Perl 6 - the future is here, just unevenly distributed

IRC log for #bioperl, 2017-09-14

| Channels | #bioperl index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:54 ilbot3 joined #bioperl
01:54 Topic for #bioperl is now Be patient! People chat here daily, but not necessarily the minute or hour you wandered in. Leave your IRC client connected. | http://bioperl.org/wiki/IRC#Getting_help | http://www.bioperl.org/wiki/Using_Git | nopaste to gist.github.com
03:37 pyrimidi_ joined #bioperl
03:55 pyrimidine joined #bioperl
04:06 pyrimidine joined #bioperl
04:07 pyrimidine joined #bioperl
04:09 pyrimidine joined #bioperl
04:14 pyrimidine joined #bioperl
04:18 pyrimidi_ joined #bioperl
04:21 pyrimidine joined #bioperl
04:24 pyrimidi_ joined #bioperl
04:29 pyrimidine joined #bioperl
04:35 pyrimidine joined #bioperl
04:40 pyrimidine joined #bioperl
04:45 pyrimidine joined #bioperl
04:54 pyrimidine joined #bioperl
05:05 pyrimidine joined #bioperl
05:17 pyrimidine joined #bioperl
05:25 pyrimidine joined #bioperl
07:37 arbenina joined #bioperl
07:49 megaLUL joined #bioperl
07:50 megaLUL joined #bioperl
07:50 Tuna_Golem joined #bioperl
07:59 Tuna_Golem joined #bioperl
11:04 scottcain joined #bioperl
13:21 carandraug pyrimidine: hi
13:26 arbenina_ joined #bioperl
13:29 pyrimidi_ joined #bioperl
14:13 pyrimidine joined #bioperl
14:16 pyrimidine morning
14:18 pyrimidine carandraug: hope the emergency situation yesterday resolved itself well :)
14:20 carandraug pyrimidine: yes. The PSU failure. All good again
14:21 carandraug s/The/A/
14:22 carandraug one of our experimental microscopes which needs 5 PC's, 4 of which are 15 years old machines
14:28 carandraug do you have time to continue yesterday's discussion?
14:35 pyrimidine yep
14:36 pyrimidine you may have to ping me by my nick so the message pops up, I'm in the middle of a microbiome analysis but can do both
14:37 carandraug pyrimidine: cool. So what modules should go into the core distribution?
14:38 pyrimidine Root, interfaces,
14:38 pyrimidine maybe the basic implementations (but only those w/o introduced dependencies)
14:39 pyrimidine Seq, SeqFeature, Annotation
14:39 pyrimidine There is one significant problem that will be hard to address: Bio::Species
14:40 carandraug all of the interfaces? Some interfaces have already been moved to their own distribution
14:41 pyrimidine I think the key interfaces like SeqI, SeqFeatureI, AnnotationI
14:42 carandraug for example the interfaces to coordinate and biblio are in their own distribution. You mentioned yesterday that you also want the interfaces in bioperl-run to remain separate.
14:42 carandraug why is Bio::Species hard to address?
14:42 pyrimidine Bio::Tree
14:43 pyrimidine Bio::Species pulls in Bio::Tree
14:44 pyrimidine Bio::Tree should really be on it's own, but Bio::Species was reimplemented to be a Bio::Tree::Node
14:44 pyrimidine use base qw(Bio::Root::Root Bio::Tree::NodeI);
14:45 carandraug so Species is dependent on Tree. Why is that an issue? And does any of them need to go into core?
