16:00:02 <geppetto> #startmeeting fpc 16:00:02 <zodbot> Meeting started Thu Oct 29 16:00:02 2020 UTC. 16:00:02 <zodbot> This meeting is logged and archived in a public location. 16:00:02 <zodbot> The chair is geppetto. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:02 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:00:02 <zodbot> The meeting name has been set to 'fpc' 16:00:02 <geppetto> #meetingname fpc 16:00:02 <geppetto> #topic Roll Call 16:00:02 <zodbot> The meeting name has been set to 'fpc' 16:00:12 * limburgher here 16:00:23 * carlwgeorge waves 16:00:38 * decathorpe is here 16:00:41 <geppetto> #chair limburgher 16:00:41 <zodbot> Current chairs: geppetto limburgher 16:00:44 <geppetto> #chair carlwgeorge 16:00:44 <zodbot> Current chairs: carlwgeorge geppetto limburgher 16:00:46 <geppetto> #chair decathorpe 16:00:46 <zodbot> Current chairs: carlwgeorge decathorpe geppetto limburgher 16:01:40 <tibbs> Hey. 16:01:47 <geppetto> #chair tibbs 16:01:47 <zodbot> Current chairs: carlwgeorge decathorpe geppetto limburgher tibbs 16:02:38 <mhroncok> hey there 16:02:49 <mhroncok> I am here but I am distracted and will have to leave soon 16:03:17 <geppetto> #chair mhroncok 16:03:17 <zodbot> Current chairs: carlwgeorge decathorpe geppetto limburgher mhroncok tibbs 16:03:17 <mhroncok> sorry about that, dealing with some missing mortgage paperwork 16:03:28 <geppetto> Ok, no problem … probably going to be a quick meeting agian 16:03:49 <geppetto> #topic Schedule 16:03:53 <geppetto> #link https://lists.fedoraproject.org/archives/list/packaging@lists.fedoraproject.org/message/VI4C336E3WHPTOZEUCTHRD63XM3PZY3N/ 16:04:00 <geppetto> #topic Open Floor 16:04:26 <geppetto> So no tickets … and I added the actions going back a few months … I think a couple of them are done, but wasn't 100% sre 16:04:28 <decathorpe> it would be great if people could vote on my Rust Guidelines update PR :) 16:05:01 <geppetto> decathorpe: Link? 16:05:26 <decathorpe> https://pagure.io/packaging-committee/pull-request/1026 16:05:44 <carlwgeorge> from that email, #topic #1020 is complete and merged 16:08:56 <tibbs> The rust stuff seems sane (and I'm really surprised the example spec used %__install) but I know very little about rust. 16:09:30 <tibbs> And this reminds me that Igor seems to have disappeared into the æther. 16:10:04 <decathorpe> well I hope he's not into the æther, but he's been unresponsive for a few weeks 16:10:59 <King_InuYasha> .hello ngompa 16:11:00 <zodbot> King_InuYasha: ngompa 'Neal Gompa' <ngompa13@gmail.com> 16:11:01 * King_InuYasha waves 16:11:02 <King_InuYasha> whoops 16:11:08 <geppetto> seems fine to me 16:11:15 <geppetto> #chair King_InuYasha 16:11:15 <zodbot> Current chairs: King_InuYasha carlwgeorge decathorpe geppetto limburgher mhroncok tibbs 16:11:39 <King_InuYasha> guess I missed everything :) 16:11:49 <mhroncok> i would not use %__install 16:12:24 <carlwgeorge> i believe the general guidelines already discourage/prohibit that 16:13:17 <decathorpe> are there any %__install left? 16:13:30 <King_InuYasha> there aren't supposed to be 16:13:56 * King_InuYasha is starting to wonder if we should look at adapting spec-cleaner to help people follow our guidelines... 16:14:22 <geppetto> I can't see any with search 16:14:33 <decathorpe> nope there are none left 16:14:49 <decathorpe> at least in the Rust pages (TIL that ripgrep is awesome) 16:17:51 <geppetto> Ok, anyone else want to vote in the pr … or we just merging it? 16:18:09 * mhroncok hadn't had time to read it 16:18:14 <geppetto> ok 16:18:16 <mhroncok> if you can vote w/out me, good 16:18:19 <decathorpe> it's not urgent 16:18:33 <decathorpe> the changes will only really be necessary once F34 is branched 16:18:52 <geppetto> carlwgeorge: You did the wording for 1020, right? 16:18:53 <King_InuYasha> we should just merge the rust policy changes 16:18:56 * mhroncok doesn't understand the commend above License 16:18:58 <King_InuYasha> they are approved at this point 16:19:17 <mhroncok> *comment 16:20:03 <tibbs> mhroncok: I think the upstream metadata includes a license field which must be translated into the Fedora equivalent. 