#metabrainz

<07munishk> Thanks @techmorningstar , will submit it

@yvanzo @bitmap hi , I have updated the proposal and made the changes as mentioned, is there anything else I should mention in my proposal? here's the link: https://community.metabrainz.org/t/gsoc-2024-in...

Will also work on sql-gen from todayand try to fix it to make it directly usable with the app we are coming up with.

rimskii[m]: i think you can add importing playlists from apple music and soundcloud to your proposal to make it 350 hours.

<lucifer> "rimskii[m]: i think you can..." <- I was planning to do that way actually

well I think I can submit the proposal now?

rimskii[m]: yes but look into the apis and add details for the both to your proposal.

sure, thank you !

may I show it to you once i finish?

sure, post the updated proposal on community, rimskii[m].

okay!

mayhem: hi! for sonic integration, do we want to support playing music from a sonic server running on the localhost and not exposed to the internet?

Lucifer: if that is possible it would be great!

mayhem[m]: due to the nature of subsonic auth requiring passwords and tokens, its a bit risky security wise but might be possible

Subsonic auth is such a disaster...

but i guess if the server is restricted to localhost then not much of an issue, maybe we'll need to implement auth for public facing and localhost servers in separate ways.

mayhem: need to add DNS records for a few subdomains on wolf to put up oauth stuff for testing.

Ok, np

dev.metabrainz.org, oauth.metabrainz.org and dev.listenbrainz.org

need DNS records similar to what datasets.listenbrainz.org has

liberachat is rebranding into libera cat and I for one aprove of it :DDD

:fish:

good thing its for april fools only, because before too long cataluyna would take offense at that site, since there is no catalan on the website.

lucifer: "dev 10800 IN CNAME datasets.listenbrainz.org." already exists.

hmm i see.

but it should work fine for what yo uneed.

ok, all done. check to make sure its what you need.

mayhem: will check and let you know

:disappointed: https://usercontent.irccloud-cdn.com/file/6Sfdj...

Eek, looks like javascript

lololol 😢

lololol @ https://x.com/FFmpeg/status/1774750713798627513 (context: bitmovin is a bay area startup which basically offers ffmpeg as a service with a nice gui for non technical users. and instead of contributing back anything, disses on ffmpeg for not doing more of their work for free)

glad to see the they are hitting their freeloaders harder. good good.

[listenbrainz-server] 14munish0838 opened pull request #2826 (03master…master): LB-1459 Add integration tests for websockets backend https://github.com/metabrainz/listenbrainz-serv...

lucifer (IRC): I uploaded my final draft ! I wanted to ask if you could take a look and give feedback :)

rimskii[m]: looks fine

yellowhatpro: your edits lgtm, just make sure to submit your application asap if you haven't yet

lucifer: i'm looking through some artist names and "Amelia TALEXIS - Sopran, Agustarello AFFRE - Tenor" this is one, i can split it using comma as they individually have items, but

"Muir Mathieson, conductor" this is also a creator

if i split then "conductor" also becomes a artist

thanks bitmap! Omw 🚀🚀

huhridge: yes, will probably need multiple ways to split and see which one yields the best results. split one way, search artists, assign a confidence score. if high enough, move on otherwise retry with another method.

no meeting today< I assume

Should be one according to https://community.metabrainz.org/t/notes-from-m...

mayhem, reosarevok ?

oh hai.

lucifer: how should i formulate the confidence scores, maybe if the results are obscenely high, i discard the artist?

yeah, we sort of agreed to have a meeting, but given that few of us are here, shall we skip it?

perhaps the meeting itself was an april fools joke?

I'm having dinner so I'm fine with skipping

and we all fell for our own silliness.

well, if no one has serious objections....

I do have one mailed in review tho

huhridge: in other places in LB, we search for the sanitized in database. then take the levensthein distance between the match and the search query for the score.

So I will post it:

<BANG>

Oh

I have four now :D

post 'em then close the meeting, lol.

Couscous said:

Well, if there is a meeting:

Sorryy

Hi folks,

Apologies for the busy week. I didn't get to work too much on the tasks.

However, I did visit my grandparents' place, and I am feeling very loved and happy. :D

Regarding my little progress, I delved deeper into the part that I overlooked when writing my proposal on revamping the listen importer, thanks to Lucifer for pointing it out.

I will address these issues in an updated version very soon.

Additionally, I received some feedback on my open PRs, and I will be working on them this week. Full throttle ahead!

Cheers!

kellnerd: go if you want while I post the next ?

ericd said:

I can give a short update too I guess

Last week finally submitted proposal. Also updated PR#2824 to support descending sort by confidence in user's fresh releases.

yvanzo said:

Last week I looked into how we can use the blockchain to replace the database.

I just asked an LLM to generate the migration code for the decentralized application.

It should be able to run on mining rigs until quantum computers become available.

The result will be triple-licensed under the AFPLv9, SSPLv1, and WTFPLv2. Fin.

He might or might not mean that one

atj said:

<3

Last week I continued learning about Solr and  working on the Ansible role. I managed to get a 2 node SolrCloud cluster running locally with the mbsssss schema and query response writer installed, which was deployed and configured using Ansible. I'm hoping to get the role more or less finished this week so we can setup the new Solr 9 environment and start performing some testing.

That's the end of mailed in reviews

bitmap, wanna go?

sure

monkey: you after if you want

👍

we deployed the new artist credit editor code to beta, give a test if you can

I also deployed the new artwork-indexer and fixed a minor issue with some spurious deindex events

shit, sorry, I didn't really want to start the meeting. my goal was to post the updates and the we all run. lol.