ChanServ changed the topic of #freedesktop to: https://www.freedesktop.org infrastructure and online services || for questions about freedesktop.org projects, please see each project's contact || for discussions about specifications, please use https://gitlab.freedesktop.org/xdg or xdg@lists.freedesktop.org
oldpcdude has quit [Remote host closed the connection]
oldpcuser has joined #freedesktop
todi has quit [Remote host closed the connection]
todi has joined #freedesktop
alpernebbi has quit [Quit: alpernebbi]
alpernebbi has joined #freedesktop
Leopold_ has joined #freedesktop
Leopold__ has quit [Ping timeout: 480 seconds]
ninja2181 has quit [Remote host closed the connection]
ninja2181 has joined #freedesktop
AbleBacon has quit [Read error: Connection reset by peer]
Lyude has quit [Quit: Bouncer restarting]
tzimmermann has joined #freedesktop
Lyude has joined #freedesktop
i509vcb has quit [Quit: Connection closed for inactivity]
alanc has quit [Remote host closed the connection]
alanc has joined #freedesktop
sima has joined #freedesktop
kxkamil2 has quit [Remote host closed the connection]
kxkamil2 has joined #freedesktop
pjakobsson has quit [Ping timeout: 480 seconds]
pjakobsson has joined #freedesktop
<pq>
I do not like that a user who does not agree to the CoC was given access. It seems like the review process has failed.
<bentiss>
pq: whot said he will fix this
<pq>
cool, thanks
<DavidHeidelberg[m]>
pq: do we care more about politics than the value of person contribution now? Just asking, because what I understood he wrote that he didn't read CoC, which isn't violation as I understood
aswar002 has quit [Remote host closed the connection]
aswar002 has joined #freedesktop
<pq>
DavidHeidelberg[m], I care about people feeling safe. That is political, yes.
<pq>
Violation no, I guess, not of CoC AFAIU, but it seems like the idea is that everyone agrees to the CoC.
<DavidHeidelberg[m]>
I don't see anywhere written he disagree, so why we do this?
<pq>
it very much looks like they intentionally deleted that line from the template
<pq>
you might as well ask why does a CoC exist at all, and I'll defer to the answer to that, which I don't feel qualified to explain here
<airlied>
DavidHeidelberg[m]: yes we do care more
<airlied>
and we should care more
<airlied>
like I'm not sure if not accepting the CoC matters or not, the site rules are what they are
<airlied>
if they turn asshat and get a CoC violation we would still remove them anways
<airlied>
but I think it's a bad faith move to begin with to make a point of not reading it :)
<DavidHeidelberg[m]>
I would still BEGIN with good faith in people. If you usually contribute to OSS project, you do it because you like to help others. (imho).
<emersion>
i think the CoC is very important, but i don't think they removed the CoC agreement line in bad faith necessarily
<whot>
pq: yeah, I missed that the CoC line was removed, I skimmed the template, saw all checkboxes ticked and acked it, didn't realize he intentionally removed that line
<emersion>
there may be other reasons, like seeing it as pointless paperwork and a "terms and conditions" kind of thing
<airlied>
emersion: seeing it as pointless is bad faith :)
<emersion>
which is wrong, but something someone unfamiliar with CoCs could think
<airlied>
like I'm sure they click through EULAs just fine
<emersion>
i don't know, i myself found CoCs pointless in the past, as in "be nice" is the only CoC i'd ever need
<emersion>
but i came around, and yes CoCs are important
<bentiss>
emersion: whether or not this was done in bad faith is not the point IMO. This tricked us, and this is something I am not very found of it.
<pq>
whot, aye, no worries. It's like there is reviewer checklist missing?
<emersion>
yeah, i can see that
<emersion>
and i completely agree
<bentiss>
we probably should add steps in the automated bot to ensure that those lines are still there
<bentiss>
(hint: it's plain python code, so anybody is welcome to contribute)
<DavidHeidelberg[m]>
Oh, I missed the part about removing CoC by hand. I takeback my "good faith" comment :D
<DavidHeidelberg[m]>
I don't want to support that kind of approach to the project :(
<whot>
yeah, it wasn't a "didn't check it, let's discuss", it was a "removed that line so everything looks like it's checked"
<bentiss>
also, I must confess that the initial idea behind that "let's validate users" step was to filter bots. But since it's in place we have seen many abuses: peolple who refuse to check the CoC line while they already agree on it while creating their account, and people trying to fake projects through user accounts because we disable group creation, and last, crypto kids
<bentiss>
so even if it's a pain to everybody, I think it has value protecting us from abuse
<bentiss>
(but it's a little bit of a pain that admins are responsible for granting permissions to users to protect from abuse)
<whot>
well, anyone who wants that job just needs to get Developer role in the freedesktop/freedesktop project
<bentiss>
exactly what I was about to say
<bentiss>
though they also need to activate notifications on this project :)
jarthur has joined #freedesktop
<DavidHeidelberg[m]>
testing new container, got 3 succ runs, 1 error "ERROR: Job failed (system failure): Error response from daemon: container bf4187d8e836135a76e55a7d7c7a89dab275380e21011c21cac2aae31b57b5bd does not exist in database: no such container (exec.go:78:0s)" https://gitlab.freedesktop.org/okias/mesa/-/jobs/42851998
<DavidHeidelberg[m]>
I guess gitlab failed to distribute the container to the runner?