14:46 pyrimidine yep, Bio::Species is used within Bio::SeqIO (GenBank/EMBL/Swissprot).  Those particular modules will be very difficult to initially pull out of core w/o raining down user/developer complaints
14:46 pyrimidine it's a time suck you don't want to immediately introduce
14:47 pyrimidine But
14:47 pyrimidine we could announce this as a large-scale reorg of the core
14:47 pyrimidine (give plenty of warning)
14:47 pyrimidine and then do it
14:48 pyrimidine if there are any complaints they should be heard out
14:48 pyrimidine if not, we can simply point to the announcement if someone complains after the fact
14:49 carandraug I don't agree completely with that statement. complaints are fine but they only need to be heard if they come with the willingness to solve the issue in another way
14:50 pyrimidine precisely
14:50 pyrimidine but you have to give them the chance to make those solutions
14:50 carandraug and I haven't seen anyone stepping forward to address the issue of bioperl's size
14:51 carandraug we did. The split has been going on discussion for more than 5 years
14:51 pyrimidine well, it's actually fairly complex due to the intermingled dependencies within the distribution
14:51 carandraug also, as long as we don't make nay code chances, no one can complain. And I have been careful to not make any changes on the new distributions I prepared
14:52 carandraug the work itself is not complex. It's complex to decide what modules should go on what distribution. But that's still not a 5 years problem
14:53 pyrimidine if the one person doing the work gets no feedback from the community and has little time to dedicate to the problem, it can easily become a 5 yr problem
14:54 pyrimidine look at how long it took to get 1.6 out
14:54 pyrimidine 1.5.2 came out in ~2004.  1.6 in 2009, 1.7 last year
14:55 pyrimidine and the only reason 1.7 came out was because we really needed a new release out
14:56 carandraug yes, I get that. Which is why I arguing that waiting for the "community" to give feedback won't get us anywhere other than delaying it further
14:56 pyrimidine Oh, I'm not expecting much feedback.
14:57 pyrimidine My whole point is that we can simply point to that post as proof we asked for feedback
14:57 pyrimidine if none was provided, we can use that as implicit consent to proceed
14:57 carandraug oh, ok. I thought that there was a post out there already
14:58 pyrimidine I posted a few things, but I think it needs to go into something a bit broader that gets more exposure (blog or similar)
14:58 carandraug but then, you need to have an actual proposal of what modules go where, to get any feedback. And that does not exist yet
14:58 pyrimidine mail lists are so 1990's :)
14:58 carandraug mailing lists is what we got. There is no bioperl blog
14:58 pyrimidine .oO ( as I'm ironically typing on irc)
14:59 pyrimidine we have the OBF news feed
14:59 carandraug mailing lists work
14:59 pyrimidine https://news.open-bio.org
14:59 carandraug ok. Whatever you think is best. BUt I would really like to move forward with making releases of the bioperl modules I care about
14:59 pyrimidine mailing lists work if people pay attention to them
15:00 carandraug and people who don't pay attention to them don't matter
15:00 pyrimidine For now, release TCoffee and Clustalw
15:00 pyrimidine let's then release a new bioperl-run
15:00 pyrimidine (1.7.2)
15:01 carandraug if the only way to send communications from a project is a mailing list, if that project clearly mentions them as the only way to get information, then people who don't follow them don't care. Having more means to send announcements won't fix the issue is that the users are not willing to follow them.
15:01 carandraug ok
15:03 pyrimidine the problem is that mail list traffic overall has dropped (not just bioperl, but everywhere).  Bio*, seqanswers, and other resources have pulled all but the most die-hard users away
15:03 pyrimidine I see more posts on Bio* than on the list
15:04 pyrimidine so, it's not that people don't care
15:05 pyrimidine it's that they don't seem to look in the one place we communicate these things
15:06 pyrimidine anyway, I primarily just want to make it very clear, to as wide an audience as possible, what the plan is.
15:07 carandraug do users expect those places to discuss development of software developed elsewhere?
15:08 carandraug ok, fine
15:08 carandraug I have pushed the TCoffee and Clustalw distributions
15:09 pyrimidine carandraug: yes, in many cases these announcements get a little more air (tweets and other things like a Bio* announcement pointing to the original post)
15:10 pyrimidine Okay, will work on run 1.7.2
15:11 pyrimidine carandraug: have you pushed the commits removing TCoffee and Clustal to bioperl-run?
15:11 carandraug also, didn't you said the other you had made me an admin on github?
15:12 pyrimidine I think on those repos, but I'll simply add you to the admin team
15:13 carandraug I pushed the removal of the example from -live but not from -run
15:14 pyrimidine I've added you as a maintainer to the core developer team
15:15 carandraug I have a commit ready but seems like there's been changes since. I noticed you added more stuff to the Changes file but did you notice the top '{{$NEXT}}' section?
15:15 pyrimidine heh. was in a hurry there.
15:16 pyrimidine did you have changes? I can revert that commit, then add them back
15:17 pyrimidine I should say, did you have changes to 'Changes
15:17 pyrimidine ?
