[06:20] <amurray> am just watching https://ftp.osuosl.org/pub/fosdem/2019/UD2.218A/usb_borne_attacks.mp4 from FOSDEM over the weekend - would be good to sync (desktop and security teams) on what the plans on in regards to this for 20.04 etc
[06:21] <amurray> s/on/are/
[08:47] <willcooke> morning
[08:59] <jibel> good morning willcooke
[09:02] <willcooke> hi jibel, good FOSDEM?
[09:03] <willcooke> jibel, love the burger receipt :)
[09:03] <jibel> willcooke, Yes, it was excellent, very good sessions
[09:03] <jibel> :)
[09:03] <jibel> the sessions were better than the burgers actually
[09:04] <willcooke> ha
[09:05] <Laney> yo
[09:05] <willcooke> hi Laney
[09:06] <Laney> hey willcooke
[09:06] <Laney> what's up
[09:08] <marcustomlinson> morning all, happy monday \o/
[09:08] <seb128> it's monday, good morning desktopers!
[09:10] <andyrock> morning all!
[09:11] <willcooke> hi seb128 andyrock
[09:14] <Laney> hey marcustomlinson seb128 andyrock
[09:14] <Laney> how's it going?
[09:14] <marcustomlinson> very good thanks (it snowed this weekend :D), yourself?
[09:15] <oSoMoN> good morning desktoppers
[09:15] <willcooke> hi oSoMoN marcustomlinson
[09:16] <Trevinho> hey guys
[09:16] <oSoMoN> hey willcooke, Trevinho
[09:16] <willcooke> o/ Trevinho
[09:16] <Trevinho> hi willcooke and oSoMoN!
[09:16] <seb128> hey oSoMoN, trevinho, willcooke, Laney, andyrock, marcustomlinson
[09:17] <seb128> oSoMoN, did you make it back without issue?
[09:17] <Trevinho> we can make a greetings chain :)
[09:17] <Laney> yeah I'm alright, quite tired this morning, basically got back from fosdem went to bed and now i'm here
[09:17] <Laney> moin Trevinho
[09:18] <seb128> I had croissant au chocolat this morning, it's the best!
[09:19] <amurray> willcooke: hey, I asked earlier but you weren't online yet - I noticed some of the GNOME folks presented about USBGuard integration with GNOME at FOSDEM - would be good to sync re plans for this and see if / how the security team can help
[09:19] <andyrock> > croissant au nutella
[09:19] <andyrock> FTFY
[09:20] <willcooke> amurray, hi! Yeah, I'll add it to the list in Malta.  I don't know much about the workings of it though.  But I'm sure we can work it out
[09:20] <amurray> willcooke: awesome - thanks
[09:20] <seb128> andyrock, that's not quite the same
[09:21] <seb128> willcooke, amurray, https://wiki.gnome.org/Internships/2018/Projects/USB-Protection
[09:21] <oSoMoN> seb128, yeah, the journey was uneventful and I got home at 9:30pm
[09:21] <oSoMoN> seb128, how about you?
[09:22] <seb128> oSoMoN, it was fine, train is great. sncf had issues and the TGV I need to take to finish the trip was not there so I had to wait another half an hour for my connection, I arrived around 7:30pm which was ok
[09:23] <amurray> seb128 willcooke: I assume there will be some bits to MIR etc as part of this, plus would also be good to just review the policy and what we would set as default etc plus to know what upstreams timeframe is and whether this is a candidate for 20.04 etc
[09:23] <oSoMoN> is croissant au chocolat a thing? in my times we called it pain au chocolat, or is that a hybrid?
[09:23] <seb128> oSoMoN, come on, don't do your Didier!
[09:24] <oSoMoN> ah, I felt I had to, since Didier didn't react :)
[09:25] <oSoMoN> he's offline, that explains it
[09:25] <willcooke> amurray, ack, we can take a look.  Looks like it's
[09:25] <willcooke> still early days, but I will review that FOSDEM session
[09:25] <didrocks> good morning from the train!
[09:26] <didrocks> livecd-rootfs migrated, nice
[09:26] <willcooke> hey didrocks
[09:26] <seb128> speaking of didrocks
[09:27] <seb128> hey didrocks
[09:27] <willcooke> didrocks, seb128 said that you like Pain au Chocolat
[09:27] <willcooke> XD
[09:28]  * seb128 slaps willcooke with an old trout
[09:29] <marcustomlinson> https://media.giphy.com/media/wv216pxQtF3Ww/giphy.gif
[09:33] <seb128> willcooke, amurray, https://ryuzakikk.github.io/ has the details and apparently the guy just opened MRs on GNOME component so good cycle is probably good timing to look at that, I might already open MIRs this cycle to have an head start
[09:33] <willcooke> seb128, ace!
[09:33] <amurray> seb128: cheers - thanks for that
[09:33] <seb128> np!
