Monday, March 20, 2017
editor@muc.xmpp.org
March
Mon Tue Wed Thu Fri Sat Sun
    1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27 28 29 30 31    
             
XEP Editor Team — logs: http://logs.xmpp.org/editor/

[00:05:54] *** Holger shows as "away" and his status message is "Auto-away (idle)"
[03:00:11] *** SamWhited shows as "online"
[06:15:17] *** SamWhited has left the room
[06:36:47] *** winfried shows as "online" and his status message is "I'am on the move, blame the LTE if I don't react"
[06:41:13] *** Holger shows as "online" and his status message is "I'm available"
[06:51:33] *** winfried shows as "away" and his status message is "sssttt... my computer fell asleep"
[07:01:20] *** Holger shows as "away" and his status message is "Auto-away (idle)"
[07:14:11] *** winfried shows as "online" and his status message is "I'am on the move, blame the LTE if I don't react"
[07:18:27] *** Holger shows as "online" and his status message is "I'm available"
[07:33:24] *** winfried has left the room
[07:52:44] *** Holger shows as "away" and his status message is "Auto-away (idle)"
[08:25:36] *** winfried shows as "online"
[08:33:25] *** Kev shows as "online"
[08:50:20] *** winfried shows as "online" and his status message is "I'am on the move, blame the LTE if I don't react"
[08:52:23] *** winfried shows as "away" and his status message is "sssssttttt! my computer fell asleep"
[09:02:23] *** winfried shows as "xa" and his status message is "sssssttttt! my computer fell asleep"
[09:04:53] *** winfried shows as "away" and his status message is "sssttt... my computer fell asleep"
[09:05:00] *** winfried shows as "online" and his status message is "I'am on the move, blame the LTE if I don't react"
[09:47:23] *** winfried shows as "away"
[10:14:36] *** Holger shows as "online" and his status message is "I'm available"
[10:14:36] *** Holger shows as "online" and his status message is "I'm available"
[10:39:40] *** winfried shows as "away" and his status message is "sssttt... my computer fell asleep"
[10:39:41] *** winfried shows as "online" and his status message is "I'am on the move, blame the LTE if I don't react"
[10:44:44] *** winfried shows as "away" and his status message is "sssttt... my computer fell asleep"
[10:44:45] *** winfried shows as "online" and his status message is "I'am on the move, blame the LTE if I don't react"
[11:00:48] *** winfried shows as "away" and his status message is "sssttt... my computer fell asleep"
[11:03:35] *** winfried shows as "online" and his status message is "I'am on the move, blame the LTE if I don't react"
[11:25:17] *** winfried shows as "away" and his status message is "sssttt... my computer fell asleep"
[11:39:48] *** Holger shows as "away" and his status message is "I'm away"
[11:39:48] *** Holger shows as "online" and his status message is "I'm available"
[11:50:17] *** winfried shows as "xa" and his status message is "sssttt... my computer fell asleep"
[12:02:54] *** winfried shows as "online" and his status message is "I'am on the move, blame the LTE if I don't react"
[12:05:46] *** Kev shows as "away"
[12:09:30] *** Kev shows as "online"
[12:15:59] *** Holger shows as "online" and his status message is "Available"
[12:16:03] *** Holger has left the room
[12:34:12] *** Kev shows as "away"
[12:48:00] *** Holger shows as "away" and his status message is "Auto-away (idle)"
[12:59:14] *** winfried shows as "away" and his status message is "sssttt... my computer fell asleep"
[12:59:48] *** Holger shows as "online" and his status message is "I'm available"
[13:08:54] *** winfried shows as "online" and his status message is "I'am on the move, blame the LTE if I don't react"
[13:25:43] *** Holger shows as "away" and his status message is "Auto-away (idle)"
[13:31:21] *** winfried has left the room
[13:41:52] *** Flow has joined the room
[13:53:16] *** Kev shows as "online"
[13:53:27] *** Flow shows as "away" and his status message is " (Away as a result of being idle more than 5 min)"
[13:55:41] *** SamWhited shows as "online"
[13:56:17] *** Flow shows as "online"
[14:16:00] *** winfried shows as "online"
[14:45:03] *** winfried shows as "away" and his status message is "sssssttttt! my computer fell asleep"
[14:46:20] *** winfried shows as "online"
[15:00:59] *** Holger shows as "online" and his status message is "I'm available"
[15:33:19] *** winfried shows as "away" and his status message is "sssssttttt! my computer fell asleep"
[15:43:18] *** winfried shows as "xa" and his status message is "sssssttttt! my computer fell asleep"
[15:59:45] <Flow> SamWhited: Are you processing the ISR-SASL2 protoXEP or shall I?