<DavidHeidelberg[m]>
it's a bit weird, because previous run was on same runner fdo-equinix-m3l-21
<MrCooper>
FWIW, I think this refers to "container" as in "podman container", not as in "podman image" (for which the term "container" is somewhat ambiguous)
<MrCooper>
looks like maybe it tried to re-use an existing container, but that failed (maybe the container was destroyed just at a bad time?)
<DavidHeidelberg[m]>
also I have to state it happen only once, other stress-tested runs are fine
Haaninjo has joined #freedesktop
GNUmoon has quit [Remote host closed the connection]
<DavidHeidelberg[m]>
SamuelMarks: well, it's take it or leave it. No-one force you to sign something, its just you acknowledge that vague definition of not being complete asshole (I read the stuff yesterday... :D )
<emersion>
yup
* SamuelMarks
replied on thread and is now off for ~an hour
<emersion>
yes, if you haven't been harassed before in a large open-source project, it's easy to see the CoC as pointless. but i assure you *many* people are very glad that it exists
<SamuelMarks>
I've worked professionally and not with many engineers, some of whom are a little rude
<SamuelMarks>
But you can just have a thicker skin
<SamuelMarks>
Think of it like having rms as a friend. You just need to not be an ass back to him
<emersion>
"have a thicker skin" is _not_ an acceptable solution to harassment, sorry
<SamuelMarks>
emersion - Yeah all these terms are rather amorphous. They can be defined in such a way that one could claim that message is harassment^
<SamuelMarks>
Simple "Don't be a dick" agreements are far superior
<emersion>
this is even more ambiguous
<SamuelMarks>
yeah but it's not trying to be legalistic
<SamuelMarks>
You've got this legalistic looking document that goes out of its way to not define its terms
<emersion>
you're complaining that it's legalese, but then you're complaining that it's not detailed enough
<emersion>
i'm confused
<SamuelMarks>
What I put in bold in that response on GitLab clearly shows terms which are ambiguous
<emersion>
doesn't sound very ambiguous to me
<SamuelMarks>
emersion - I'm anti-CoC entirely. You don't need people to sign a document agreeing that they won't be a dick
<SamuelMarks>
emersion: Define "inclusive". Define "best for the community". Define, in context, "empathy". Define, in context, "experiences".
<emersion>
well we do. otherwise it's much harder to take action when someone is acting badly
<SamuelMarks>
FreeDesktop reserves the right to kick people out of any repo or org or kick their entire user off the platform at their own discretion
<emersion>
and people may not be treated equally etc
<SamuelMarks>
Why can't you just have a line like that? - Maybe say "own discretion" needs a certain number of people agreeing, but I don't see the issue otherwise
<SamuelMarks>
inverting to the user seems bad
<emersion>
honstely i've had enough of this discussion, sorry
<emersion>
if you don't like CoCs, don't use FDO
<SamuelMarks>
well as I said 8 minutes ago I was running off for an hour now anyway
<emersion>
it's a bit too easy for a user to claim that CoCs are useless, when they haven't been on the other side of the fence
<SamuelMarks>
emersion - I've been an open-source and free software contributor for a long time. See my GitHub and old Bitbucket for proof. Also I didn't say they were useless, I said they are ambiguous and just an excuse to not take responsibility for managerial action. Just have the balls to "kick people out [...] at their own discretion"
* SamuelMarks
sets status to away for ~1 hour
AbleBacon has joined #freedesktop
<eric_engestrom>
SamuelMarks: anyone can code, but clearly not everyone can behave while doing it
<eric_engestrom>
I have to say SM' behaviour towards the idea of not being an asshole is not filling me with confidence that this is the kind of person we want to allow into our community
ximion has quit [Quit: Detached from the Matrix]
<gry>
emersion: if there is no "coc" and i get harassed by X, doesnt it mean you can kick out X just because common sense?
<gry>
you mention "dont be a dick" is ambiguous; as long as the admins know what it means, i would be cool with it
<pq>
me neither - it could have been easy by just checking the box, but starting by arguing against the values of the community you want to join is not a good sign of fitting in.
<gry>
maybe the main point is to highlight to everyone what are the red flags which they need to report?
