Not too sure about the BPM tap along idea. It could be added, and it probably is a good idea to do this if the server side infrastructure exists for this, but I'm not too sure Picard is the best spot for this.
mayhem
yeah, not sure if this is a good idea at all.
outsidecontext
You usually don't use Picard for playback, rather maybe to listening to a short quickly to identify it. I think a better place would be actual players
Or maybe there is a way to gamify this a bit. Make it fun to tap along. Could work well as a mobile app I think, with a touch screen
Also if this feature needs actual real time access to the audio stream during the playback this would need some more work on Picard. The current player is really simple, using QMediaPlayer to handle everything
mayhem
ohhh. oooh. yes.
that would make sense.
and that could work for the streaming case too.
outsidecontext
I have no concrete idea yet how exactly this would work, but I could imagine tapping along on your phone to some cool music could be some fun way to kill some time if it is presented right :D
alastairp
an interesting possibility for a game would be to use some kind of recommendation system, so that we try and give users something that they're interested in listening to, but we make specific recommendations of either tracks for which we don't have data, or for which we want an extra confirmation in the case that we have conflicting existing data
outsidecontext
+1
For me this sounds like BP would be the natural first place to have this feature, with the possibility of a mobile app and maybe even player plugins
MRiddickW joined the channel
also because it was asked above how many users Picard has: as mayhem said there are no general numbers. We have some numbers for the installations via Canonical's Snap and from MS store. But I suspect most users are using one of the binary downloads for Windows and macOS, or for Linux users their distribution's packages
atj
outsidecontext: totally agree about Picard, I just don't think it's a good fit
when you click start next to Data safety, it'll display definitions of various terms. we don't collect any data so fill no on next page i sufficient then submit.
I kinda liked the "but we don't actually care about yo biz to peek" was kidna funny
couldwe . idunno, add aa "besides we don't actually cre enough to .." addition as well? :D
lmao "A Touhou hijack occured" -> "Bad Apple has transcended media"
oh wait
i thoguht it was a gif of osmething. but it's a video playing but i had the sound off os don't reaslie
wtf autoplay shit
jokes on thme i didn't hear it
wtf if i scroll up it stops playing what is this shit
ugh modern internet
is it any wonder i prefer older browsers
i thoguht it was a gif of osmething unrelated, that "tohou" was something else random name of something, and peopel refrenced "bad appole" becasue of the name was the same
KassOtsimine: I generally agree, but for a privacy document I think it's important to be as clear and formal as possible :)
lucifer
alastairp: that PR doesn't modify dumps yet. i intended to dumps in a separate PR. i haven't tested the PR in prod currently, will do now.
mayhem: oh yeah, one would think public clouds earning billions in revenue could add measures to prevent situations like this but afaik none does.
KassOtsimine
reosarevok: yea i guess
alastairp
lucifer: cool, thanks. I have my local setup with the migrated db too, so I can try it too
lucifer: what is the state of the listen counts in this PR? We'll keep the same system but use ids instead of usernames? Or we'll get the username anyway and continue using them?
I made a bunch of suggestions - I think that there are some more things that can be improved, but I didn't want to be too picky about wording. I think that it's good that we have _something_, we can improve each section bit by bit as we work on related bits of code
talking about documentation, is listen counts documented somewhere? I saw a bunch of discussion last week about it, and it sounds like it has a few tricks in the way that it works that would be good to be written down
lucifer
thanks.
no it isn't yet. i'll do that as part of the listen count PR.
reosarevok: Not urgent either: Added doc for handling SIR Queues alert.
reosarevok
I saw, will read in a bit :)
bitmap
akshaaatt: re: design.metabrainz.org, it seems like a lot of work for something that won't be used very often. do you have an example of how this would be used?
akshaaatt
If it's anything complicated, we can leave it out since I already have it hosted through chromatic on a free tier bitmap
For now, it's just the storybook
bitmap
ok. well, we'd have to write a Dockerfile for it and then set up some services. but it's very easy to run locally
akshaaatt
Yeah right
monkey
akshaaatt: Did you look into deploying to Github pages, by any chance?
akshaaatt
I could do that monkey, sounds simple
monkey
I'm asusming it can be done automatically using github actions
the4oo4 "Checkout code", "NPM install" and "Deploy Docs to GH pages" steps should probably work for the design system pretty much as-is. Now it's a matter of figuring out the step to build the storybook to the right directory
akshaaatt
Right
monkey
I've just spent a couple of hours today with my head in GH actions, so happy to help if you have questions
akshaaatt
Wonderful! Will do the set up and ask if I face any issues. Thanks monkey :)
Ok. So, I can see the point with that, I was just struggling because there's an "editing disabled" flag, which is not just the same as "editing is disabled" :)
bitmap nods
I guess "spammer automatically sets the others" is an option, if you feel that's the simplest?
yvanzo
I have it on my review list as well.
bitmap
I think it might be. otherwise we have this situation where one flag implies other flags, which we'd have to handle wherever the implied flags are used
so just setting them all outright simplifies that
reosarevok
I guess my only worry is that means you can unset the others without unsetting spammer
But I guess that might be fine for an admin
Or we could actively freeze the checkboxes, I guess
well we could certainly check that any "implied" flags aren't being unset in the controller. but that seems better still because it's only one place to check it
reosarevok
Ok
So how would you imagine this working? Clicking spammer locking the others as checked seems ideal, but I'm not sure I can be bothered implementing that on a TT page and if we want to merge today porting the page is also not reasonable
So maybe it should just be automatic in the backend for now
bitmap
yes, in the backend sounds fine. if you'd like you can just put a message on the page saying the spammer one will enable the others
reosarevok
So we undo the changes to is_editing_enabled
bitmap
yeah
reosarevok
And similarly, stuff like " !$editor->is_adding_notes_disabled && !$editor->is_spammer;" just goes back to just the first half
And then I guess in update_privileges in Data::Editor
bitmap
that should be redundant, though I don't see an issue with keeping it just in case
reosarevok
We do what, + ($values->{editing_disabled} || $values->{spammer} // 0) * $EDITING_DISABLED_FLAG ?
No mailed-in reviews today, so the usual roster are up: bitmap, lucifer, yvanzo, reosarevok, alastairp, mayhem, akshaaatt, monkey, zas, CatQuest, Freso – anyone else who want to give review, let me know ASAP!
bitmap: Go!
bitmap
hey
last week I worked on updating some PRs to fix MB container warnings
and improving the MusicBrainzAlertDiagnostics.md doc in syswiki
I submitted a PR to enable logging of requests from MB website/web service containers to aid in handling alerts. that was for one of the review comments