22:01:39 <FranciscoD> #startmeeting NeuroFedora team meeting - 2019-03-26
22:01:39 <zodbot> Meeting started Tue Mar 26 22:01:39 2019 UTC.
22:01:39 <zodbot> This meeting is logged and archived in a public location.
22:01:39 <zodbot> The chair is FranciscoD. Information about MeetBot at http://wiki.debian.org/MeetBot.
22:01:39 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
22:01:39 <zodbot> The meeting name has been set to 'neurofedora_team_meeting_-_2019-03-26'
22:01:45 <FranciscoD> #chair zbyszek
22:01:45 <zodbot> Current chairs: FranciscoD zbyszek
22:01:47 <FranciscoD> yep :)
22:01:49 <zbyszek> I guess yes.
22:01:57 <tg-fedneuro> <L​oKoMurdoK> Here
22:01:59 <FranciscoD> #meetingname neurofedora
22:01:59 <zodbot> The meeting name has been set to 'neurofedora'
22:02:12 <FranciscoD> #topic Roll call
22:02:16 <FranciscoD> #info ankursinha
22:02:21 <FranciscoD> #info lbazan
22:02:25 <FranciscoD> #info zbyszek
22:02:27 <tg-fedneuro> <L​oKoMurdoK> Thanks
22:02:37 <FranciscoD> #info blackfile is driving
22:02:56 <FranciscoD> #info pac23 should be sleeping
22:03:10 <FranciscoD> That's all I guess
22:03:19 <FranciscoD> #topic Current status
22:03:31 <FranciscoD> #info We've gotten a lot of computational neuroscience software done
22:03:45 <FranciscoD> #info https://docs.fedoraproject.org/en-US/neurofedora/compneuro-tools/
22:03:53 <FranciscoD> #info our docs are at neuro.fedoraproject.org
22:04:03 <FranciscoD> #info our blog is at neurofedora.github.io
22:04:23 <tg-fedneuro> <L​oKoMurdoK> Good!
22:04:30 <FranciscoD> for comp-neuro, only NEURON and PyNN are the two major bits that are pending.
22:04:56 <tg-fedneuro> <L​oKoMurdoK> Neuron work in progress?
22:05:04 <tg-fedneuro> <L​oKoMurdoK> Pynn?
22:05:15 <FranciscoD> #info NEURON has not yet been udpated to use the new cvode version (branch exists, but has not yet been merged): https://github.com/neuronsimulator/nrn/issues/113
22:05:55 <FranciscoD> So, while I do have a neuron package, I've not yet provided the python bindings or MPI versions. I figured I'll wait until the code changes are complete
22:06:42 <tg-fedneuro> <L​oKoMurdoK> Ok
22:06:56 <FranciscoD> #info PyNN does not currently support Brian2: WIP, but no updates yet: https://github.com/NeuralEnsemble/PyNN/pull/617#issuecomment-456741238
22:07:45 <FranciscoD> We do provide brian1 in our COPR so we could provide PyNN. It also supports NEURON but that part will have to wait until the NEURON packaging is complete
22:08:32 <FranciscoD> #info TVB library will only be py3 complete by the end of 2019, so we'll have to wait until then
22:08:58 <FranciscoD> #undo
22:08:58 <zodbot> Removing item from minutes: INFO by FranciscoD at 22:08:32 : TVB library will only be py3 complete by the end of 2019, so we'll have to wait until then
22:09:15 <FranciscoD> #info TVB library will only be py3 complete by the end of 2019, so we'll have to wait until then: https://github.com/LTS5/cfflib/pull/8#issuecomment-473934882
22:09:46 <fm-neuro> pagure.issue.comment.added -- ankursinha commented on ticket neuro-sig/NeuroFedora#68: "Package tvb-library" https://pagure.io/neuro-sig/NeuroFedora/issue/68#comment-562891
22:10:10 <FranciscoD> Are there any other tools for comp-neuro that we should prioritise from our list: https://pagure.io/neuro-sig/NeuroFedora/issues?tags=F%3A+Computational+neuroscience ?