16:20:11 <mhroncok> also, why is echo '/usr/bin/asciidoctor' preferred over specifying it as manual BuildRequires 16:20:37 <decathorpe> that's a quirk of rust2rpm 16:20:44 <carlwgeorge> geppetto: yes, and it's already merged 16:21:16 <mhroncok> This one I don't understand 16:21:17 <mhroncok> + # * ASL 2.0 or Boost 16:21:17 <mhroncok> + # * ASL 2.0 or MIT 16:21:17 <mhroncok> + # * MIT 16:21:17 <mhroncok> + # * MIT or ASL 2.0 16:21:17 <mhroncok> + # * Unlicense or MIT 16:21:17 <mhroncok> + License: MIT and (Boost or ASL 2.0) 16:21:21 <decathorpe> regarding License, any subpackages that contain Rust binaries need a separate License tag with the resulting license of statically linking everything 16:21:30 <mhroncok> # Upstream license specification: MIT OR Apache-2.0 -> this one I do 16:22:20 <mhroncok> is the license tag a combination of the above? how so? 16:22:45 <decathorpe> basically boolean logic. at least that's how it was explained to me 16:23:09 <mhroncok> boolean logic in legal, can we do that 16:23:23 <mhroncok> also unlicense is skipped becasue it allows anything? 16:23:42 <decathorpe> I mean, in this example: MIT must be in the License tag since one of the dependencies is MIT-only. 16:23:48 <decathorpe> this covers the last three lines 16:23:56 <decathorpe> (last four) 16:24:04 <decathorpe> the rest is (ASL 2.0 or Boost) 16:24:06 <mhroncok> I ma not sure it works that way 16:24:14 <mhroncok> but maybe it does 16:24:15 <tibbs> Honestly I don't know. The License: tag there seems to be an allowable distillation of the licenses of the source files, but I don't see why it would be reduced. 16:24:16 <carlwgeorge> yeah that doesn't sound right 16:24:43 <decathorpe> well "OR" does mean that you can choose either license, right? 16:24:56 <mhroncok> yes, but as a packager, you try to preserve both options 16:25:00 <mhroncok> if legally possible 16:25:32 <decathorpe> well, this basically drops the impossible options 16:25:33 <carlwgeorge> i would think that one would have to be `(ASL 2.0 or Boost) and (ASL 2.0 or MIT) and MIT and (Unlicense or MIT)` 16:25:44 <mhroncok> I agree 16:26:18 <carlwgeorge> the only thing to deduplicate would be the double `ASL 2.0 or MIT` 16:26:39 <tibbs> We don't really have legal analysis of compatible licenses except in the case of the GPL family. 16:27:14 <decathorpe> in this example, why do you need to include Unlicense? MIT is already required, and adding Unlicense doesn't add anything 16:28:43 <carlwgeorge> https://docs.fedoraproject.org/en-US/packaging-guidelines/LicensingGuidelines/#_combined_dual_and_multiple_licensing_scenario doesn't mention anything about trimming licenses out 16:29:22 <decathorpe> don't ask me why Rust packaging does it this way. it's done like this for all Rust binary packages 16:29:49 <decathorpe> it just wasn't included in the Packaging Guidelines before 16:30:08 <tibbs> It's really a question for the legal folks. 16:30:19 <decathorpe> right 16:30:56 <decathorpe> tough, I mean, MIT and (MIT or Unlicense) compared to MIT does not give you any more freedom 16:31:02 <decathorpe> it's just noise 16:31:40 <decathorpe> it's not like you can decompile the binary to get the Unlicense bits out 16:31:41 <tibbs> I don't see much point in getting too hung up on it here as it seems to be just a detail in a specfile example. 16:31:45 <geppetto> I agree … but I'm not legal folk :) 16:32:09 <King_InuYasha> decathorpe: spot asked us to do it 16:32:10 <King_InuYasha> that's why 16:32:15 <mhroncok> I don't feel comfortable having this in the guidelines w/out legal blessings 16:32:16 <tibbs> The "License for binary packages" doesn't appear to talk about this kind of reduction. 16:32:36 <mhroncok> King_InuYasha: do you have a link? 16:32:53 <carlwgeorge> 100% agree about deferring to spot on this 16:33:11 <mhroncok> spot doesn't do this any more 16:33:17 <King_InuYasha> https://src.fedoraproject.org/rpms/rust-rpick/pull-request/1#comment-34760 16:33:55 <mhroncok> King_InuYasha: "list the licenses of all of the code inside the static binary" is very much what I agree with 16:34:19 <mhroncok> "reduce the license list to a minimal form that is compatbile" not so much 16:34:30 <King_InuYasha> we probably can do compatible reduction 16:34:36 <King_InuYasha> we do this already for everything else anyway 16:34:42 <decathorpe> this is what was eventuelly committed to rpick: https://src.fedoraproject.org/rpms/rust-rpick/blob/master/f/rust-rpick.spec#_32-38 16:35:02 <mhroncok> I don't think this can be solved here 16:35:44 <King_InuYasha> in any case, at least now y'all know why 16:36:24 * geppetto nods 16:36:46 <geppetto> Ok, I'm going to close the meeting … let us al go do other things and not think about legal problems ;) 16:37:04 <geppetto> But one last thing … I won't be around next week 16:37:23 <geppetto> We can probably skip, but if anyone else wants to run te meeting feel free to volunteer 16:37:55 <decathorpe> skipping is fine with me, there's not much to discuss 16:38:50 * geppetto nods 16:39:21 <geppetto> Well anyone else feel free to email the list, or whatever 16:39:27 <geppetto> #endmeeting