#musicbrainz

/

      • KingJ joined the channel
      • hibiscuskazeneko has quit
      • xplt
        Xof: Freso is right
      • Freso
        (I mean, it'll probably still be better to make that edit go through and then fix the release country.)
      • xplt
        [Worldwide] are usually digital releases
      • hibiscuskazeneko joined the channel
      • arbenina_ joined the channel
      • timbl joined the channel
      • timbl_ joined the channel
      • Xof
        you guys and your implicit ontologies
      • timbl__ joined the channel
      • timbl___ joined the channel
      • there ought to be a rule among people designing interfaces for knowledge representation, something like graceful degradation... don't make me understand your entire world model before I can make plausible contributions :-)
      • timbl_ has quit
      • timbl has quit
      • will happily fix to something more plausible or stylish
      • (presumably UK, since that glorious land is where Hyperion is based)
      • SothoTalKer
        it's not a big issue
      • Xof
        if it's not a big issue, don't make me spend two elapsed days on it
      • SothoTalKer
        it's already in 3:0 state
      • Xof
        well it is now :-)
      • hibiscuskazeneko
        There are 2 Abstain votes, though
      • I abstained because the other changes you're making in that edit are too important to lose
      • Xof
        I mean, look, I'm as guilty as anyone of trying to represent the world in my projects
      • hibiscuskazeneko
        It would be easier to let that edit pass and then change the release country
      • Xof
        and I understand that quality control is important
      • xplt
        Xof: Errr... no, everything is fine. IIRC, the edit will pass within next hours (Abstain votes don't matter)
      • SothoTalKer
        i learned to be patient
      • xplt
        Xof: A few years ago, you'd have to wait 2 weeks %)
      • Xof
        but as a moderately-knowledgable time-poor occasional contributor, basically every interaction goes like this: I submit data, it's a bit wrong according to this year's style guides, and I get to wait seven days (or less if I act arsey on IRC) to retag my stuff
      • xplt: do you know who did the data analysis to motivate the change from 2 weeks?
      • xplt
        No
      • Xof
        I did
      • xplt
        :)
      • Xof
        I produced evidence to show that almost all votes came either in the first or last day of the open period
      • I wonder where those graphs are
      • on some dust-accumulating laptop somewhere
      • this is how I remember it, anyway :-)
      • SothoTalKer
        at least you got votes :p
      • xplt
        On the other hand, on comics.org [open database about comic books] ALL the edits are going through peer-review :P [so, it could be worse...]
      • Leftmost
        If you have suggestions for improving the way MB displays its models, etc. to users, _please_ submit them. The complexity of editing is a known issue, and people are working on it, but it's also a hard problem.
      • SothoTalKer
        np-hard?
      • Xof
        don't get me started on how long I had to argue that "contratenor" was not what the style guide thought it was
      • Leftmost: and believe me, I know it's a hard problem.
      • I have sympathy, even if it's somewhat frustrated sympathy :)
      • arbenina_ has quit
      • the main thing that annoys *me* is that the editing interface allows you to make edits that will subsequently be found to be wrong based on statically knowable stuff
      • like: it's a physical release, so no "worldwide"
      • or no "purchase for download"
      • I'd be happy if at the "submit edits" stage those consistency checks were carried out if that meant a shorter open period or lower voting threshold or just generally less round-tripping
      • and if there were a way of capturing those round-trips in comments and distilling them to more consistency checks, great.
      • hibiscuskazeneko
      • BrainzBot
        MBS-9499: Disable "purchase for download" URL relationship for non-Digital Media releases
      • SothoTalKer
        there is nothing that speaks against a physical release being worldwide
      • xplt
        And if "English" is chosen, then select "Latin" script... :P
      • SothoTalKer
        its just a very very unlikely case for older releases
      • hibiscuskazeneko
        This.
      • There was some discussion about it a while back (I don't know if I can still find it; it might have been on the old forums for all I know)
      • Xof
        SothoTalKer: so, make that a warning at the submit edit stage, requiring confirmation
      • or use the date of the release event as a filter
      • always allow it to be overridden, but perform as many reasonable checks as possible before submit
      • (unless it's an ontology violation like MBS-9499 in which case disallow it completely)
      • BrainzBot
        MBS-9499: Disable "purchase for download" URL relationship for non-Digital Media releases https://tickets.metabrainz.org/browse/MBS-9499
      • SothoTalKer
        if you use the release editor, release event comes before you enter the type of the release.
      • so there is not a good way to check for that :)
      • Xof
        at the end it shows you all your edits, batched up
      • hibiscuskazeneko
        True, but that doesn't mean there can't be a prompt
      • Xof
        right before you type your edit note and submit
      • that's the point for all these consistency checks, failing a more integrated one
      • SothoTalKer
        yep, let's make every edit require an additional prompt.
      • Xof
        no, only those with ontology-violating edits?
      • SothoTalKer
        "are you sure blabla". and then another, just to be extra safe. "are you really really sure?"
      • Xof
        the highly-competent case, with a user who has fully absorbed the musicbrainz ontology and has made no edits with surprising content: no additional prompt (hooray! good!)