22:10:50 <tg-fedneuro> <L​oKoMurdoK> Lfpy
22:11:08 <FranciscoD> you're working on that, no? Please assign the review to me when its ready :)
22:11:35 <FranciscoD> #info STEPS needs lots of unbundling and updating: upstream has said they'll do it in the new release: https://github.com/CNS-OIST/STEPS/issues/15
22:11:43 <tg-fedneuro> <L​oKoMurdoK> Ok
22:12:12 <zbyszek> Doesn't lfpy depend on neuron?
22:12:30 <FranciscoD> zbyszek: I haven't checked tbh. @LoKoMurdoK: have you checked yet?
22:12:52 <zbyszek> https://github.com/LFPy/LFPy/blob/master/requirements.txt says so...
22:13:05 <FranciscoD> meh, then it's blocked too
22:13:13 <FranciscoD> quite a few are blocked by neuron, unfortunately
22:13:33 <FranciscoD> ah, the ticket already has a dep on nrnpython
22:14:03 <tg-fedneuro> <L​oKoMurdoK> :(
22:14:20 <fm-neuro> pagure.issue.edit -- ankursinha edited the priority fields of ticket neuro-sig/NeuroFedora#182 https://pagure.io/neuro-sig/NeuroFedora/issue/182
22:14:21 <fm-neuro> pagure.issue.comment.added -- ankursinha commented on ticket neuro-sig/NeuroFedora#182: "Package nrnpython: python bindings for neuron" https://pagure.io/neuro-sig/NeuroFedora/issue/182#comment-562895
22:14:27 <tg-fedneuro> <L​oKoMurdoK> No problem i will continue with other package
22:14:39 <FranciscoD> I'll look at the neuron master branch again and see if I can package a snapshot and proceed
22:15:02 <FranciscoD> #action FranciscoD look at the neuron master branch again and see if it is possible to proceed with a snapshot
22:15:19 <FranciscoD> How are we doing on the other fronts?
22:16:12 <FranciscoD> I was thinking we could first do up the comp-neuro bits first, and then focus on neuro-imaging etc
22:16:34 <FranciscoD> Since we have some momentum going with comp-neuro, and we have CNS coming up where we'll have a poster
22:16:49 <FranciscoD> A comp-neuro lab by then would be *very nice*
22:17:13 <FranciscoD> #info CNS*2019 is in mid july: https://www.cnsorg.org/cns-2019
22:17:25 <tg-fedneuro> <L​oKoMurdoK> Flock in august
22:17:25 <FranciscoD> that gives us 3 months to get a lab image
22:17:34 <FranciscoD> #info Flock in August
22:17:57 <FranciscoD> We did have someone volunteer to look into the lab image bit, but that's not been followed up
22:17:58 <zbyszek> I guess that's doable.
22:18:20 <FranciscoD> zbyszek: do you know the lab process? Can we release images mid release for example?
22:18:40 <FranciscoD> or will it only become "official" at the F31 release
22:18:41 <zbyszek> No, I don't know.
22:19:23 <FranciscoD> #info F30 releases on 2019-05-07 at the moment: https://fedoraproject.org/wiki/Releases/30/Schedule
22:19:46 <FranciscoD> Yeh, we'll need to look into the lab process then. I'll take that up next, and document it all in a blog post maybe
22:20:05 <FranciscoD> #action FranciscoD investigate Fedora lab process, document in a neurofedora blog post.
22:20:17 <FranciscoD> That may act as a starting document for a volunteer too
22:20:40 <FranciscoD> So, here's the proposal then:
22:20:46 <tg-fedneuro> <L​oKoMurdoK> Question: fedora neuro iso or dnf group install?