[16:15:11] <SamWhited> Flow: Go ahead; you sent an update that was still broken. Please make sure CI passes before merging.
[16:26:23] *** Flow shows as "away" and his status message is " (Away as a result of being idle more than 5 min)"
[16:28:18] *** winfried shows as "away"
[16:35:16] *** Holger shows as "away" and his status message is "Auto-away (idle)"
[16:36:13] *** Flow shows as "online"
[16:42:01] <Flow> SamWhited: Does CI also run for inbox?
[16:42:47] <SamWhited> Flow: No, I couldn't get building things in the inbox to work
[16:43:10] *** Holger shows as "online" and his status message is "I'm available"
[16:53:45] <Flow> SamWhited: https://github.com/xsf/xeps/pull/454/commits/314b0207d109d8fbc13725d964f2d46a8d6d8784
If there are no objections, then I'm going to merge it
[16:54:06] <SamWhited> Flow: I think adding stuff to link is going to break things
[16:54:07] <Flow> turns out it was the dtd who is broken
[16:54:41] <SamWhited> I'm not sure that it is; I don't fully understand the consequences of that change, but I doubt stuff in link other than text was left out purely by mistake
[16:57:37] <SamWhited> Flow: In this particular case, why don't you just use &xep0198;?
[16:57:54] <SamWhited> (and not link directly to the section since it's in the text anyways)
[16:57:57] <Flow> sylistic reasons
[16:58:09] <Flow> It work for HTML, i.e. it does what you would expect
[16:58:16] <SamWhited> Does it work for PDFs?
[16:58:20] <SamWhited> And references?
[16:58:29] <Flow> there are no references
[16:58:37] <Flow> it's a direct link
[16:58:48] <Flow> and I'm missing texml here locally to generate the pdf
[16:59:12] <SamWhited> You should probably just follow the DTD then if you can't test the changes
[16:59:38] <Flow> or I test the changes, and if it works I change the dtd
[16:59:56] <SamWhited> sure, although we should probably still ask for guidance; I just don't have confidence that this won't break things.
[17:00:06] <SamWhited> I'm sure somebody understands our build process better though
[17:00:09] <SamWhited> Kev maybe
[17:21:40] *** SamWhited has left the room
[17:24:49] <Flow> So I build the PDF successfully, I don't think there is a reason not to merge this. And even discover later issues, we can revert the change.
[17:25:49] <Flow> uh, and it appears that the "put the reference only once" change does not work for PDFs
[17:36:27] *** SamWhited shows as "online"
[17:36:56] <SamWhited> In PDFs references are end notes not footnotes, so it doesn't need to put them only once
[17:37:07] <SamWhited> err, footnotes not end notes, I mean
[17:37:32] *** SamWhited shows as "online"
[17:37:57] <Flow> SamWhited: same problem: If you have multiple references on the same page you end up with multiple footnotes of the same reference on the same page
[17:38:09] <SamWhited> Ah, sorry, I see; yah, you're right
[17:38:20] <SamWhited> I thought it actually was fixed in that case, but I guess not
[17:38:28] <SamWhited> (as in, I thought LaTeX deduped them for you)
[17:38:47] <Flow> latex certainly does, I guess your input is just different references
[17:41:56] <Flow> SamWhited: So, since I was able to generate the HTML and PDF, I'm going to merge 454 and if things really fall apart, what I don't expect, we can revert.
[17:43:57] <SamWhited> I still think we should ask someone who knows what they're doing first or think about it; I don't understand our build process, so maybe I'm just being paranoid, but allowing stuff in links still seems like a huge change.
[17:44:07] <SamWhited> think about it for a while first, I mean
[17:44:40] <SamWhited> but ¯\_(ツ)_/¯ I guess if it works it's up to you
[17:47:25] <Flow> SamWhited: the build process in this case was gen.py on perseus
[17:48:38] <SamWhited> Oh, gen.py doesn't actually check anything against the DTD IIRC
[17:52:19] <Flow> that's right, and IMHO should be changed. But I also did a 'make'. I still think it would be vastly beneficial if we would put the tooling into a xep-tooling repo
[17:54:44] <SamWhited> Why didn't we before? I can't remember
[17:54:56] <SamWhited> Pretty sure there was something important that would break, but I don't know what it was
[17:57:45] *** Flow shows as "away" and his status message is " (Away as a result of being idle more than 5 min)"
[17:58:12] <SamWhited> Flow: I *think* I've added you to the Trello, FYI. Not 100% sure…
[17:59:46] *** Kev shows as "away"
[18:07:45] *** Flow shows as "xa" and his status message is " (Not available as a result of being idle more than 15 min)"
[18:09:05] *** winfried shows as "online"
[18:12:59] *** Flow shows as "online"
[18:13:48] <Flow> SamWhited: Nothing would break if we make it a two phase step: 1 freeze every tool in the existing xeps repo 2. once the tools in xsf-tools are ready, switch
[18:14:13] *** winfried shows as "away" and his status message is "sssssttttt! my computer fell asleep"
[18:14:17] *** winfried shows as "online"
[18:15:20] <SamWhited> That seems fine; was that why we didn't do it before?