      • the occasional user case, with someone who is a bit out of date but would like to get on with tagging their stuff: prompts in likely cases of mistakes, allowing them to fix them directly: shorter open edit period (hooray! good!)
      • SothoTalKer
        what made you assume, it's a worldwide release in the first place? just because some people doubt, it is, does not mean they are right :)
      • Xof
        finally, the downside case: the highly competent user with an unusual worldwide jewel case release. Oh no! one extra click
      • so this is an engineering question: how often does the downside case happen compared to the occasional user case?
      • and what are the relative costs
      • SothoTalKer: well I am a guest here; my ontology of what a worldwide release means doesn't necessarily match the MB community's and I'm happy, as a guest, to go with my host's interpretation
      • SothoTalKer
        you could just answer my question, instead :)
      • Xof
        Western parochialism? You could get it in the shops here so it was probably available everywhere?
      • rose-tinted sunglasses?
      • the sleeve notes don't say "not for sale in the USA"
      • SothoTalKer
        even digital releases are not "worldwide", since north korea is excluded usually :)
      • Xof
        the point is that I don't want to make a strong claim about the release country. Definitely the UK, almost definitely Europe, probably some other places too; worldwide for me is the weakest thing I can say.
      • But that is apparently not how worldwide is interpreted here, which is what is important
      • the worldwide statement is a strong one ("it is defintely released worldwide") rather than a weak one ("it's released in more than one country vaguely around this time")
      • it's not that my assumption is wrong; it's that what I said in my edit doesn't mean to musicbrainz what I thought it meant
      • mea maxima culpa. [Latin, Latin]
      • SothoTalKer
        nobody blames you. it's nothing that can't be fixed later.
      • Freso
        Xof: You may be interested in https://youtu.be/MQ8wDsRkKeE?t=26m28s
      • Our UX talk/discussion at the summit this very Sunday.
      • Xof
        :-) I'm not feeling blamed, either
      • SothoTalKer: it's somewhat refreshing being the clueless user for a change
      • xplt
        Xof: And your edit got applied. Hooray!
      • SothoTalKer
        you only make this error once, then you know the drill ;)
      • Xof
        I guess I'm trying to give the benefit of (gulp) too many years' experience trying to manage contributions to complex stuff: make it easier for users not to get it wrong
      • xplt: also, yay
      • SothoTalKer: there is that, though the danger is that I come back in a year's time and the guidance / ontology / style has changed and I make a related but different error then, too
      • SothoTalKer
        Freso: is it possible to get captions? :)
      • Freso
        SothoTalKer: Are you writing them?
      • SothoTalKer
        nope
      • Freso
        That's your answer right there then.
      • SothoTalKer
        such meanieness.
      • Xof
        xplt: of course I have spent all my musicbrainz time budget this evening arguing on IRC, so... :)
      • reosarevok
        Xof: that's a very good point, thanks. Quesito ^ for the UX talk
      • (quite a long read but ;) )
      • Xof
        howdy reosarevok
      • it's been a while since I hung out here
      • reosarevok
        Xof: I don't think we can make everything autoedits, tempting as that would be, until we can revert things more easily, so the few days issue will still be an issue
      • Xof
        I am now nostalgically trying to find my evidence that the 14-day voting period was too long. How can it have disappeared from the Internet?
      • reosarevok
        But the giving more prompts is definitely something I agree with
      • Freso
        ^
      • Xof
        I can live with a voting period if I don't have to argue with anyone ;-)
      • SothoTalKer
        i suggested to make votes more prominent already :)
      • Freso
        Xof: FWIW, I think one of the reasons that "attributes for all entities" is being held up is exactly that yvanzo is implementing safe guards to not allow e.g. a release "rpm" attribute being applied to a "digital media" or "cd" release.
      • SothoTalKer: You and dozens of other people in at least the last 10 years...
      • SothoTalKer
        now, get programming, Freso :)
      • make us (the community) happy :3
      • Xof
      • BrainzBot
        STYLE-55: "Contra-tenor" in vocal lead performance
      • Xof
        hey look this one is me
      • Freso
        SothoTalKer: I'm not employed as a programmer.
      • SothoTalKer
        i'm not employed as an editor and i edit anyway :)
      • Freso
        SothoTalKer: But I will happily tell you, being a community member, where and how to provide your feedback to developers and even help you get started making PRs yourself.
      • chhavi_ joined the channel
      • Xof
        aha!
      • BrainzBot
        MBS-5757: Consider diminishing time edits stay open
      • Xof
        ok, nostalgia done
      • thanks for the therapy everyone
      • SothoTalKer
        14 days was a bit long, yes
      • Freso
        Xof: Cheers.
      • SothoTalKer
        Freso: maybe you could enable this for the MetaBrainz yt channel? https://support.google.com/youtube/answer/6052538
      • Freso
        SothoTalKer: Absolutely! Thanks.
      • Done.
      • SothoTalKer
        yay :)
      • kyan joined the channel
      • hibiscuskazeneko has quit
      • timbl___ has quit
      • timbl__ has quit
      • timbl joined the channel
      • timbl_ joined the channel
      • drsaunders joined the channel
      • timbl_ has quit
      • timbl has quit
      • Jormangeud has quit
      • colbydray has quit
      • loujine
      • reosarevok
        loujine: this is fantastic