15:17 carandraug yes, but I have no problem merging that back in. It's just that you ignore my entry and repeated it so I thought maybe on -live and -run I was supposed to use some special format or tool
15:19 pyrimidine nope, just missed reading it b/c it was under {{$NEXT}} instead of a 1.7.x header.  However, there is no dzil.ini in bioperl-run or bioperl-live yet
15:21 pyrimidine I think one could be added to bioperl-run (though the build script will complicate it a bit)
15:21 pyrimidine same with live
15:24 carandraug I just used '{{$NEXT}}' as placeholder for whatever would be the next version. I dislike putting a version number because it is confusing for those building from source
15:24 pyrimidine Works for me
15:25 carandraug I don't want to set -live and -run with dzil. The bioperl pluginbundle requires some changes to the modules unless we disable most of the plugins which makes it kind of pointless
15:26 pyrimidine agreed, and it's not too difficult to make a release currently
15:27 pyrimidine though I will say the underlying code is a little brittle, used to rely on some Module::Build internals (not anymore)
15:28 pyrimidine just when we thought we were in the clear: https://travis-ci.org/bioperl/bioperl-run/builds/275514548?utm_source=email&utm_medium=notification
15:31 pyrimidine carandraug: looks like ProtDist and ProtPars tests in bioperl-run rely on Bio::Tools::Run::Align::ClustalW.  I'll check these but they shouldn't have a direct dependency (hope not)
15:32 carandraug pyrimidine: makes sense to me if bioperl-run is dependent on the new distributions. It will have to happen soon when I remove WrapperBase and friends
15:39 carandraug pyrimidine: I pushed another change that removed clustalw and tcoffee from the multiple files mentioning dependencies. I left it on the travis configuration file because I'm guessing you will soon have to configure it to install the new distro from cpan too
15:40 pyrimidine carandraug: looks like the phylip *tests* rely on the clustalw wrapper, but not the modules themselves.  I can have it skip those if ClustalW isn't installed to prevent a circular dependency.
15:41 pyrimidine but we should probably move the Phylip modules out at some point as well
15:42 pyrimidine not a huge issue but we can make an issue on github to track that
15:43 carandraug I'm looking at the bioperl teams at the moment. What's the meaning of that structure?
15:45 pyrimidine it's a bit of a legacy structure from the original github organization
15:45 carandraug I thought about cleaning up the permissions, since it's something I done recently at work
15:46 pyrimidine works for me
15:48 carandraug pyrimidine: from the list of repos we have, I would guess we should just have a team for all members and maybe a team for the website. Note that admins are no longer based on teams
15:48 pyrimidine Yes, this is the legacy part
15:50 pyrimidine we could easily get rid of the 'admin' groups and just assign maint to those in the team
15:52 carandraug pyrimidine: I still don't have permissinos.  You need to go to https://github.com/orgs/bioperl/people , find my username, and then select "Change Role" and which point you can add me as "Owner"
15:52 carandraug there is no longer admin groups or teams. That is now a "role"
15:53 pyrimidine carandraug: changed
15:54 pyrimidine Yes, the 'old-school owners' should have "Owner" roles, then that group can go
15:54 pyrimidine as well as Core Developers (the two overlap)
15:55 carandraug are they all still active?
15:57 pyrimidine not completely.  Lincoln, Jason, Hilmar, you, and myself should be 'Owners'
15:58 pyrimidine Jay, Rob, and Chris D. are members
15:58 pyrimidine as are the rest
16:03 carandraug pyrimidine: I'm still not owner. Do I appear as such on https://github.com/orgs/bioperl/people or do I appear as "Member" ?
16:03 pyrimidine that
16:03 pyrimidine that's odd.  Just did it again, you appear as owner now
16:08 pyrimidine The 'Core Developer' team is a little honorary since it originally defined the group that were like the original core (Lincoln, Jason, Chris D, Hilmar, myself, with a few additional contribs who helped a lot)
16:15 pyrimidine have to go for a while, and have meeting this afternoon.  I'll be back on around 11am tomorrow
16:17 carandraug pyrimidine: The core developer and old school owners are secret teams
16:17 carandraug not only those teams have no special purpose or powers, they are also secret so there's also no kudos gained from it
16:18 pyrimidine funny, the core one used to be public.
16:18 pyrimidine may as well get rid of them then
16:18 carandraug so I will have a single team for all members, and put Lincoln, Jason, Hilmar, you, and myself as owners
16:18 pyrimidine yep
16:19 carandraug then I will go through the repos and make sure there's nothing special on them. Any repo collaborator I will put as team member
16:20 pyrimidine bp-utils was an odd one, it was a separate project with scripts that is using the core code. members for that currently only have access to that repo
16:20 pyrimidine I don't see any reason to keep it separate though
16:25 pyrimidine did you manage to merge in the users from bp-utils into the main team?
