#musicbrainz-devel

/

      • rembo10
        haha
      • 2012-04-20 11142, 2012

      • reosarevok
        I guess the next schema will be DSNS anyway, for Deep Space Nine? :p
      • 2012-04-20 11104, 2012

      • ianmcorvidae
        MusicBrainz: We Get Music, But C'Mon, Sequential Numbers Are Hard, Give Us A Break
      • 2012-04-20 11107, 2012

      • ianmcorvidae
        ;)
      • 2012-04-20 11114, 2012

      • kepstin-netbook joined the channel
      • 2012-04-20 11128, 2012

      • ghuntley joined the channel
      • 2012-04-20 11148, 2012

      • ianmcorvidae
        clearly we should identify our webservice iterations with UUIDs
      • 2012-04-20 11151, 2012

      • ianmcorvidae
        ;)
      • 2012-04-20 11100, 2012

      • ocharles
        phew, just about ready to start doing my own work
      • 2012-04-20 11106, 2012

      • ocharles
        darn contributors!
      • 2012-04-20 11107, 2012

      • ocharles
        :)
      • 2012-04-20 11114, 2012

      • ianmcorvidae
        pesky ;)
      • 2012-04-20 11135, 2012

      • ocharles
        ianmcorvidae: I'll have a stab at your email today, but leaving it til I'm not feeling creative enough to code
      • 2012-04-20 11156, 2012

      • ianmcorvidae
        ocharles: cool, works for me :)
      • 2012-04-20 11110, 2012

      • ocharles
        not sure what we'll do about the letter head, but I plan to just get a draft to you today
      • 2012-04-20 11115, 2012

      • ijabz
        warp hi
      • 2012-04-20 11147, 2012

      • ianmcorvidae
        ocharles: I'll just make something up, unless ruaok has metabrainz letterhead for fancy business customers or something
      • 2012-04-20 11113, 2012

      • ocharles
        he probably does
      • 2012-04-20 11118, 2012

      • ocharles
        Does anyone have some time to talk about the 'primary' attribute for aliases?
      • 2012-04-20 11157, 2012

      • ianmcorvidae
        ocharles: not sure what you mean, but given a link or explanation I can discuss
      • 2012-04-20 11125, 2012

      • reosarevok
        ocharles: guess so
      • 2012-04-20 11142, 2012

      • reosarevok
        Since I don't have the keys for this house and thus I can't go buy myself some chocolate :(
      • 2012-04-20 11157, 2012

      • ocharles
        http://tickets.musicbrainz.org/browse/MBS-3788 talks about a 'primary' attribute for aliases
      • 2012-04-20 11159, 2012

      • reosarevok
        Yes
      • 2012-04-20 11106, 2012

      • ocharles
        I was wondering how we want to display this, and he want to edit it
      • 2012-04-20 11120, 2012

      • ocharles
        do we literally display a "Primary for locale" column, with a checkbox in or something?
      • 2012-04-20 11129, 2012

      • ocharles
        Do we represent it as a checkbox on the edit alias page?
      • 2012-04-20 11143, 2012

      • reosarevok
        For editing, a checkbox seems ok
      • 2012-04-20 11110, 2012

      • reosarevok
        For displaying... I guess some mark could do?
      • 2012-04-20 11125, 2012

      • reosarevok
        a bold (P) after the locale or something
      • 2012-04-20 11148, 2012

      • reosarevok
        (a checkbox in the edit alias page I mean)
      • 2012-04-20 11108, 2012

      • ocharles
        i'd rather explicitly say (primary) I think
      • 2012-04-20 11116, 2012

      • ocharles
        but primary is such a vague term
      • 2012-04-20 11120, 2012

      • MBJenkins
        Project musicbrainz-server_master build #169: FAILURE in 10 min: http://ci.musicbrainz.org/job/musicbrainz-server_…
      • 2012-04-20 11121, 2012

      • MBJenkins
        * oliver.g.charles: Make the default max-height of annotations 100px, not 400
      • 2012-04-20 11122, 2012

      • MBJenkins
        * pavan: MBS-4545: Add support for google analytics
      • 2012-04-20 11123, 2012

      • MBJenkins
        * pavan: MBS-4545: Explain analytics config option
      • 2012-04-20 11141, 2012

