fpc
LOGS
17:00:06 <geppetto> #startmeeting fpc
17:00:06 <zodbot> Meeting started Thu Jan  3 17:00:06 2019 UTC.
17:00:06 <zodbot> This meeting is logged and archived in a public location.
17:00:06 <zodbot> The chair is geppetto. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:06 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
17:00:06 <zodbot> The meeting name has been set to 'fpc'
17:00:06 <geppetto> #meetingname fpc
17:00:06 <geppetto> #topic Roll Call
17:00:06 <zodbot> The meeting name has been set to 'fpc'
17:01:49 * limburgher here
17:01:52 <mhroncok> hi hi
17:01:57 <geppetto> #chair limburgher
17:01:57 <zodbot> Current chairs: geppetto limburgher
17:02:01 <geppetto> #chair mhroncok
17:02:01 <zodbot> Current chairs: geppetto limburgher mhroncok
17:02:03 <geppetto> hey
17:02:16 <ignatenkobrain> .hello2
17:02:17 <zodbot> ignatenkobrain: ignatenkobrain 'Igor Gnatenko' <i.gnatenko.brain@gmail.com>
17:02:23 <geppetto> #chair ignatenkobrain
17:02:23 <zodbot> Current chairs: geppetto ignatenkobrain limburgher mhroncok
17:04:03 <decathorpe> hello 😁
17:04:11 <geppetto> #chair decathorpe
17:04:11 <zodbot> Current chairs: decathorpe geppetto ignatenkobrain limburgher mhroncok
17:04:18 <geppetto> hey, number 5 :)
17:04:33 <geppetto> 👍
17:05:01 <decathorpe> (glad I could make it, IRC client kept crashing)
17:05:17 <geppetto> #topic Schedule
17:05:20 <geppetto> #link https://lists.fedoraproject.org/archives/list/packaging@lists.fedoraproject.org/message/2FXUUFLJGXIICBF64S563C7DK4KMXULM/
17:05:35 <geppetto> #topic #841 Change or remove "Want to help" footer
17:05:45 <geppetto> .fpc 841
17:05:46 <zodbot> geppetto: Issue #841: Change or remove "Want to help" footer - packaging-committee - Pagure.io - https://pagure.io/packaging-committee/issue/841
17:05:49 <tibbs> Hey, folks.
17:05:59 <tibbs> Just got back to work; it's busy here.
17:06:21 <decathorpe> #chair tibbs
17:06:21 <zodbot> Current chairs: decathorpe geppetto ignatenkobrain limburgher mhroncok tibbs
17:06:26 <geppetto> #chair tibbs
17:06:26 <zodbot> Current chairs: decathorpe geppetto ignatenkobrain limburgher mhroncok tibbs
17:06:31 <redi> hi
17:06:35 <geppetto> decathorpe: You win :)
17:06:36 <tibbs> I filed that ticket just so I wouldn't lose track of it.
17:06:38 <decathorpe> sorry ;)
17:06:41 <geppetto> #chair redi
17:06:41 <zodbot> Current chairs: decathorpe geppetto ignatenkobrain limburgher mhroncok redi tibbs
17:07:13 <geppetto> tibbs: yeh, it's fine … if anyone else knows how it's there they can speak now … redi?
17:07:57 <ignatenkobrain> This will need some work with docs team
17:07:59 <tibbs> I need to try #fedora-docs again; I don't really know the best way to contact the docs folks.
17:08:28 <ignatenkobrain> I'll ping bexelbie and asamalik myself afterwards/)
17:08:33 <tibbs> It's possible that we can just override some default, or maybe we can't.  I don't know quite enough about how that footer gets added.
17:09:10 <tibbs> I did figure out how to do macros for fedora versions like the wiki had.
17:09:57 <decathorpe> that's nice
17:10:07 <tibbs> Anyway, no reason to spend much time on 841.
17:10:21 <geppetto> ok, not much else though :)
17:10:29 <geppetto> #topic #719 Simplify packaging of forge-hosted projects
17:10:33 <geppetto> .fpc 719
17:10:35 <zodbot> geppetto: Issue #719: Simplify packaging of forge-hosted projects - packaging-committee - Pagure.io - https://pagure.io/packaging-committee/issue/719
17:10:56 <geppetto> So this hasn't updated since we last talked about it
17:11:01 <mhroncok> sorry I was discrated, reading the backlog
17:11:27 <decathorpe> I haven't managed to write the accompanying text yet ... holidays, short vacation, being ill ... you know how it is
17:11:47 <geppetto> Which is kind of expected given the dates involved … no worried decathorpe
17:12:07 <geppetto> #topic Open Floor
17:12:20 <geppetto> So … anything else we want to talk about this week?
17:12:36 <tibbs> Well, there are a couple of things.
17:12:41 <tibbs> .fpc 840
17:12:45 <zodbot> tibbs: Issue #840: Broken links in JavaScript guidelines - packaging-committee - Pagure.io - https://pagure.io/packaging-committee/issue/840
17:13:30 <limburgher> Eek
17:13:45 <decathorpe> do you mean to say that the documentation is broken, or that JavaScript packaging is broken? ;)
17:13:46 <tibbs> I cleaned up the page as best I could, but that page is weird because it tried to address the javascript bundling situation in a way that no longer makes sense after FESCo changed the bundling rules.
