#metabrainz

Also reviewed MBS PRs, and started to work on moving test.mb.o search to Solr.

!m yvanzo

You're doing good work, yvanzo!

That’s all from me, go iliekcomputers!

iliekcomputers already went.

zas: Go!

Hey

zas then?

(People still up: zas, reosarevok, Mr_Monkey, Freso, spellew, aidanlw17, amCap1712, Cyna, CatQuest – anyone else, please let me know explicitly ASAP.)

(Minus zas.)

Mainly kept things up, upgrades (all systems, plus grafana), I also fixed few things related to admin stuff

Also did quite a lot code reviews (picard, picard plugins, picard website): we plan a release before summit

We'll prolly announce feature freeze this week

That's all for me. reosarevok?

Hi!

I worked on MBS-10299

MBS-10299: “Voting Reports” (pre-set edit searches) https://tickets.metabrainz.org/browse/MBS-10299

MBS-10320

MBS-10320: Don’t wrongly nag local users of MB slave server https://tickets.metabrainz.org/browse/MBS-10320

And a bunch of shorter things

This week I'm looking at old tickets again

I also did quite a bit of fixing editing messes

CatQuest: what about you?

hi

Meow! `😺`

Have had a busy week! My foot's really starting to heal now so I've been walking about, enjoying movemental freedom and getting things done I couldn't before

I've also worked on instruments: added the reyong (INST-709), the calung (INST-698) (especially this was difficult, still not completely happy about it, see INST-832)

also started some seecrit work ♪😽

Finally summer has done a little comeback here so therefore I'm at Hovedøya right now :D

That's it for me `😸`

findus!

INST-709: Reyong https://tickets.metabrainz.org/browse/INST-709

INST-698: Calung https://tickets.metabrainz.org/browse/INST-698

INST-832: Translate article Calung https://tickets.metabrainz.org/browse/INST-832

next...

am?

erh

amCap1712:

Or maybe Mr_Monkey ?

Hallo !

Last week I continued helping akhilesh along to complete the API project, as he described.

(People still up: Freso, spellew, aidanlw17, Cyna, amCap1712(?) – anyone else, please let me know explicitly ASAP.)

I also started figuring out how to add elasticsearch to travis for CI tests of the search endpoints.

And that's it for me !

spellew: go !

Or maybe aidanlw17?

Hi all!

I can go

Being the last week of GSoC, I spent most of my time cleaning things up and making sure everything was working well, across all aspects of the similarity pipeline.

Did some final testing and got some pesky things worked out.

Also wrote my evaluation and blog post

I’m happy with what we achieved and I can’t wait to see where it goes in the future.

:D

Pretty sad that the summer is ending :( thanks to everyone for being part of this great community!

I’m honoured to have been part of it and I plan to stick around :)

Fin

Cyna, go?

Hi

This week was most productive for me. I finally converted all the MB entity's editforms and added previews to them

(Only myself left on the list—and maybe spellew and/or amCap1712 if they show up. Last call for anyone else who may want to go.)

Most of the time I spent on completing the gsoc project which is mostly now complete. I couldn't add test cases which I'll be starting to work on once the main code written during the gsoc period is committed

and all the issues with them are fixed

Thanks for helping me during the gsoc period bitmap, couldn't have done it without you :)

Fin !

Freso: wanna go next ?

Well, not really any other options currently. :p

But sure. Thanks. :)

(amCap1712, spellew: final ping; otherwise I’ll move on once I’m done with my review.)

🙋

..

So, by end of last week I notably tried to organise all y’all GSoC students’ blog posts.

!m Freso

You're doing good work, Freso!

aidanlw17, spellew, pristine__, Cyna, and akhilesh have their posts up on https://blog.metabrainz.org/

amCap1712 might have a post for a later day.

!m GSoCers

You're doing good work, GSoCers!

^

I also looked into an issue with Jira, but unfortunately couldn’t find resolve for it. :(

was that he label thing?

Spent some time trying to get an overview of all the reports that came in over last weekend.

And other than that, trying to catch up and be around/about etc.

fin.