22:21:02 <zbyszek> There's also https://labs.fedoraproject.org/en/scientific/
22:21:17 <tg-fedneuro> <L​oKoMurdoK> Ok
22:21:35 <FranciscoD> scientific is pretty much in maintenance mode. Amit (the maintainer) had said so at some point
22:21:44 <zbyszek> It might be easier to just add some missing packages to that
22:21:51 <FranciscoD> Scientific is quite a good collection, but it's quite general.
22:22:01 <FranciscoD> zbyszek: it doesn't contain any neuro tools at the moment
22:22:47 <FranciscoD> I was thinking of downloadable ISOs for: a) comp-neuro b) neuro-imaging (when we make progress on that front)
22:22:56 <FranciscoD> so that folks can download + install -> get started
22:23:13 <tg-fedneuro> <L​oKoMurdoK> +1
22:23:46 <FranciscoD> to permit folks to also use groupinstall, we could set up a comps group first (and use that in the labs kickstart)
22:24:26 <zbyszek> Yeah, a comps group is much less work and generally useful at the same time.
22:24:55 <FranciscoD> OK, let's make that first priority then
22:25:05 <FranciscoD> once we have that, we can look at the usefulness of the ISO
22:25:09 <FranciscoD> does that sound OK?
22:25:23 <zbyszek> +1
22:25:43 <fm-neuro> pagure.issue.assigned.added -- ankursinha assigned ticket neuro-sig/NeuroFedora#198 to ankursinha https://pagure.io/neuro-sig/NeuroFedora/issue/198
22:26:05 <fm-neuro> pagure.issue.edit -- ankursinha edited the milestone and priority fields of ticket neuro-sig/NeuroFedora#198 https://pagure.io/neuro-sig/NeuroFedora/issue/198
22:26:20 <FranciscoD> #action FranciscoD set up comp-neuro comps group
22:26:31 <tg-fedneuro> <L​oKoMurdoK> +1
22:26:41 <FranciscoD> #info Look into labs image after comps group is set up
22:26:54 <FranciscoD> So, in general, are we agreed that we focus on comp-neuro at least till CNS?
22:27:08 <FranciscoD> Hopefully by then mhough will be back to lead the neuro-imaging effort
22:27:44 <zbyszek> https://fedoraproject.org/wiki/Spins_SIG seems to be the docs for spins
22:28:28 <FranciscoD> I *think* I'm on the ML already. Haven't had the time to investigate the whole process in detail, yet
22:29:15 <zbyszek> "Submit spin for review at least 3 weeks before Feature Freeze..."
22:29:24 <zbyszek> So that answer the question we had earlier.
22:29:47 <FranciscoD> ah, well, so it won't happen before F30 release then
22:30:03 <FranciscoD> I guess if we're ready enough, we could have it up immediately after F30 release
22:30:11 <zbyszek> F31
22:30:19 <tg-fedneuro> <L​oKoMurdoK> F31
22:30:23 <tg-fedneuro> <L​oKoMurdoK> +1
22:30:58 <FranciscoD> I mean get it all ready and submitted for review or whatever after F30 releases on 2019-05-07
22:31:19 <FranciscoD> It'll only be officially released with F31, but we can point to the nightly images in docs etc until then
22:32:06 <FranciscoD> there's 5 weeks between F30 release and CNS to get nightly builds of the image up
22:32:33 <tg-fedneuro> <L​oKoMurdoK> we need to recruit more packagers
22:32:54 <FranciscoD> Let's make that a new topic
22:33:01 <FranciscoD> #topic Expanding the team
22:33:31 <FranciscoD> Ideas on getting more people to join the team?
22:33:49 <FranciscoD> I guess after our poster at CNS, and the talk at flock, we'll have a better chance, but until then?
22:33:53 <tg-fedneuro> <L​oKoMurdoK> Packaher ml email
22:34:04 <tg-fedneuro> <L​oKoMurdoK> Flock
22:34:24 <FranciscoD> @LoKoMurdoK: i always send "review swap" etc mails to the devel ML hoping to get more people involved
22:34:34 <tg-fedneuro> <L​oKoMurdoK> Ok
22:35:07 <FranciscoD> maybe we should do that then: first send a review-swap mail, if no new person takes it up in 2--3 days, someone from the team can step in and review
22:35:44 <tg-fedneuro> <L​oKoMurdoK> Ok
22:35:47 <FranciscoD> I.e., reduce "internal reviewing" to get more community participation?