17:14:05 * geppetto nods
17:14:21 <geppetto> Removing that section and your proposed wording seem fine to me
17:14:33 <mhroncok> "I'm tempted to just remove all mention of bundling from the page and let the existing bundling policy dominate, perhaps leaving only a note that bundling in javascript is extremely common and linking to the main policy." +1 to that
17:14:37 <ignatenkobrain> +1 to what tibbs proposed in pagure
17:15:05 <tibbs> That would involve removing the text of the "Static Inclusion of Libraries" section and replacing it with something like:
17:15:43 <tibbs> It is extremely common for a single minified JavaScript file to contain bundled code from other JavaScript libraries.  Please refer to <xxx> for the proper way to handle this.
17:16:23 <geppetto> +1
17:16:35 <limburgher> +1
17:16:42 <tibbs> Need to check to see what in the distro actually provides "js-includes.*" and fix them to provide bundled(xxx) instead, I guess.
17:16:51 <decathorpe> +1
17:17:01 <tibbs> +1, obviously.
17:18:44 <tibbs> Doesn't look like "repoquery --whatprovides 'js-includes*'" returns anything.  Obviously these guidelines were closely followed.
17:19:05 <mhroncok> :D
17:19:28 <tibbs> Those guidelines also link to https://fedoraproject.org/wiki/JavaScript/Minification_Issues
17:19:45 <tibbs> But I have no idea if the information there is remotely current.
17:19:53 <decathorpe> 😂
17:20:32 <redi> I'm having some hardware issues, going to have to miss this meeting
17:20:33 <tibbs> Obviously the jsmin license is still non free.  That's Crockford and his purposefully trolling license, isn't it?
17:21:39 <tibbs> Anyway, I know very little about Javascript these days; I suspect doing this right would require mentioning webpack and the like but I know even less about that.
17:22:44 <tibbs> I don't even know what Fedora folks to ask about whether what we have still makes sense.
17:23:26 <tibbs> And I went to do writeups and ran into
17:23:30 <tibbs> .fpc 784
17:23:32 <zodbot> tibbs: Issue #784: forbid use of globs for shared libraries in %files that conceal soname versions - packaging-committee - Pagure.io - https://pagure.io/packaging-committee/issue/784
17:23:37 <tibbs> which has no announcement text.
17:24:05 <tibbs> But that's redi, who unfortunately just dropped off.
17:24:15 <decathorpe> nope, that was me
17:24:39 <decathorpe> should I write 2-3 sentences to explain the change and the reasons for it?
17:24:55 <tibbs> Please.
17:25:10 <decathorpe> will do
17:25:17 <tibbs> I could certainly do it but I've forgotten all of the context.
17:25:36 <tibbs> And the one other thing....
17:25:50 <tibbs> .fpc 667
17:25:52 <zodbot> tibbs: Issue #667: Recommend use of systemd sandboxing directives - packaging-committee - Pagure.io - https://pagure.io/packaging-committee/issue/667
17:26:09 <tibbs> This one fell through the cracks.
17:26:31 <tibbs> At the time I thought it was just "apply these changes" but... it's not.
17:26:51 <tibbs> https://fedoraproject.org/wiki/User:Zbyszek/ProtectionsPolicyDraft was the draft
17:27:14 <tibbs> but that's not really a "draft".  It seems to be more a summary of the systemd documentation.
17:28:03 <decathorpe> so this is about patching systemd unit files?
17:28:11 <tibbs> Yes.
17:28:33 <tibbs> I think the bit under "Proposed FESCo decision" is the part we're supposed to implement.
17:28:56 <tibbs> But it's not clear to me where that is supposed to go.
17:29:10 <decathorpe> right
17:29:35 <tibbs> And our existing guidelines in https://docs.fedoraproject.org/en-US/packaging-guidelines/Systemd/ also read very much like documentation.
17:30:19 <tibbs> At the time they were adopted, I don't think much of FPC knew all that much about systemd; it was very new, after all.
17:31:05 <tibbs> I can try to put something together and send a PR, I guess, but it will need to be reviewed by someone who actually understands the details.
17:31:15 <decathorpe> yeah, this page reads like a Guide how to write your one unit file for packages that don't ship one
17:32:41 <tibbs> It's not an uncommon event for me to start to do a writeup of something that isn't really a draft and then realize that the whole page needs an overhaul.
17:33:17 <tibbs> Fortunately now that drafts can be PRs, there is less burden on me to figure out details like where something is supposed to go.
17:33:42 <tibbs> But we should do something soon since FESCo has asked is to deal with it.
17:34:39 <tibbs> Anyway, I'll work on it.
17:34:56 <tibbs> And get an announcement out as soon as can.
17:35:18 <geppetto> ok
17:35:29 <geppetto> I hope you all had a nice holiday :)
17:35:34 <tibbs> I think that's all I had.  Will close off all of the tilde-related tickets as soon as I send that announcement.
17:35:43 <decathorpe> great :)
17:35:45 <tibbs> That will get our total down to a more manageable level.
