code though
Link MauveIndeed, that would be an useful addition.
Link MauveIndeed, that would be a useful addition.
jonas’now, if we had namespaced attributes...
Link MauveThere is one, just in this list. :p
jonas’(thinking of stuff like <code xhtml-im:language="c"/>)
jonas’yeah, xml:lang doesn’t count
Link MauveI’ve used a @class hack for that in the past.
ralphmI'd like bold, italic, strike, literal inline, literal block, bullet list and quotes.
Link MauveStrike requires CSS if we keep it to XHTML-IM.
Zashralphm: "literal inline" ?
ralphmNo color, I think, no font choices, so I guess no css.
ralphmYes, like `this`.
Zashralphm: So <code>-ish?
jonas’Link Mauve, <del/>?
Link MauveIsn’t that <code/>?
ralphmVersus: ```
Block of literal.
```.
Link Mauvejonas’, oh, indeed.
Link Mauveralphm, <pre><code/></pre>.
jonas’Link Mauve, although I’m not 100% convinced that <del/> has the right s emantics
Alexwe need a markdown extension ;-)
Link Mauve“The HTML <del> element represents a range of text that has been deleted from a document. This can be used when rendering "track changes" or source code diff information, for example.”
ralphmYeah, one block, one inline. That's the HTML terminology.
ZashAlex: Bother your client about markdown to xhtml-im support? :)
Link MauveAlex, no, we need to allow clients to use Markdown as an input format, and transmit it in some standard way (*hint* XHTML-IM) over the wire.
Zash<code xml:lang="c">int main (void) { }</code> mouhahhaaha
jonas’Zash, that hurt me a little.
ralphmMarkdown is like RSS. There are many dialects and bad interop.
Link MauveThat’s something which can’t be done using XEP-0393 due to incompatibilities in the markup used.
ralphmThat's why Slack says they don't support it.
Link Mauveralphm, hence why clients should be able to pick the one they like, and convert it to XHTML-IM.
Alexwhat I do not like in HTML-IM is that you need to validate for allowed tags and scripts
jonas’fun fact: with namespaced attributes we could get references right
Zashjonas’: We need to get computer languages their own language tags!
Alexotherwise clients are volnerable
ralphmAlex: you need to validate any format
Link MauveAlex, no, you need to pick a subset and convert it to whichever rendering library you’re using.
ZashAlex: And you are aware that Markdown allows HTML pass-through?
Link MauveThat’s very different from validation.
jonas’Alex, not really. only if you pass the XHTML-IM 1:1 to the (code executing) renderer. If you did what you would do for any other format (parse and re-emit as HTML), it would be fine.
ralphmZash: and that
Link Mauvejonas’, let’s just use namespaced attributes, and see what breaks!
Link MauveI’d bet on not much, if anything.
ZashThe real problem is the distinct lackof a renderer without mandatory JS support and remote content fetching
jonas’(thinking of <body xmlns="xhtml" xmlns:xhtml-im="..."><span xhtml-im:id="abc">@Link Mauve</span></body><references><ref id="abc"><jid>linkmauve@...</jid></ref></references>)
jonas’although that could probably also be solved with some href actually
jonas’but you get the idea
jonas’(same for blockquotes for example)
Link Mauvejonas’, looks very nice, but @id already exists in XHTML, we could just add it to the profile.
jonas’Link Mauve, @id is even in XML and has some semantics I’m afraid
Link Mauve@xml:id is different from @xhtml:id in semantics.
Link MauveHere, you create a new XHTML document in each message (possibly more than one even, with @xml:lang).
jonas’right
jonas’that could actually work
Link MauveSo you could perfectly use @xhtml:id.
jonas’and the @xhtml:id needs to be unique anyways for references to work... or does it?
Link MauveLet’s keep it simple. :)
jonas’ok :)
Alexwhich client are you working on?
Link MauvePersonally, at the moment mostly Converse (when I’ll stop being ill), but I don’t have a fixed client to work on, I just do what I feel is useful.
ZashI should submit the XEP for the thing I did with JabberCat at some point
mightyBroccolihas left
mightyBroccolihas joined
jonas’alright, heading to bed, tele-type to you later ;-)
XSF Discussion - 2018-11-22