22:36:32 <FranciscoD> zbyszek: any ideas on how we can get more academics involved? Even undergrad/post-grad students will do
22:36:52 <zbyszek> It's a hard problem.
22:36:54 <FranciscoD> researchers unfortunately are hard---they usually have the "I'll do the minimum required for my work" attitude
22:38:13 <tg-fedneuro> <L​oKoMurdoK> I can check in latam channels
22:38:22 <FranciscoD> +1
22:38:39 <FranciscoD> zbyszek: do you think it's easier to get techies over to neuro than it is to get neuro people over to tech?
22:38:52 <FranciscoD> as in, should we focus on getting foss folks to neuro intead of trying to get neuro folks to foss?
22:39:14 <zbyszek> I think neuro folks are generally favorable towards foss
22:39:41 <FranciscoD> yes, but whoever I speak to is happy to use foss as long as they dont have to do the coding bits somehow
22:40:05 <FranciscoD> it's understandable given their workloads
22:40:33 <FranciscoD> pynn etc seem to be only one--two people dev teams, even when they are so widely used
22:40:53 <FranciscoD> Only the nest people have some community participation but they do open meetings every week etc
22:41:05 <zbyszek> I have seen various discussions in which various "important" people were generally pro-foss, as long as it didn't stand in the way of research, publishing, and other priorities.
22:41:31 <FranciscoD> +1
22:42:02 <FranciscoD> a citation for neurofedora (which is not directly science related) seems to hold less importance than a scientific citation too, no?
22:42:09 <tg-fedneuro> <L​oKoMurdoK> +1
22:42:12 <zbyszek> So I think it's good to make the whole processes smoother, e.g. by packaging stuff and making it easy to consume the whole foss ecosystem, and (I hope) in the longer term this will cause neuro people to use what we provide.
22:42:52 <FranciscoD> OK, I agree. It's hard to build a contributor base without first building a user base
22:43:07 <zbyszek> Going the other way, i.e. asking non-neuroscience people to work on neuroscience tools, seems strange to me. Most poeple wouldn't have any interest.
22:43:33 <FranciscoD> zbyszek: I intend to test that at flock when I speak about how foss should also be driving open science
22:43:43 <FranciscoD> the feedback from there will be quite interesting
22:43:52 <zbyszek> Yes.
22:44:06 <FranciscoD> Ok, so to summarise:
22:44:25 <FranciscoD> #info Current focus on building a user base, and then working towards converting users to contributors
22:44:41 <FranciscoD> #topic Current usability review and suggestions for improvements
22:44:55 <FranciscoD> How are we doing from a usability perspective at the moment?
22:45:03 <FranciscoD> Anything at all that we can improve?
22:45:20 <FranciscoD> Marketing/docs/software?
22:45:31 <FranciscoD> I forget one:
22:46:21 <FranciscoD> #info In general, when a new package is ready for review, we first e-mail the devel list for a review swap to get more community participation. If the review is not picked up by a "non-team member" in 3--4 days, another team member can step in and do the review.
22:46:47 <FranciscoD> #info https://docs.fedoraproject.org/en-US/neurofedora/overview/ -> current docs
22:47:08 <FranciscoD> #info Please remember to add software to the lists in docs, and document their usage if they are complex (MPI, for example)
22:47:24 <tg-fedneuro> <L​oKoMurdoK> Ok
22:47:54 <zbyszek> Also, it's helpful to do random reviews of packages from https://fedoraproject.org/PackageReviewStatus/NEW.html, people tend to return the favour even without asking.
