#metabrainz

/

      • legoktm
        oh boy, renaming users
      • 2018-05-21 14104, 2018

      • legoktm
        that's always a blast
      • 2018-05-21 14125, 2018

      • central4 has quit
      • 2018-05-21 14144, 2018

      • d4rkie joined the channel
      • 2018-05-21 14135, 2018

      • D4RK-PH0ENiX has quit
      • 2018-05-21 14148, 2018

      • D4RK-PH0ENiX joined the channel
      • 2018-05-21 14142, 2018

      • d4rkie has quit
      • 2018-05-21 14131, 2018

      • D4RK-PH0ENiX has quit
      • 2018-05-21 14159, 2018

      • D4RK-PH0ENiX joined the channel
      • 2018-05-21 14107, 2018

      • Leo__Verto has quit
      • 2018-05-21 14118, 2018

      • Slurpee has quit
      • 2018-05-21 14145, 2018

      • sentriz has quit
      • 2018-05-21 14140, 2018

      • sentriz joined the channel
      • 2018-05-21 14139, 2018

      • UmkaDK joined the channel
      • 2018-05-21 14135, 2018

      • samj1912
        Morning people o/
      • 2018-05-21 14147, 2018

      • samj1912
        And evening Major_Lurker o/
      • 2018-05-21 14150, 2018

      • Major_Lurker
        good evening samj1912
      • 2018-05-21 14138, 2018

      • Major_Lurker
        ha you beat me
      • 2018-05-21 14145, 2018

      • obtext joined the channel
      • 2018-05-21 14137, 2018

      • samj1912[m] joined the channel
      • 2018-05-21 14102, 2018

      • UmkaDK has quit
      • 2018-05-21 14149, 2018

      • UmkaDK joined the channel
      • 2018-05-21 14108, 2018

      • UmkaDK_ joined the channel
      • 2018-05-21 14118, 2018

      • UmkaDK has quit
      • 2018-05-21 14123, 2018

      • ruaok
        \ø moooin!
      • 2018-05-21 14100, 2018

      • CatQuest
        yello
      • 2018-05-21 14153, 2018

      • iliekcomputers
        moin!
      • 2018-05-21 14124, 2018

      • d4rkie joined the channel
      • 2018-05-21 14139, 2018

      • D4RK-PH0ENiX has quit
      • 2018-05-21 14152, 2018

      • samj1912
        ruaok, can you take a look the reminding sir PRs?
      • 2018-05-21 14108, 2018

      • samj1912
        *remaining
      • 2018-05-21 14102, 2018

      • drsaunder has quit
      • 2018-05-21 14114, 2018

      • ruaok
        samj1912: in https://github.com/metabrainz/sir/pull/78 you are using the same refcount trick to make more productive labels float to the top?
      • 2018-05-21 14107, 2018

      • samj1912
        Yup
      • 2018-05-21 14113, 2018

      • samj1912
        I tested it out, seemed to work well for things like Sony and Universal
      • 2018-05-21 14130, 2018

      • Freso
        samj1912: Does refcount use relationships too?
      • 2018-05-21 14149, 2018

      • samj1912
        For some it does
      • 2018-05-21 14109, 2018

      • Freso
        A lot of labels are specifically not for Release Label use, but are still used a lot (e.g., n©b, GEMA, etc.). Not sure if it'll have any effect, but if I write "ncb", I'd expect n©b to pop up first. Is it still live on test for testing?
      • 2018-05-21 14133, 2018

      • samj1912
        yup
      • 2018-05-21 14151, 2018

      • Freso
        «yes, feel free to wreak havoc» you mean? :D
      • 2018-05-21 14102, 2018

      • samj1912
      • 2018-05-21 14122, 2018

      • samj1912
        just use the advanced query method since only that uses the right boosts currently
      • 2018-05-21 14136, 2018

      • Freso
        👍
      • 2018-05-21 14111, 2018

      • samj1912
      • 2018-05-21 14115, 2018

      • Freso
        samj1912: https://test.musicbrainz.org/search?query=publish… has "Technosis" 1st, which seems to only be used in one relationship, while the 2nd one is used for 27 Works and has a Discogs link.
      • 2018-05-21 14152, 2018

      • samj1912
        we are now returning just a count instead of returning all the ids and then counting them in python
      • 2018-05-21 14117, 2018

      • samj1912
        Freso: I dont think the ref count part is live yet for labels
      • 2018-05-21 14123, 2018

      • Freso
        Ah, okay.
      • 2018-05-21 14131, 2018

      • ruaok
        samj1912: got it -- makes sense.
      • 2018-05-21 14138, 2018

      • github joined the channel
      • 2018-05-21 14138, 2018

      • github
        [sir] samj1912 closed pull request #79: Improve query efficiency when calculating ref_counts (master...count) https://git.io/vppNk
      • 2018-05-21 14138, 2018