      • ocharles
        sigh, MORE dbdefs changes
      • 2012-04-20 11120, 2012

      • nikki
        ocharles: btw, there's a weird blank line above the annotation
      • 2012-04-20 11133, 2012

      • nikki
        and I still disagree with making annotations so small
      • 2012-04-20 11141, 2012

      • ianmcorvidae
        nikki: it's a margin-top: 1em on '.annotation', I suspect
      • 2012-04-20 11154, 2012

      • ianmcorvidae
        ocharles: do we have a ticket for "make our config system less fiddly"?
      • 2012-04-20 11155, 2012

      • reosarevok doesn't mind the size, but the margin top seems unnecessarily big
      • 2012-04-20 11112, 2012

      • ocharles
        ianmcorvidae: I think so
      • 2012-04-20 11118, 2012

      • reosarevok
        (the size of the actual annotation is what I don't mind, clearly I mind the margin)
      • 2012-04-20 11118, 2012

      • ocharles
        i have plans, lack time to do the work
      • 2012-04-20 11150, 2012

      • reosarevok
        Maybe half an em would do?
      • 2012-04-20 11151, 2012

      • ianmcorvidae
        ocharles: don't we all -- just wanted to make sure it's in jira, heh
      • 2012-04-20 11112, 2012

      • ocharles
        i'm not sure it is, and even if it is, I should explain my thoughts so someone else can try and do it
      • 2012-04-20 11117, 2012

      • nikki
        I don't know of a ticket and searching for dbdefs and config doesn't find anything
      • 2012-04-20 11135, 2012

      • ocharles
        MBS-4578 created
      • 2012-04-20 11157, 2012

      • ocharles
        beta fixed (again)
      • 2012-04-20 11130, 2012

      • voiceinsideyou joined the channel
      • 2012-04-20 11152, 2012

      • rembo10 joined the channel
      • 2012-04-20 11127, 2012

      • ocharles
        i'm not sure why that margin is there either
      • 2012-04-20 11125, 2012

      • ianmcorvidae
        bde201dc root/static/styles/entity.css (Pavan Chander 2010-06-02 14:51:15 -0400 87) margin-top: 1em;
      • 2012-04-20 11105, 2012

      • ianmcorvidae
        spring-cleaning of CSS, seeing if it's sourced from somewhere older in that commit
      • 2012-04-20 11105, 2012

      • ocharles
        and the log message?
      • 2012-04-20 11109, 2012

      • ocharles
        hum
      • 2012-04-20 11142, 2012

      • ianmcorvidae
        "Even more CSS spring cleaning\nOur layout.css was getting a tad unwieldly so I split it up. The deleted image and CSS files were all unused"
      • 2012-04-20 11158, 2012

      • ocharles
        If we have one English locale, should it be primary be default? Or is it ok to have English locale aliases, and no "primary" one?
      • 2012-04-20 11132, 2012

      • ianmcorvidae
        (that commit was moving it from layout.css to entity.css: 08d3aba9 root/static/styles/layout.css (Lukáš Lalinský 2009-06-02 15:11:55 +0000 499) margin-top: 1em; previous source, checking that now)
      • 2012-04-20 11150, 2012

      • ianmcorvidae
        "Merge in the first version of the new design"
      • 2012-04-20 11109, 2012

      • ianmcorvidae
        I think that space has been there essentially all of NGS, wonder why we're just noticing now
      • 2012-04-20 11114, 2012

      • reosarevok
        ocharles: you mean when we convert the current ones?
      • 2012-04-20 11137, 2012

      • reosarevok
        Or in general?
      • 2012-04-20 11129, 2012

      • ocharles
        reosarevok: both
      • 2012-04-20 11101, 2012

      • reosarevok
        For converting I'd say yes, since the current guides for it suggest it's to be used for the primary name in that locale
      • 2012-04-20 11108, 2012

      • navap
        ianmcorvidae: On the main server the margin-top for .annotation and for the p overlap and there's a net of 1em, which looks fine. On test they don't overlap and the net is 2em, which doesn't look fine.
      • 2012-04-20 11116, 2012