22:48:02 <FranciscoD> +1
22:48:15 <tg-fedneuro> <L​oKoMurdoK> +1
22:48:45 <FranciscoD> #info Continue reviewing community tickets to build relationships within the packager group
22:50:54 <FranciscoD> I can't think of anything at the moment. Slowly improving the package set and gaining users seems to be the way to go
22:51:13 <FranciscoD> #topic Open floor
22:51:17 <FranciscoD> Anything else?
22:51:25 <FranciscoD> #action FranciscoD blog about meeting notes
22:51:37 <tg-fedneuro> <L​oKoMurdoK> Neuro badge
22:51:42 <tg-fedneuro> <L​oKoMurdoK> :)
22:51:54 <FranciscoD> Ah! For SIG members, or users?
22:52:10 <tg-fedneuro> <L​oKoMurdoK> Mmmm sig members i think
22:52:25 <tg-fedneuro> <L​oKoMurdoK> Or sig users xD
22:52:35 <tg-fedneuro> <L​oKoMurdoK> Sig members
22:52:35 <FranciscoD> It doesn't do much then---SIG members are getting a citation to begin with
22:52:46 <FranciscoD> it adds a bit of fun, yeh
22:53:27 <FranciscoD> @LoKoMurdoK: could you open a ticket on our tracker for this so we can think about it a bit?
22:53:34 <zbyszek> Somebody would have to design the badge, it's actually quite a bit of work...
22:53:39 <tg-fedneuro> <L​oKoMurdoK> Yes
22:53:54 <FranciscoD> zbyszek: I asked the design team to do our stickers: they'll hopefully be ready soon
22:54:02 <tg-fedneuro> <L​oKoMurdoK> Blackfile
22:54:04 <FranciscoD> https://pagure.io/design/issue/627
22:54:16 <tg-fedneuro> <L​oKoMurdoK> Can you design the badge
22:54:24 <tg-fedneuro> <L​oKoMurdoK> ?
22:54:27 <FranciscoD> #info NeuroFedora stickers: https://pagure.io/design/issue/627
22:54:57 <FranciscoD> I guess we can work on T-shirts next: have them ready and send them out to users and labs
22:54:58 <tg-fedneuro> <L​oKoMurdoK> @Blackfile
22:55:04 <FranciscoD> he's probably still driving :)
22:56:05 <FranciscoD> #action LoKoMurdoK open ticket about neurofedora badge
22:56:17 <tg-fedneuro> <L​oKoMurdoK> Stickers
22:56:36 <FranciscoD> they'll be done soon hopefully, and I'll take them to CNS and bring them to flock
22:56:37 <tg-fedneuro> <L​oKoMurdoK> +1
22:56:45 <FranciscoD> Ok, almost out of time.
22:56:55 <FranciscoD> Should we have another meeting in 4--6 weeks and see how we're doing?
22:56:58 <zbyszek> The logo from 627 could be used as a basis for a badge. Then a badge seems doable.
22:57:07 <FranciscoD> +1
22:57:13 <tg-fedneuro> <L​oKoMurdoK> +1
22:57:32 <FranciscoD> #info FYI: all NeuroFedora artwork is kept here: https://pagure.io/neuro-sig/NeuroFedora/blob/master/f/Artwork
22:57:41 <FranciscoD> when the stickers are ready, I'll put them there too
22:57:52 <FranciscoD> (the stickers are using our logo)
22:58:34 <tg-fedneuro> <L​oKoMurdoK> +1
22:59:51 * FranciscoD hadnt realised all the work that goes into making stickers
23:00:07 <FranciscoD> #action FranciscoD send out logs to ML
23:00:15 <FranciscoD> We're out of time. Let's close
23:00:20 <tg-fedneuro> <L​oKoMurdoK> +1
23:00:27 <tg-fedneuro> <L​oKoMurdoK> Go ahead!
23:00:34 <FranciscoD> Thanks everyone for coming. We'll find another time for the next meeting in a month or so;
23:00:37 <FranciscoD> #endmeeting