[18:19:50] *** winfried shows as "away" and his status message is "sssssttttt! my computer fell asleep"
[18:20:05] *** winfried shows as "online"
[18:24:24] <Flow> SamWhited: I don't remember, maybe someone was opposing the idea to split code and data in two repos. I like the idea
[18:24:46] <Flow> I remember mopharisisbest offering his help
[18:25:14] *** winfried shows as "away" and his status message is "sssssttttt! my computer fell asleep"
[18:26:55] <SamWhited> I think whomever offered to help wanted to do a rewrite instead of just splitting it out into a new repo; maybe scope creep is why it didn't happen. That's why tooling improvements normally don't happen.
[18:29:56] *** Flow shows as "away" and his status message is " (Away as a result of being idle more than 5 min)"
[18:32:03] <SamWhited> Also it would need someone on the infrastructure team side to fix the build process since the tooling wouldn't just be in the repo anymore
[18:32:30] <SamWhited> Which I suspect would involve getting rid of the hg repo thing (which would make me very happy, but we'd still have to find someone to do that)
[18:35:14] *** winfried shows as "xa" and his status message is "sssssttttt! my computer fell asleep"
[18:39:56] *** Flow shows as "xa" and his status message is " (Not available as a result of being idle more than 15 min)"
[18:41:14] *** Holger shows as "away" and his status message is "I'm away"
[18:49:44] *** Flow shows as "online"
[18:50:39] <Flow> SamWhited: why do we need someone from infra?
[18:50:50] <Flow> We already do most of the build and publishing process by hand
[18:54:53] <SamWhited> Flow: If we move gen.py out of the xeps repo, it won't be in the hg thing to run it, and anything doing automatic stuff (which I thhink exists?) will break.
[18:55:18] <SamWhited> And we should not mess with anything on perseus without Kev or someone making sure the infrastructure won't just break
[18:55:32] <Kev> Hmm? Highlight?
[18:55:36] <Flow> as I said, we won't move anything in the first step
[18:56:06] <Flow> I'm also unsure if there exists anything that runs gen.py automatically
[18:56:40] <SamWhited> Flow: Not necessarily gen.py; any tool, doesn't matter. We are unsure, that's correct, and that's why we'd need to pull in other people.
[18:57:03] <SamWhited> I'm not saying we shouldn't do it, just that we would need to coordinate with others.
[18:57:13] <SamWhited> Kev: Oops, sorry, just discussing moving tools out of the xeps repo again
[18:57:23] <Flow> probably, but I don't see this as an issue
[18:57:25] <SamWhited> I was just saying that we shouldn't touch anything on perseus without your buy in
[18:57:33] <Kev> Ah, ok.
[18:57:43] <Kev> Sorry, I'll be heading out in a moment.
[18:58:03] <Kev> But I'll repeat my usual comment that incremental change is good, complete rewrites are bad (and is what stopped it all last time).
[18:58:33] <SamWhited> Flow: I don't think it's a problem either, I just think it's a thing that requires doing and is worth keeping in mind, which is all I said originally.
[18:59:45] <Flow> I don't think complete rewrites are bad as long as the can act as drop in replacements
[19:00:17] <SamWhited> They are bad, because every time we have this discussion someone says they're going to do a complete rewrite, and then realizes that it would be a full time job and doesn't and nothing gets done.
[19:00:50] <Flow> and what is bad if that happens?
[19:01:33] <SamWhited> Nothing's bad if it happens, but it never does; trying it yet again and thinking it will be different this time is literally what happens every single time and for some reason people are still arguing for it.