      • ocharles
        it's probably going to get weird if we have a checkbox that can be unchecked, but also to have a requirement that *something* must be checked
      • 2012-04-20 11138, 2012

      • ianmcorvidae
        navap: what changed, then?
      • 2012-04-20 11101, 2012

      • nikki
        ocharles: for the conversion, I said somewhere that all current ones should be set to primary, since that's what it currently means
      • 2012-04-20 11122, 2012

      • ocharles
        ah, so you did!
      • 2012-04-20 11129, 2012

      • ocharles
        I better read your comment again and see if I've missed anything else :)
      • 2012-04-20 11102, 2012

      • ocharles
        "Search hints should not have dates, sort names, locales or the primary flag." that makes me want to add a artist_search_hint table
      • 2012-04-20 11144, 2012

      • navap
        ianmcorvidae: I think its because overflow:auto was set on .annotation-body
      • 2012-04-20 11101, 2012

      • ianmcorvidae
        hm
      • 2012-04-20 11109, 2012

      • ianmcorvidae
        well, in any case, that margin is huge and it should be changed :P
      • 2012-04-20 11116, 2012

      • MBJenkins
        Project musicbrainz-server_master build #170: STILL FAILING in 15 min: http://ci.musicbrainz.org/job/musicbrainz-server_…
      • 2012-04-20 11137, 2012

      • navap
        Yeah, the .annotation style can just go
      • 2012-04-20 11150, 2012

      • ocharles
        oops, forgot to update tests with new annotation stuff
      • 2012-04-20 11133, 2012

      • navap thinks about going back to bed
      • 2012-04-20 11154, 2012

      • djce joined the channel
      • 2012-04-20 11114, 2012

      • kurtjx joined the channel
      • 2012-04-20 11101, 2012

      • ocharles
      • 2012-04-20 11120, 2012

      • reosarevok
        ocharles: guess so, but let's ask nikki
      • 2012-04-20 11128, 2012

      • reosarevok
        nikki, does that look OK?
      • 2012-04-20 11142, 2012

      • ocharles
        I think I'm going to group aliases by type in the same way we group releases in a release group
      • 2012-04-20 11155, 2012

      • ocharles
      • 2012-04-20 11105, 2012

      • reosarevok
        It'd make sense to me
      • 2012-04-20 11114, 2012

      • reosarevok
        At least I'd want to group all the search hints at the bottom
      • 2012-04-20 11125, 2012

      • reosarevok
        So I guess this is a good way of doing so
      • 2012-04-20 11153, 2012

      • ocharles
        unknown will go at the bottom atm
      • 2012-04-20 11125, 2012

      • reosarevok
        Well, yeah, I meant after all the ones selected as something else
      • 2012-04-20 11139, 2012

      • reosarevok
        Unknown at the bottom is consistent with our discography display, I'm fine with it
      • 2012-04-20 11139, 2012

      • ocharles nods
      • 2012-04-20 11143, 2012

      • ocharles
        cool
      • 2012-04-20 11105, 2012

      • ocharles
        meh, I can't decide if search hints should be aliases with the invariant sort_name=name, begin_date=null, end_date=null, type=search_hint, or if i stick it in a new table
      • 2012-04-20 11114, 2012

      • reosarevok
        ocharles, nikki: how do unknowns work? Are they allowed to have all that?
      • 2012-04-20 11136, 2012

      • rembo10 joined the channel
      • 2012-04-20 11101, 2012

      • nikki
        yes
      • 2012-04-20 11149, 2012

      • reosarevok
        Hmm
      • 2012-04-20 11153, 2012

      • reosarevok
        Does that make sense? :/
      • 2012-04-20 11112, 2012

      • nikki
        yes
      • 2012-04-20 11135, 2012

      • nikki
        maybe you're not sure what type it is. maybe it's something you feel doesn't fit under the existing types
      • 2012-04-20 11143, 2012

      • nikki
        primary is probably the only one which might not make sense, because having a name as the primary unknown type of name is a bit... odd. you wouldn't know when you could use it
      • 2012-04-20 11147, 2012

      • ocharles
        hmm, grouping by type makes primary a little weird
      • 2012-04-20 11105, 2012

      • ocharles
        you can only have one primary alias per *locale*, but the grouping is per *type*
      • 2012-04-20 11122, 2012