17:35:54 * geppetto nods
17:36:03 <decathorpe> I have another two things (one small, one big)
17:36:20 <geppetto> ok
17:36:27 <decathorpe> .fpc 693
17:36:29 <zodbot> decathorpe: Issue #693: Wiki:Packaging:RPMMacros - packaging-committee - Pagure.io - https://pagure.io/packaging-committee/issue/693
17:36:49 <decathorpe> this somehow fell through the cracks. I thought we merged this?
17:38:09 <tibbs> I don't think so.
17:38:20 <decathorpe> my PR is also gone, probably because I deleted my fork after thinking that it got merged :/
17:39:12 <decathorpe> I can dig up my changes and re-send a PR
17:39:42 <geppetto> ok, you still have the work locally?
17:40:33 <tibbs> Surely the PR would still exist but show as cancelled or something.
17:41:29 <decathorpe> no idea. I couldn't find it. I find the new pagure interface confusing ...
17:41:39 <mhroncok> https://pagure.io/packaging-committee/pull-requests?status=all&author=decathorpe
17:41:45 <mhroncok> I don't see it
17:42:11 <decathorpe> ok, I'll submit my changes (again)
17:42:24 <mhroncok> decathorpe: there was no PR, just a diff link
17:42:29 <mhroncok> and that is now 404
17:43:30 <decathorpe> I have the changes locally, I'll just need to submit a PR (again, or for real, this time)
17:44:57 <decathorpe> second thing: tibbs brought something up that I've been thinking about too. some of the Guideline pages are either completely outdated or became irrelevant ... it'd be really nice if we could start a "Project" to clean that stuff up, maybe also ask relevant "experts" for input on current best practices for different things?
17:45:30 <tibbs> Yes, there are dark corners that just don't get looked over very often.
17:46:49 <tibbs> asciidoc supports plenty of document metadata.
17:47:11 <tibbs> We should probably leverage that to indicate when documents were last reviewed for correctness.
17:47:26 <decathorpe> yeah that'd be a good start
17:47:32 <geppetto> yeh
17:47:34 <tibbs> Then it should be trivial to find the documents which haven't been looked at in a few years.
17:48:16 <tibbs> https://asciidoctor.org/docs/asciidoc-writers-guide/#document-attributes for example
17:48:49 <tibbs> Though... attributes are more like macros.
17:48:52 <geppetto> cool
17:50:03 <tibbs> Anyway, this deserves a ticket, I guess.
17:50:20 <decathorpe> I'll open one
17:50:26 <geppetto> sure, can't hurt
17:50:53 <geppetto> I think that's it though, for this week.
17:51:02 <tibbs> We don't have to actually use the attributes in the document; we can just have :lastreviewed: 2019-01-01 and then grep that out.
17:51:30 <tibbs> I did also make a little wiki template and add it to most of the old wiki pages.
17:51:52 <tibbs> Need to finish doing that.
17:52:01 <mhroncok> tibbs: any plans for actually removing the content so people need to stop using the wiki ?
17:52:18 <mhroncok> otherwise I'm afraid they will still read the wiki and they may get outdated info
17:52:20 <tibbs> I didn't want to do that until we're sure all of the pages are actually in good shape.
17:52:26 <mhroncok> makes sense
17:52:37 <mhroncok> maybe we can do it page by page - together with shape review
17:52:55 <tibbs> Yes, first should just be the main page and the ones which have received the most work lately.
17:53:15 <decathorpe> that sounds like a good plan
17:53:37 <mhroncok> tibbs: so if I see that asciidoc page looks OK, I may remove the wiki content to indicate that? or would you like not to do that yet?
17:53:55 <tibbs> Let's let decathorpe open that ticket and then we can coordinate there.
17:54:38 <mhroncok> ack
17:54:57 <tibbs> And I wouldn't delete the pages from the wiki.  We'll just leave the header in place saying that the new content is ---> that way.
17:55:18 <mhroncok> sure, I meant that
17:55:25 <mhroncok> removing just the content, leaving the header
17:55:50 <tibbs> But you're right; we do need to get rid of the old content because people might never notice the header saying that it's old.
17:56:33 <tibbs> Anyway, lots to do it seems.
17:57:03 * geppetto nods
17:57:03 <decathorpe> something like a New Year's resolution ;)
17:57:10 <tibbs> I've been getting more fluent with asciidoc (tables, in particular) and it's... overcomplicated but powerful.
17:57:36 <tibbs> I still get pain typing `+  +` all the time.
17:57:50 <decathorpe> yeah that's awful ...
17:58:12 <tibbs> It's less painful to type `pass: ` but more verbose.
17:58:23 <mhroncok> that does the same?
17:58:26 <tibbs> Yes.
17:58:53 <tibbs> asciidoc seems to be a bit like perl.  There's generally about five ways to accomplish anything.
17:59:28 <geppetto> :)
18:00:02 <geppetto> Anyway, we've hit time now … thanks for coming everyone.
18:00:03 <tibbs> [cols=",,",options="header",] is what  the converter did for a wiki table.
18:00:06 <geppetto> #endmeeting