[19:01:34] <Flow> If we have someone who says: "He I will help you" then you say, no don't do it because complete rewrites are bad
[19:01:48] <Flow> I don't buy that argument
[19:02:01] <SamWhited> No one said that, that's why someone volunteered last time, and I still don't see a complete rewrite or any other progress
[19:02:41] <Flow> last time mopharisbest volunteerd we said he can't start working because people thought it was a bad idea to split the tools into an extra repo
[19:03:15] <SamWhited> oh, maybe I'm misremembering then; I thought it was like Kev said, that we wanted to just move the tools and then people said "no, let's do a complete rewrite instead"
[19:03:35] <Flow> but anyway, what I propose is that we create an xsf-tools repo and copy the tooling from the xeps repo over
[19:03:50] <Flow> and then start working improving the tooling towards what we want in the new xsf-tools
[19:04:05] <SamWhited> yes, I think that might be a sane starting point
[19:04:35] <Kev> As long as they're tightly coupled somehow (presumably submodules), so you can always work out what version of the tools were used to build that version of the XEPs, I don't much care.
[19:04:44] <SamWhited> ahh, that's what it was
[19:04:44] <Flow> I would have done that months ago if I had been sure that this effort would lead to something which was actually used by someone in the end
[19:05:18] <Flow> I think it's trivial to add the git revisions of the involved repos into the produced artifacts
[19:05:27] <Kev> That's not the same thing.
[19:05:43] <Flow> and I note that we alrady don't know which version was used to build the artifacts
[19:06:25] <SamWhited> That's a good point; it wouldn't be a regression, it would be exactly the same as what we have today
[19:06:27] <Flow> I think I don't have any objections making xsf-tools a submodule of xeps
[19:06:33] <Kev> SamWhited: No, I don't think that's true.
[19:06:53] <Kev> Currently you can check out a commit of the XEP source and know exactly what version of the tools you need to build that commit.
[19:06:57] <Kev> (because they're tied)
[19:07:01] <Flow> that way the xsf-tools repo could try to autodetect the official xeps repo as being one level up the directory tree
[19:07:01] <SamWhited> oh, I see, right
[19:07:29] <SamWhited> sorry, I know we've had this exact discussion before, it's just been too long
[19:08:45] <Flow> so do we have consensus on 1. creating xsf-tools, 2. copying the code from xsf repo over 3. making xsf-tools/ a submodule of xsf ?
[19:09:09] <Kev> I don't think it buys us much, but I don't object to it if it's going to make incremental improvements to the tools easier for you :)
[19:09:12] <SamWhited> I don't like the idea of the submodule, but for the sake of not going through all this yet again then fuck it, if you want to do the work then go for it
[19:09:41] <SamWhited> I'm no so against it that I would object, I just doubt it will ever actually get updated.
[19:09:56] <Flow> I can't promise to start tomorrow, i've a freelance job left for this month that has priority, but after that, count me in
[19:10:53] <Flow> uh, another nit, can we go for python3?
[19:11:03] <SamWhited> Flow: Incremental improvement please
[19:11:18] <SamWhited> (I would love it to be python3 compatible, but that's yet another thing the iteam would have to work on, we'd have to compatibility test, etc.)
[19:11:21] <Flow> what does that mean exaclty?
[19:11:27] <Flow> ah com
[19:11:30] <SamWhited> Flow: Just do one thing to start, then we can think about python3
[19:11:32] <Flow> apt install python3
[19:12:27] <Flow> SamWhited: let's ask iteam if python3 on perseus would be possible
[19:12:43] <Flow> if yes, we go for python3
[19:12:51] <SamWhited> Let's move the files to a new repo first.
[19:13:08] <SamWhited> And then try experimenting with python3.
[19:20:14] *** winfried shows as "away"
[19:26:16] *** Flow shows as "away" and his status message is " (Away as a result of being idle more than 5 min)"
[19:36:16] *** Flow shows as "xa" and his status message is " (Not available as a result of being idle more than 15 min)"
[19:50:37] *** Flow shows as "online"
[20:07:33] *** winfried shows as "online"
[20:34:47] *** Flow shows as "away" and his status message is " (Away as a result of being idle more than 5 min)"
[20:44:47] *** Flow shows as "xa" and his status message is " (Not available as a result of being idle more than 15 min)"
[21:07:51] *** winfried shows as "online" and his status message is "I'am on the move, blame the LTE if I don't react"
[21:27:15] *** winfried has left the room
[21:49:11] *** Flow shows as "online"
[21:50:19] *** SamWhited has left the room
[22:04:59] *** bear has left the room
[22:13:33] *** bear has joined the room
[22:13:34] *** bear shows as "away" and his status message is "Away"
[22:16:16] *** Flow shows as "online"
[22:18:48] *** Flow has left the room
[22:23:33] *** Flow has left the room
[22:31:55] *** SamWhited shows as "online"
[23:41:00] *** winfried shows as "online" and his status message is "I'am on the move, blame the LTE if I don't react"
[23:43:58] *** winfried has left the room
[23:58:05] *** winfried has left the room