      • github has left the channel
      • 2018-05-21 14147, 2018

      • samj1912
        hmm, this is new, GH now removes PR approvals if you push/rebase
      • 2018-05-21 14154, 2018

      • samj1912
        which kinda makes sense
      • 2018-05-21 14156, 2018

      • github joined the channel
      • 2018-05-21 14156, 2018

      • github
        [sir] samj1912 closed pull request #78: Add release counts for labels (master...label) https://git.io/vpxhl
      • 2018-05-21 14156, 2018

      • github has left the channel
      • 2018-05-21 14100, 2018

      • iliekcomputers
        ruaok: re the GDPR thingy what do you think we should do, add logic for not creating an account if I disagree is ticked?
      • 2018-05-21 14111, 2018

      • drsaunder joined the channel
      • 2018-05-21 14134, 2018

      • iliekcomputers
        (I'm not sure why anyone without an account would login to go to the deletion page)
      • 2018-05-21 14104, 2018

      • samj1912
        Freso: now the refcount for labels should be live
      • 2018-05-21 14101, 2018

      • Tykling has quit
      • 2018-05-21 14149, 2018

      • samj1912
        reosarevok: ping
      • 2018-05-21 14156, 2018

      • Tykling joined the channel
      • 2018-05-21 14152, 2018

      • samj1912
        hmm, seems liek hes not here
      • 2018-05-21 14107, 2018

      • samj1912
        Freso: you wanna help :P?
      • 2018-05-21 14148, 2018

      • reosarevok
        samj1912: I'm out for a while longer, I'll be home in an hour or so I guess
      • 2018-05-21 14113, 2018

      • reosarevok
        If you have some short questions I can take a look, but no continuous checking rn :)
      • 2018-05-21 14151, 2018

      • Freso
        samj1912: https://test.musicbrainz.org/search?query=publish… looks a lot more sane at a quick glance now at least. :)
      • 2018-05-21 14150, 2018

      • samj1912
        reosarevok: wanted to tweak other entities
      • 2018-05-21 14155, 2018

      • samj1912
        lets do that when you are home
      • 2018-05-21 14106, 2018

      • D4RK-PH0ENiX joined the channel
      • 2018-05-21 14115, 2018

      • zas
        alastairp: are you ready to move acoustibrainz stuff from spike to another server ?
      • 2018-05-21 14106, 2018

      • d4rkie has quit
      • 2018-05-21 14158, 2018

      • D4RK-PH0ENiX has quit
      • 2018-05-21 14150, 2018

      • drsaunder has quit
      • 2018-05-21 14155, 2018

      • Leo__Verto joined the channel
      • 2018-05-21 14152, 2018

      • D4RK-PH0ENiX joined the channel
      • 2018-05-21 14115, 2018

      • D4RK-PH0ENiX has quit
      • 2018-05-21 14144, 2018

      • Slurpee joined the channel
      • 2018-05-21 14149, 2018

      • D4RK-PH0ENiX joined the channel
      • 2018-05-21 14154, 2018

      • reosarevok
        samj1912: around
      • 2018-05-21 14106, 2018

      • samj1912
        reosarevok: hey
      • 2018-05-21 14110, 2018

      • samj1912
        wanna do places?
      • 2018-05-21 14132, 2018

      • ruaok
        iliekcomputers: hmmm. I've been thinking about this and I'm really not sure how to proceed.
      • 2018-05-21 14140, 2018

      • ruaok
        lets make sure we see the same problem first.
      • 2018-05-21 14104, 2018

      • ruaok
        if I read the PR correctly, if they agree, they log in and all good.
      • 2018-05-21 14101, 2018

      • ruaok
        if they disagree, then we log them in and take them to the delete my account page.
      • 2018-05-21 14119, 2018

      • reosarevok
        samj1912: sure :)
      • 2018-05-21 14121, 2018

      • ruaok
        from there, the can just go about their business and never have agreed to the GDPR flag.
      • 2018-05-21 14101, 2018

      • samj1912
      • 2018-05-21 14102, 2018

      • ruaok
        I think it is a bad idea to just log the user in if they do not agree to the GDPR flag.
      • 2018-05-21 14106, 2018

      • samj1912
        this is the current state
      • 2018-05-21 14142, 2018

      • samj1912
        it currently has no boosts whatsoever
      • 2018-05-21 14155, 2018

      • samj1912
        address alias area comment ngram pid place type
      • 2018-05-21 14117, 2018

      • samj1912
        I guess we might want to boost the place and alias
      • 2018-05-21 14121, 2018