[09:35] <willcooke> seb128, you said you had problems with conn.check on the train?  Did you find bugs?
[09:40] <seb128> willcooke, turned out that a systemctl restart NetworkManager was not enough to get the new 1.15 version fully in place, it started working after a reboot :/
[09:40] <seb128> upstream pointed a fix though to better deal with some weird cases, i need to check if we need that one backported to other series
[09:49] <willcooke> kk
[09:49] <willcooke> thx
[09:56] <didrocks> flaky train network
[09:57]  * Laney glues didrocks back together
[11:27] <willcooke> clobrano, hey!  I'm trying the Communitheme Dark theme for the first time in a long time, it's great!!! thanks to you guys
[11:28] <willcooke> the snap that is, hence the Communitheme
[11:43] <clobrano> willcooke: :D I'm glad you like it!
[11:44] <clobrano> It's still less mature than the light theme, so I'd be happy if you can find the time to report any problem
[11:58] <willcooke> clobrano, will do :)
[13:37] <didrocks> and back home!
[13:43] <seb128> wb didrocks!
[13:47] <didrocks> thx ;)
[14:01] <jibel> didrocks, hey, how was the trip?
[14:02] <jibel> didrocks, I had a look at pycallgraph2 and it looks okay for what we want to do. I fixed the examples and it can generate some fairly large call charts.
[14:02] <didrocks> jibel: uneventful, only 10 minutes late, which is ok ;)
[14:03] <didrocks> jibel: oh nice, did you get it to run on subiquity without any issue?
[14:03] <didrocks> remember it was working on some simple files
[14:03] <didrocks> but not subiquity for $strange_reasons
[14:03] <jibel> I got 8€ for my cancelled train, you cannot even buy a sandwich while you wait for next train :/
[14:03] <didrocks> (you installed the pip version as well, and had to change the import?)
[14:04] <didrocks> jibel: a cookie + a coke at most ;)
[14:04] <jibel> didrocks, no I run from trunk
[14:04] <didrocks> "thx SNCF"
[14:04] <jibel> the pip version is buggy
[14:04] <didrocks> ohhhh
[14:04] <didrocks> that's why then
[14:04] <didrocks> we should have started with that, indeed
[14:04]  * didrocks trusted too much "released versions"
[14:04] <jibel> yeah it didn't finish the renaming from pycallgraph to pycallgraph2
[14:05] <didrocks> excellent news then!
[14:05] <didrocks> we should try profiling multiple install scenarios
[14:05] <didrocks> there is no text file produced, correct? So that we can diff the scenarios
[14:05] <jibel> I'm looking at gephi to view the charts. It seems better than graphviz for large ones
[14:06] <jibel> it is not packaged apparently
[14:06] <jibel> didrocks, no only a png
[14:06] <didrocks> trusting you on gephi vs graphviz ;)
[14:07] <didrocks> I don't remember what we used for large bootchart images we had in 2010
[14:07] <didrocks> just that eog wasn't able to handle it
[14:07] <jibel> didrocks, I looked at pytrace too which provides the same level of info but in text format. It's a bit harder to read but we could compare runs
[14:07] <didrocks> yeah
[14:07] <didrocks> I guess for diffing, that's OK
[14:07] <jibel> large svg usually kill eog
[14:07] <didrocks> to see the general path, it's way harder as we discussed
[14:08] <jibel> the best tool to watch those is ..... chrome :)
[14:09] <didrocks> gephi is a good snap candidate, as being java
[14:09] <didrocks> yeah ;)
[14:09] <jibel> ah gephi is in java, good candidate for a snap
[14:09] <jibel> exactly :D
[14:09] <didrocks> heh
[14:09]  * jibel registers the name
[14:09] <didrocks> you will start to be the java snap expert!
[14:10] <jibel> well, I made a template, put a zip or a jar in the same directory than snapcraft.yaml and it'll make a snap out of it
[14:11] <jibel> Success: gephi registered
[14:11] <didrocks> no deps needed? only the jre?
[14:13] <jibel> didrocks, all the deps are in the upstream jar
[14:13] <jibel> just add the jre
[14:14] <didrocks> sounds easy enough :)
[14:14] <jibel> actually it take a bit more time because you need an icon, a desktop file, a description, do a nice page on the store ...
[14:15] <didrocks> all this non-technical work… :)
[14:15] <jibel> hm, pycallgraph's CI is broken
[14:15] <didrocks> btw, I saw some sun this morning in Brussels and blue sky (for a short while) before taking the train
[14:15] <didrocks> no kidding
[14:15] <didrocks> ah?
[14:18] <cpaelzer> didrocks: thanks for your update on that byobu bug
[14:18] <jibel> didrocks, pluggy.PluginValidationError: Plugin 'pytest_cov' could not be loaded: (pytest 3.3.0 (/home/travis/virtualenv/python2.7.14/lib/python2.7/site-packages), Requirement.parse('pytest>=3.6'))!