16:27 carandraug pyrimidine: yes, already done
16:28 pyrimidine cool, I just saw the emails going out.  They're the only group I can think of that may not have members on the team
16:28 carandraug pyrimidine: well, I sent them an invitation to join the team ,they have to accept it
16:29 pyrimidine that's fine.  I don't think they should have an issue with that
16:29 carandraug actually, maybe you should send a ping asking to if they want to keep on the project?
16:29 pyrimidine ok
16:30 carandraug on a project I was maintaining before, every year I would send an email to everyone who hadn't made a commit in at least one year. If they don't reply within 1 month, they get removed. They only needed to reply they are still interested, no need to actually make any new contribution
16:31 carandraug I have seen similar things on other projects. For example, debian perl team does the same but only ping anyone who hasn't contributed in the last two years
16:32 carandraug others simply announce on the mailing list
16:32 pyrimidine It's largely been a non-issue for us tbh
16:33 pyrimidine not a bad policy though
16:34 pyrimidine so there are seven contributors to p5-bpwrapper (that is the former bp-util team)
16:36 carandraug I have just added Lincoln as owner
16:37 pyrimidine grr, forgot they removed private messaging.  hmm
16:37 pyrimidine I need to head out
16:38 carandraug pyrimidine: ok, see you tomorrow
16:38 pyrimidine I'm going to leave an issue so they are messaged in that repo, and see what they want to do
16:38 pyrimidine bye!
16:38 carandraug pyrimidine: just email the mailing list :p
16:42 pyrimidine I can tag their Github user name in the Issue, it's actually more direct
16:50 carandraug are we using all of this https://github.com/organizations/bioperl/settings/oauth_application_policy ?
17:07 scottcain_ joined #bioperl
17:15 scottcain_ joined #bioperl
17:18 carandraug joined #bioperl
17:26 pyrimidine joined #bioperl
17:32 pyrimidine joined #bioperl
17:36 pyrimidine joined #bioperl
17:41 pyrimidine joined #bioperl
17:56 pyrimidine joined #bioperl
17:59 pyrimidi_ joined #bioperl
18:09 pyrimidine joined #bioperl
18:13 pyrimidine carandraug: we could remove readthedocs
18:14 pyrimidine the others (coveralls, quay, travis) are used.
18:15 pyrimidine I'll check zenodo but we sometimes use this for a DOI snapshot
18:59 pyrimidine joined #bioperl
19:16 pyrimidine joined #bioperl
19:25 pyrimidine joined #bioperl
19:27 pyrimidine joined #bioperl
19:32 pyrimidine joined #bioperl
19:35 pyrimidine joined #bioperl
19:39 pyrimidine joined #bioperl
19:47 pyrimidine joined #bioperl
19:50 pyrimidi_ joined #bioperl
19:57 pyrimidine joined #bioperl
20:02 pyrimidine joined #bioperl
20:05 pyrimidine joined #bioperl
20:10 pyrimidine joined #bioperl
20:17 pyrimidine joined #bioperl
20:21 pyrimidi_ joined #bioperl
21:19 pyrimidine joined #bioperl
21:51 pyrimidine wow one would think the network here would be great, but I keep disconnecting :(
22:06 pyrimidine joined #bioperl
22:08 pyrimidi_ joined #bioperl
22:16 pyrimidine joined #bioperl
22:19 pyrimidine joined #bioperl
22:29 pyrimidine joined #bioperl
22:33 pyrimidine joined #bioperl
22:36 pyrimidine joined #bioperl
22:39 pyrimidi_ joined #bioperl
22:45 pyrimidine joined #bioperl
22:48 pyrimidi_ joined #bioperl
22:58 pyrimidine joined #bioperl
23:01 pyrimidi_ joined #bioperl
23:05 pyrimidine joined #bioperl
23:08 pyrimidine joined #bioperl
23:11 pyrimidine joined #bioperl
23:15 pyrimidine joined #bioperl
23:32 pyrimidine joined #bioperl
23:35 pyrimidine joined #bioperl
23:38 pyrimidine joined #bioperl
23:42 pyrimidi_ joined #bioperl
23:51 pyrimidine joined #bioperl
23:58 pyrimidine joined #bioperl

| Channels | #bioperl index | Today | | Search | Google Search | Plain-Text | summary