      • samj1912
        to 2 and 1.5?
      • 2018-05-21 14123, 2018

      • samj1912
        anything else?
      • 2018-05-21 14159, 2018

      • reosarevok
        Area is important
      • 2018-05-21 14113, 2018

      • samj1912
        okay, so 1.5 to that?
      • 2018-05-21 14115, 2018

      • samj1912
        or 1.2?
      • 2018-05-21 14119, 2018

      • reosarevok
        Maybe try that 1.5 too, yes
      • 2018-05-21 14124, 2018

      • samj1912
        can you do a test query
      • 2018-05-21 14125, 2018

      • reosarevok
        LEt's see if that makes a mess, if it does, 1.2
      • 2018-05-21 14134, 2018

      • samj1912
        then I can change the boosts and you can try again
      • 2018-05-21 14140, 2018

      • reosarevok
        heh
      • 2018-05-21 14141, 2018

      • reosarevok
      • 2018-05-21 14153, 2018

      • reosarevok
        church paris returns paris*h* churches
      • 2018-05-21 14106, 2018

      • samj1912
        lol XD
      • 2018-05-21 14132, 2018

      • samj1912
        that's prolly due to ngarm
      • 2018-05-21 14129, 2018

      • samj1912
        reosarevok: should I do any address boosts?
      • 2018-05-21 14138, 2018

      • reosarevok
        I'd say no, at least for now
      • 2018-05-21 14150, 2018

      • reosarevok
        Did the current search boost it?
      • 2018-05-21 14100, 2018

      • samj1912
        I dont know
      • 2018-05-21 14115, 2018

      • samj1912
        I dont think the current search does any of its boosting the way we are currently doing it
      • 2018-05-21 14132, 2018

      • reosarevok
        Oh ok
      • 2018-05-21 14144, 2018

      • samj1912
        reosarevok: done
      • 2018-05-21 14147, 2018

      • samj1912
        weirder results
      • 2018-05-21 14143, 2018

      • samj1912
        I guess I need to boost area
      • 2018-05-21 14152, 2018

      • reosarevok
        Back in 5
      • 2018-05-21 14123, 2018

      • samj1912
        yo
      • 2018-05-21 14130, 2018

      • reosarevok
        wtf :D
      • 2018-05-21 14151, 2018

      • reosarevok
        The first doesn't even have Paris?
      • 2018-05-21 14156, 2018

      • reosarevok
        (but area is Church)
      • 2018-05-21 14105, 2018

      • reosarevok
        I think area + address might be too strong
      • 2018-05-21 14107, 2018

      • samj1912
        okay
      • 2018-05-21 14137, 2018

      • samj1912
        reosarevok: try now?
      • 2018-05-21 14130, 2018

      • reosarevok
        The first is Paris Theatre... which is great, except it has nothing to do with a church :D
      • 2018-05-21 14149, 2018

      • reosarevok
        Let me try London instead, to be using name more than alias
      • 2018-05-21 14102, 2018

      • reosarevok
        hah
      • 2018-05-21 14103, 2018

      • reosarevok
      • 2018-05-21 14117, 2018

      • reosarevok
        Well top 1 looks about right :D
      • 2018-05-21 14129, 2018

      • samj1912
        lol :P
      • 2018-05-21 14155, 2018

      • reosarevok
        But I would expect St James's Church, Piccadilly, London (which even has both words in the name!) to score higher than "London Coliseum"
      • 2018-05-21 14137, 2018

      • samj1912
        well its got the same score
      • 2018-05-21 14100, 2018

      • reosarevok
        Sure, but why, if both terms match? Shouldn't that matter? :/
      • 2018-05-21 14111, 2018

      • samj1912
        hmm
      • 2018-05-21 14115, 2018

      • samj1912
        let me try and boost that
      • 2018-05-21 14106, 2018

      • samj1912
        reosarevok: the problem with that is that st. james doesnt have any alias
      • 2018-05-21 14115, 2018

      • samj1912
        other docs have aliases that are getting boosted I think
      • 2018-05-21 14142, 2018

      • reosarevok
        https://test.musicbrainz.org/place/abad3c8c-fe74-… has one, but it doesn't have London nor Church in it
      • 2018-05-21 14105, 2018

      • reosarevok
        And both have London in the address and the area
      • 2018-05-21 14138, 2018

      • samj1912
        the problem is that they are given the same score by solr
      • 2018-05-21 14142, 2018

      • samj1912
        which is weird
      • 2018-05-21 14118, 2018

      • reosarevok
        yeah
      • 2018-05-21 14124, 2018

      • reosarevok
      • 2018-05-21 14154, 2018

      • samj1912
        well, that makes sense, since its edismax
      • 2018-05-21 14114, 2018

      • samj1912
        the main problem is phrase search
      • 2018-05-21 14126, 2018

      • samj1912
        hmm
      • 2018-05-21 14122, 2018

      • ruaok
        iliekcomputers: can we log a user in for 5 minutes with no automated re-logins?