#bookbrainz

Hello Mr_Monkey

Hi !

What can I do for you akashgp09 ?

Currently there is no logic to calculate editor's repuation. All the editor repuation value is 0 in the database.

Moreover the reputation value is hard coded `as <dt>Reputation</dt>

     <dd>0</dd>` in the bookbrainz-site.

How about incrementing repuation value with +3 points when the editor applies a revision.

Yes, that's correct. The idea is that at some point we'll be able to revert revisions (that's currently not implemented), and that should update the users reputation (with the assumption that a revision that needs reverting is a bad revision, hence -1 reputation)

But I agree that we can already start the incrementing part. Perhaps +1 for a non-reverted revision

I will take this over and implement the incrementing part for now. Thank you : )

No, thank you !

hmmmmhmp... i can't help but feel. if you'll let me say it. that this "reputation" thing is maybe a bit misaligned in priorities..

a bit of a thing is that we see in eg. lb how ther eare lsiten-spammers now.

i mena it's easy ot get a "high" reputation by doing things like many small implemntations. or intentionally doing something wrong or halfway, jsut ot enter evne more revisions later...

what I mean is that I think really that some form of anti-"gaming" the reputation system needs ot be in place for mthe start

the same is true for reverted = -1 in reputation

say

say if the datamodle changes, the styleguidlines are latered. etc etc. which owudl mean 10 years alter old revisions that were made wich were *correct* at the time are reverted or chnaged becase now, they are not

in mb, i have edits that where voted down 13 years ago, but are the actual way to do it now. it was wrong for the datamodel then (a-head of its time if you will) but correct now

another 15 years in the future an who knowns what *brainz datamodel will look like

not to mention grudge reverting to make someone elses reputation plumit, etc

all the issues with voting in mb will be transferred to reputation

i don't thingk it's a bad idea to have a statistic on people's profiles which show how many of theri revisions whre reverted mind. but like, let the stats and the revisions speak for themselves

In case there are any 🤓s here who’d be interested in a free ebook about Palestine… https://www.haymarketbooks.org/blogs/307-free-e...

(Or non-🤓s for that matter.)

uh. what is 🤓 anyway? a bespectabled person with big teeth??

(free books yay)

U+1F913NERD FACE

:P nerd face

(Had to give Unicode codepoint too, of course, as a true 🤓.)

:D

actually I'd be interested in your view about the above ^^as yo uworke with comunity and spammers and such

I think it’s fine to experiment with things and try things out, and figure out how to deal with it if/when issues arise.

No system is perfect, so let’s not let the hunt for perfect prevent us from finding something good.

I agree with that sentiment very much

i'm not actulaly looking for perfect, I'm looking for "let's make sure to learn of our previous experiences"

perhaps. a robust fundation to build the experiemnting on. right now we still lack rudimentary infrastructure :D

(eg, reversion, series)

now at first I tink the reputation thing won't be an issue at all, sicne the community is stll fairly small and dedicated 🤓s :D

i just don't wnat us t oget i na possition where we have to redo everything or something becasue the systme we wrote didn't have salability in mind (here scalability is: more users and spammers)

I main, push come to shove, we just DROP (or ignore) that column.

I guess :)

I don’t think this is necessarily such an invasive experiment that it would involve a lot of work to "undo" it.

that's true

i don't want to get into a situiation where people game it and use this as soem sort of "I'm higher reputation thna you o stfu" or the like

Yeah. If anyone does that anywhere in MeB (e.g. an autoeditor to a "beginner" editor), please do report them/let me know. :)

indeed

usually i get the opposite, fiarly new editors acting liek i don't know siht XD

usually i get the opposite, fairly new editors acting like i don't know shit XD

Maybe the issue is more the naming? Maybe "reputation" suggests a value judgement and we could do with "applied revisions" and "reverted revision" (like on MB)

Yeah. :)

yea! that's a good point

Unless we want the system to evolve into some other form of reputation calculations that actually reflect a user's standing in the community, but nothing I can think of right now

I try really hard thne not to drag my history aroudn and be like "yo, I'mma let you finish, but i've been her efor 15 years now and.."

hmm yea

it's hard to *programmaticaly* calculate these things

Indeed

which is what I meant with "letting the stats speak for themselves"

for AE electios. I look at failed edits. see how long ago they where. see what kidn of arguments they made.