<gry>
say if a coc doesnt say that asking me in privmsg for a ASL is rude, then i will not know to report my attacker to an op
<emersion>
one could argue whether an action is worth banning someone or not
<gry>
probably thats a most valuable impact of a coc
<emersion>
one could be a bad admin who harasses other people
<gry>
ops need to decide, so my argument doesnt really matter
<emersion>
one admin could ignore complains
<gry>
right, thats a coc for admin cooperation :P
<gry>
separate business
<emersion>
the CoC is just here to set things straight
<emersion>
no treatment of favor, clear team handling the reports, etc
SamuelMarks has quit [Ping timeout: 480 seconds]
pjakobsson has quit [Remote host closed the connection]
<emersion>
also less room for banned people to complain
<emersion>
also a guarantee for everyone that we're trying to be decent human beings
<gry>
yeah i guess a half of it sounds like ops cooperstion
<gry>
cooperation
<gry>
that SM guy can just fork and keep his own fork; if it is useful, a signed contributir from here can pick and merge
<gry>
have a lovely day
SamuelMarks has joined #freedesktop
<SamuelMarks>
eric_engestrom - Community? - I don't understand. I just want to send some code over
<eric_engestrom>
SamuelMarks: the fact you don't understand the concept of a community is exactly the problem here
<eric_engestrom>
SamuelMarks: I recommend doing your homework, reading up a bit, and coming back once you're ready :)
<SamuelMarks>
at lunch the other day rms was telling me how he's Saint IGNUcius
<SamuelMarks>
I guess community can be religion
<SamuelMarks>
(though this was just parody; you seem to be taking it seriously)
ilmais1n has quit [Remote host closed the connection]
vkareh has quit [Quit: WeeChat 3.6]
vkareh has joined #freedesktop
todi has quit [Remote host closed the connection]
ximion has joined #freedesktop
<pq>
I've forgot all the very few cases when I've been harassed, but it is really comforting to know that should anything happen, the community will take it very seriously. No-one will be able to shrug their bad behavior off as a joke.
* SamuelMarks
is rewriting it in CMake and would be good to see a few variants. For some reason it wouldn't work from Cygwin (I'll reboot into Linux sometime to check there)
<SamuelMarks>
Though it's only ~400 lines so I guess I can hand reverse-engineer it
todi has quit [Remote host closed the connection]
todi has joined #freedesktop
todi has quit [Remote host closed the connection]
todi has joined #freedesktop
gert31 has joined #freedesktop
oldpcuser has quit [Remote host closed the connection]
oldpcuser has joined #freedesktop
todi has quit [Remote host closed the connection]
todi has joined #freedesktop
todi has quit [Remote host closed the connection]
todi has joined #freedesktop
SamuelMarks has quit [Ping timeout: 480 seconds]
todi has quit [Remote host closed the connection]
SamuelMarks has joined #freedesktop
tzimmermann has quit [Quit: Leaving]
systwi has joined #freedesktop
todi has joined #freedesktop
ximion has quit [Quit: Detached from the Matrix]
todi has quit [Remote host closed the connection]
todi has joined #freedesktop
todi has quit [Remote host closed the connection]
todi has joined #freedesktop
SamuelMarks has quit [Ping timeout: 480 seconds]
todi has quit [Remote host closed the connection]
todi has joined #freedesktop
Kayden has quit [Quit: -> JF]
gchini has joined #freedesktop
Kayden has joined #freedesktop
SamuelMarks has joined #freedesktop
ximion has joined #freedesktop
systwi_ has joined #freedesktop
systwi has quit [Ping timeout: 480 seconds]
sima has quit [Ping timeout: 480 seconds]
vkareh has quit [Quit: WeeChat 3.6]
SamuelMarks has quit [Ping timeout: 480 seconds]
<DavidHeidelberg[m]>
gitlab pro's, question: if I have a prog1..n.yml files which I include with include: local prog1.yml; prog2.yml in .gitlab-ci.yml, can I with some clean syntax make it conditional depending on changes to these prog1..n.yml files? Current impl. is here, and currently I have to fire a container to check, which is... quick, but still wasteful: https://gitlab.freedesktop.org/gfx-ci/ci-deb-repo/
<DavidHeidelberg[m]>
I tried use CI_JOB_NAME but due to usage of parallel matrix, the job names doesn't correspond to the $JOB.yml file and I cannot do variable postprocessing in gitlab-ci directly.
karolherbst has quit [Remote host closed the connection]
karolherbst has joined #freedesktop
<whot>
DavidHeidelberg[m]: if you're willing to use the jinja templates, you can probably set something up there? see ci-fairy generate-templates
<DavidHeidelberg[m]>
whot: I'm trying to avoid that stuff really, hard, but I guess at some point I'll have to give up and use it...
<whot>
what's the reason for avoiding it? other than it's a two-step thing but...?
<DavidHeidelberg[m]>
I hate adding layers, since they (usually) compicate thing in long-term
<DavidHeidelberg[m]>
I'll check it tomorrow, too late this timezone (1:51am) for it :D
<whot>
in your case adding a package would be a three-liner in the .gitlab-ci/config.yaml like - { name: app_name, url: foo, sha: abcd }
<whot>
it's a basic text processing, so I'd add the layer now, get things working as you want them and remove the layers if it becomes a problem later
<DavidHeidelberg[m]>
sounds good, I'll explore tomorrow, maybe ask you for some advices 😉
<whot>
libinput and libei both use it (copy/paste), the latter one is less complicated and probably the best one to check out
<whot>
but in the end it's mostly a bunch of {% for foo in foos %}{{foo.somevalue}}{% endfor %} and a lot of cursing at vim because it doesn't syntax-highlight the templates properly
oldpcuser has quit [Quit: minha internet e uma desgraça]