KassOtsimine: Yes.

No word from spellew or amCap1712.

So that’s it for reviews!

One more item on today’s agenda:

So I was hoping to have this up for discussion at the summit, but maybe it’s great to start the discussion/thinking now and we may be able to have some more fruitful discussion/planning come summit time.

...

(also not everyone can Make the summit)

zas, bitmap, yvanzo, reosarevok: This will likely be mostly oriented towards technical "solutions", so please be around. :)

I do

KassOtsimine: The discussion is mostly between me and MBS devs/sysadmins (ie., the four people I just pinged).

The problem currently is…

jeez

Sometimes we have people coming in to MB acting in bad faith.

And they get reported—sometimes massively so—and I will eventually either disable their editing/voting rights or just straight up delete their account.

But they are free to immediately turn around and make a new account. In fact, many people where I’ve disabled their editing/voting rights delete their own accounts and create new ones and continue on with their behaviour.

We currently have no way to block e‐mail addresses or IP address (ranges), no way to find editors via e‐mail, or any way to see what IP address a given editor comes from.

Would these (or some of these) be feasible to implement?

Any other ideas for tools that we might be able to implement?

i don't see a problem this for user admin like freso

we have a way to block ips on the gateway, don't we? though I don't think we'd want to block them permanently

We have, but it's not easy to manage atm

bitmap: Yeah, but we can’t associate editors with IPs.

So we won’t know which IPs to block to begin with.

Throw-away email addresses are easy to obtain, and you can't always rely on an ip address (spoofing, vpn, etc.)

To prevent creation of new accounts we could start by blocking emails they previously used

(And yeah, I agree that we won’t want the blocks to be permanent.)

but yes, we'll need to show editor ips to admins in mbs

rdswift: Yes, but we can make it harder.

in fact I think there is a ticket for that

We can associate editors with IPs in Sentry.

instead of blocking emails (that will mean they realise they just can use a throwaway

if they happen to trigger an error that gets posted to sentry, I guess

notify admin when new user uses email on "blacklist"

rdswift: Right now you can registers 100’s of accounts with a single e‐mail address. (Which in general is fine, but we also can’t search via the website interface for accounts by e‐mail address, so I can’t find all the addresses with the same mail.)

Freso, I agree it will help. Just won't catch all of them.

Anyway, I’m not sure I necessarily want a specific actionable path forward—but I would love if y’all could think about this and then maybe we can talk about it in a week or two and then again at the summit?

bitmap: yes, we can make that happen when needed.

*actionable path forward right now/today

but zas’ suggestion seems simpler and less efforts to implement.

Sounds like a great first step.

showing the ips in mbs doesn't seem too hard

But since I can’t search for e‐mail address pattern, I can’t see if there are more sleeper accounts matching this pattern.

Being able to see different accounts associated with a given IP would also be a great help.

Also the new feature MBS-6791 should help with finding revenants: https://musicbrainz.org/search/edits?auto_edit_...

MBS-6791: Search for edits made by beginners https://tickets.metabrainz.org/browse/MBS-6791

what means "revenant" :o

KassOtsimine: "one that returns after death or a long absence "

cool

(source: https://www.merriam-webster.com/dictionary/reve... )

I've seen that word used occasionally

bitmap, yvanzo, zas: Can you ruminate on this and then we can discuss it again in a week or two and again at the summit?

sure

(If you haven’t implemented things by then already. 😇)

hahaha

Alright, I think that’s all for this topic for tonight then.

Thanks for your input on this item. It’s something that’s been bothering me for a while. :)

And we have no more items on today’s agenda.

Thanks for your time everyone!

</BANG>

Fanks !

Longer term, this might be worth looking at zas's proposal in https://community.metabrainz.org/t/idea-editor-...

rdswift: Something like that, yeah. I’m hoping for something more short term to begin with. :)

This link to search edits is much faster: https://musicbrainz.org/search/edits?auto_edit_...

Agreed. This is a much longer term solution.

I assume the discussion was partly driven by MBS-10323