CatQuest: Re: that, I also think a big part of it is exactly the community/social aspect. Communities have certain norms for behaviour and those can be really hard to change. My job is essentially to try and steer the community’s behaviours/norms in a "good" direction and help prevent bad actors from taking them in a "bad" direction. New people who are not "socialised" yet obviously haven’t "internalised" (or learned to

mimic) these norms/behaviours yet. :)

exactly!

and.. i gus i t thing that having a "reputation" mecanism, sort of attracts "bad" direction a bit? idk

:D

eventually we're going to need a comm. man. on bb as well (not it!) but that is stil lfor the future

I think for now, as was mentionned before, it's OK to start trying out this feature.

If we feel like "reputation" is too big a word for glorified addition and subtraction, we can always just rename it.

good point

i.e. applied revisions/ reverted revisions

I mean, I’m CommMan for all of MeB AFAIK (which is also why I’m poking at the LB sockpuppets), so I assume I’m also for BB… :)

oooh :D

well then

then since I have you both here. there is one thing: looking at https://beta.bookbrainz.org/statistics you'll see both "goldilox" and "agathacrustie" (wich was apparently the same person too) and this person deleted their account (twice)

But yeah. I’m fine with experimentation, as long as the community is okay with it, and as long as we’re not going down a path that will be (really) hard to diverge from.

since that was done it shoudl be removed from bb too right?

Yeah, seems like their accounts are gone from MB.

There’s a ticket for this…

ah good

BB-260: Implement the user account deletion endpoint

i susspect i'm already voting/watching this

huh, no. well am now :)

(Only the LB endpoint is currently implemented, so the MBS side of things is still waiting for BB, AB, CB, MEB.)

Yeah.

I'm not so sure what should happen when a user is deleted. I think currently that'll break a bunch of things

i suspect a "Deleted user #number" rename and inhability to edit under hat name

s/edit/make revisions

But I suppose a "soft deletion" (marking as deleted in the DB and displaying "deleted editor #124" on the web pages) would work

I assume that's how MB does it too

basically I think that's what mb does tbf, since aoutright removing there too would break things

hah

But we should also probably scrape all details from said account too

right now bb revisioners are rowid linked as well

Yeah, MB currently renames account to "Deleted Editor #${rowid}" if the editor has any edits, notes, or votes.

A, so rename directly in the DB, yeah that makes sense

If there is nothing that will stick around after deletion, editor is just deleted.

cool i didn't know that

Makes sense

Mr_Monkey: do you want a ticket on this?

Mr_Monkey: Do you have access to see the MB row ids for https://beta.bookbrainz.org/editor/1898 and https://beta.bookbrainz.org/editor/1820 ? (Not sure if that’s stored on BB’s side.)

It is indeed stored

Yes please CatQuest :)

Actually, come to think of it, one question could be whether we want to synchronise the editor renaming (ie., so it’s the same "Deleted Editor #xyz" on both) or whether it’d be better to _not_ do that, to increase the "anonymisation" of the deleted accounts. 🤔

On one hand it might be good to be able to link them across the sites. OTOH, being able to link across sites means greater chance of being able to identify the editor post-mortem.

I’m kind of leaning towards keeping them in sync if possible, to be able to keep track of deleted bad actors.

That's sometimes useful for us for tracking vandalism and such, I suppose, but we do have the MB row id as reference

Right.

Maybe a question for the main channel at next meeting?

The question is kind of hurting my brain right now. 😂

my gut instinct is to keep the bb rowid on bb deleteds for *now* but to be able (for account admins) to see the mb rowid easily

I think I’ll consider it a bit more. CatQuest, can you link ticket when you make it? I’ll try and put the question there and then maybe mull it over some more and maybe bring it up on Monday.

https://tickets.metabrainz.org/browse/BB-624 was jsut about to

BB-624: Mark/rename deleted users

OH, you linked chatlog already. 😂

:D

Thanks folx !

I've had experience making tickets for bb by now :D

Isn't that your full-time job? :p

hahahaha

now to get peopel to pay me for it! :P

"but people are only willing to produce value when given monetary incentive, hence why capitalism is the only system that can work2#!$@!?#$!?!?#$!@>#!#$<!$!@" 🙄

I mean, yes. 👀

uh I mean hat I was kinda joing :P (I'd still do it evne if never paid :P))

joking*

since monkey sid "job" :P

said*

anyway that capitalism incentive also means people *only* do it for the money, so they don't care about it at all.

so it's less quality work.

only if someone enjoys what they do

I know. :) I’m just in a very (outwardly) anti-capitalist mood lately. :p

it'll be the thing they happily do. yea

yea

it was more a reaction to the "job" comment

(on the other hand i 'd want peopel who do a good job have thier wrok rewarded/valued)

ugh jazzhands