      • ocharles
        that mismatch seems confusing
      • 2012-04-20 11155, 2012

      • nikki
        one per locale per type probably makes more sense
      • 2012-04-20 11100, 2012

      • ocharles
        yea
      • 2012-04-20 11107, 2012

      • ocharles
        shall we go with that?
      • 2012-04-20 11147, 2012

      • nikki
        and as long as it doesn't result in multiple matches when we want to use the data to display localised names, it should be ok
      • 2012-04-20 11149, 2012

      • nikki
        sure
      • 2012-04-20 11158, 2012

      • ocharles
        multiple matches in what sense?
      • 2012-04-20 11106, 2012

      • reosarevok
        nikki: how wouldn't it?
      • 2012-04-20 11108, 2012

      • ocharles
        presumably for localised names, we'll be selecting from a specific type?
      • 2012-04-20 11121, 2012

      • nikki
        yes, for artists we'd select the type 'artist name'
      • 2012-04-20 11130, 2012

      • reosarevok
        But can't that be legal or performance?
      • 2012-04-20 11133, 2012

      • ocharles
        ok, lunch now
      • 2012-04-20 11141, 2012

      • reosarevok
        Or if a legal name is also a performance name you'd add it twice?
      • 2012-04-20 11149, 2012

      • nikki
        hmm.
      • 2012-04-20 11114, 2012

      • nikki
        I wouldn't have a problem with people doing that if they wanted
      • 2012-04-20 11105, 2012

      • warp
        hello!
      • 2012-04-20 11110, 2012

      • reosarevok
        Hi warp
      • 2012-04-20 11108, 2012

      • reosarevok
        Dunno, I would expect each locale to have just one primary. and usually all the primaries would be in the same type, but hey, if there's a case where the artist is known by a performance name only in one locale... that also works
      • 2012-04-20 11140, 2012

      • nikki
        given that we're starting off with just "entity name" and "search hint", I guess either approach works
      • 2012-04-20 11159, 2012

      • nikki
        we could stick with just one primary name regardless of type for now and see what happens later
      • 2012-04-20 11121, 2012

      • reosarevok
        Oh
      • 2012-04-20 11123, 2012

      • reosarevok
        Are we?
      • 2012-04-20 11127, 2012

      • reosarevok
        I missed that discussion :)
      • 2012-04-20 11135, 2012

      • reosarevok
        Fine then
      • 2012-04-20 11114, 2012

      • kurtjx joined the channel
      • 2012-04-20 11136, 2012

      • acerHDMI joined the channel
      • 2012-04-20 11154, 2012

      • acerHDMI
        what am im doing wrong
      • 2012-04-20 11158, 2012

      • acerHDMI
        chown: invalid user: `/var/log/musicbrainz'
      • 2012-04-20 11139, 2012

      • rembo10
        sudo chown musicbrainz /var/{log,run}/musicbrainz
      • 2012-04-20 11140, 2012

      • rembo10
        :-)
      • 2012-04-20 11124, 2012

      • acerHDMI
        its ur code
      • 2012-04-20 11151, 2012

      • rembo10
        haha typo, sorry :-)
      • 2012-04-20 11145, 2012

      • acerHDMI
        guess the same for this one too sudo chown /usr/bin/mbcontrol
      • 2012-04-20 11100, 2012

      • rembo10
        yes
      • 2012-04-20 11102, 2012

      • cduez has left the channel
      • 2012-04-20 11115, 2012

      • ocharles
        nikki, reoafk, so is there a decision?
      • 2012-04-20 11122, 2012

      • ocharles
        is primary unique over locale, or type and locale?
      • 2012-04-20 11107, 2012

      • reoafk would rather have locale, but I'm willing to accept type and locale if people think it's better - it just seems harder to use later for its clearer use, displaying localised artist names
      • 2012-04-20 11122, 2012

      • nikki
        it seems like if we go with locale now, it's still possible to switch to type+locale later. the other way round isn't easily doable
      • 2012-04-20 11129, 2012

      • ocharles
        yep
      • 2012-04-20 11135, 2012

      • ocharles
        but i think my grouping idea might not be very good
      • 2012-04-20 11104, 2012

      • ocharles
        i might just revert that