[14:20] <cpaelzer> didrocks: did I understand you right that Desktops "try to detect which program is running", and for that will scan through the .desktop files - and this fails in the bug due to LXQt not honoring StartupWMClass, so detecting it as byobu is a false positve for that
[14:21] <didrocks> cpaelzer: exactly! This is my guess on why all gnome-terminal is matched by LXQt shell against byobu (just based on the Exec=gnome-terminal…)
[14:21] <cpaelzer> ok thanks
[14:22] <cpaelzer> as I'm blank on that area I was trying to wrap my head around desktop'y things when I see them to learn a bit at least :-)
[14:22] <didrocks> jibel: hum, you are using python2? unsure how it loads the python3 subiquity-based then
[14:23] <didrocks> cpaelzer: yeah, can be quite complex :) I think some googling/talk with the upstream LXQt devs to know which LXQt components is doing the matching
[14:23] <didrocks> in Unity, this was bamf, in GNOME, it's GNOME Shell itself
[14:24] <didrocks> (also, I wonder why someone is running gnome-terminal on LXQt, I'm sure there is a more integrated terminal in that DE?)
[14:26] <jibel> didrocks, the migration to python3 is not completely finished apparently
[14:27] <didrocks> jibel: how is subiquity running then with pycallgraph?
[14:27] <didrocks> as it's python3 only?
[14:31] <jibel> the tool itself works with py3, not the tests :/
[14:31] <jibel> or the examples
[14:31] <didrocks> ahhhh
[14:32] <didrocks> got you now
[14:49] <clobrano> willcooke: are you on edge channel or stable?
[14:54] <willcooke> clobrano, stable
[15:01] <clobrano> willcooke: okay, there should be some improvement on edge I will release tomorrow
[15:03] <willcooke> woot
[16:39] <k_alam> Hi desktopers, good afternoon
[16:49] <k_alam> seb128, jbicha: u there ?
[16:57] <jbicha> good evening
[16:58] <willcooke> clobrano, Do you want comments on github or here? re: dark theme
[17:01] <clobrano> willcooke: if it's a bug, on github would be good, otherwise here is fine
[17:01] <k_alam> jbicha: I opened a merge for rhythmbox-zeigeist...https://bugs.launchpad.net/ubuntu/+source/rhythmbox/+bug/1813813
[17:03] <willcooke> clobrano, I'm not really sure.  I'll comment here, see what you think.  In Chrom[e|ium], it's hard to differentiate between active and inactive tabs.  It might be a Cr. thing, it might be a theme thing, not sure.
[17:04] <clobrano> willcooke: yes, we had that bug opened. We didn't find a way to style it, but I can have a look again
[17:04] <jbicha> yes, the rb thing is a bit complicated
[17:04] <willcooke> clobrano, ah you've already seen it, ace!
[17:05] <willcooke> clobrano, I'm not surprised it's hard to fix
[17:05] <willcooke> if even possible
[17:05] <k_alam> jbicha: And please review this https://bugs.launchpad.net/ubuntu/+source/unity/+bug/1814506
[17:07] <jbicha> go ahead and subscribe ubuntu-sponsors to that bug
[17:08] <jbicha> I don't really have the time/concentration now to review Unity stuff, sorry
[17:08] <clobrano> Uhm, it seems that matrix doesn't sync my messages fast enough
[17:08] <clobrano> My following message might be duplicated sooner or later :D
[17:08] <clobrano> I should have a look at chromium project itself. Being a Firefox fan kept me away from this issue :D
[17:09] <willcooke> :)
[17:13] <oSoMoN> clobrano, if that is a chromium bug, please keep me in the loop, and let's ensure it's reported upstream
[17:32] <clobrano> oSoMoN: I don't think it's a bug actually. It's just that chromium uses it's own css classes
[17:34] <Laney> what is zfs desktop?
[17:34] <Laney> (from jibel/didrocks reports)
[17:41] <willcooke> Laney, we were looking for somewhere to try zfs as root part.  So as part of the experimental image jibel and didrocks are making for the new installer work, we figured we'd try out zfs as the root fs too
[17:51] <Laney> ah
[17:57] <sarnold> oh sweeeeet
[17:57] <sarnold> zfs on root is enough work that I've never tried
[17:58] <sarnold> having distro support for that would be a huge differentiator
[18:04] <willcooke> sarnold, it's something we wanted to tag on as a test, so don't get your hopes up too much :)
[18:04] <willcooke> but ya know, if you wanna help....
[18:06] <sarnold> willcooke: cooool :) pencil my name in for "happy to review patches" :)
[18:06] <willcooke> :)
[22:14] <jbicha> wow, the rhythmbox zeitgiest plugin bug was reported in April 2014 😭
[22:15] <jbicha> bug 1313914