so you get to se the pattern of how old they are contra how old they where whne they joined
Quesito
Leo_Verto: do we wanna do some graphs perhaps for a "presentation of results" or such?
CatQuest
yes!
Leo_Verto
I know, I'd love to do that, but users responded to the survey under the impression that I wouldn't cross-reference the demographic data with anything else (e.g. to find out the MB usernames or email addresses of industry people)
yes
definitely
CatQuest
those are cool
Leo_Verto
graphs all the way
and a map
Quesito
maybe close down the survey a few weeks ahead of Oct 1st then....
CatQuest
Leo_Verto: but isn't "demographics" also the sort of things like "members by age, gender(orwhatever) shoesize, time joined" etc?
maybe i understnad the word wrong..
Leo_Verto
the language and country questions will need some cleanup but I'm really looking forward to the map of users worldwide
CatQuest
:O
that soudns really cool
Quesito
me too Leo_Verto!!!
sound super cool
CatQuest
oh so not "avarage age by continent" is not ok?
Quesito working on artists worldwide right now....
Leo_Verto
CatQuest: the "for how long have you been using MB"-question is not part of the demo data
CatQuest
aww, but I love silly things liek that :O
Leo_Verto
"average age by continent" is fine
CatQuest
Leo_Verto: it wasn't???
oh.
i thoguht it was
Leo_Verto
but I'm going to seperate the demo data from the rest and scramble the order of both
CatQuest
no wai, I remember explicitly telling for how long
Leo_Verto
well yeah, it's part of the first section
CatQuest
uhh.. ok i don't understand and probably it's becasue attentino or english
Leo_Verto
hmm, let me try to explain it
we've got the general MB questions and the demo questions
in the survey, I promised to process and analyze the demo data seperately from the rest
CatQuest
yea the test survey
wait? that was to be included?
Leo_Verto
with the idea that people may be more open if they know that their personal data isn't referenced to their other answers, including their username or email
so what I'm going to do is split them into two datasets, one of general questions and for demo
CatQuest
but the demo was jsut for testing???
Leo_Verto
both of those are going to be released (without usernames and emails of course) and I'll analyze both in seperate blog posts, using only the data that is publicly released
and that means I can't cross-reference them or people would be able to figure out which demo data entry is associated to which general data entry
CatQuest
but what if someone added bogus info i nthe test version+ I mena i basically copy pasted my answers
Leo_Verto
oh, I'm not really talking about that right now
the test version answers are not going to be included
CatQuest
uh.. so what is "demo"
Leo_Verto
I've got the response date as part of the data after all
demographics data
CatQuest
oh
uh bit as i took it i was unaware that some parts were "demograpic" and others wher "nope"
anyway I am not suggestings 1-1 cross refrencials. I am talking about a grap that shows the general trend of "how old" vs "how long in mb"
those shouldn't be able to refrence to things lie kusernames
nor should you be able to pinpoint that *this* 50 year old is *this* long in mb* is *this spesifc person with other answers*
it's bsically to answer this question i was wondering: the majority of our 50 or ouler, did they start mb when 50 or older or wher ethey 40 or older at the time :D
Leo_Verto
that'd require at least some fuzzying of the data or outliers (like the one 81-year-old) would still be indetifiable
CatQuest
hmm..
well aren't most grpahs a it fuzzy?
:D
Quesito
now we get into data ethics....:D
CatQuest
but ok, I think I get your objectino now
i was very confused :)
Quesito
Leo_Verto: I'll keep trying with the reminders on twitter and such...I go on a bit of holiday till after you return...so maybe we formulate a plan for the closure timing and graph work first week in September?
Leo_Verto
sure
Quesito
right on
saifukbkhan joined the channel
github joined the channel
github
[listenbrainz-server] paramsingh opened pull request #243: Make info view accessible only for logged in current user (production...login-required) https://git.io/v7237
github has left the channel
anshuman73 joined the channel
anshuman73 has quit
anshuman73 joined the channel
anshuman73 has quit
anshuman73 joined the channel
Mineo joined the channel
github joined the channel
[sir] gentlecat opened pull request #24: Additional fixes for the wscompat.convert module (master...converter) https://git.io/v72lz
github has left the channel
agentsim has quit
dseomn has quit
agentsim joined the channel
dseomn joined the channel
dseomn has quit
dseomn joined the channel
github joined the channel
[sir] gentlecat opened pull request #25: Capture log messages with Sentry (master...logging) https://git.io/v72Bg
I'm trying to see if there are any issues with the implementation
have some issues with release updates
with release-group core and `releases` path I hit an error because I get a PK `release_group` for this selection: `SELECT release_group.id FROM musicbrainz.release_group WHERE release_group.id IN (:ids)`
so I don't have a reference to pick a PK from the original message `{"_table": "release", "id": 1978055}`
CatQuest
Zastai: a screensgort of wh you mean would be ace becasue that sounds vey weird
Zastai
I created a ticket for it. Basically the tracklist is rendered in grey with the stippled underline and the "deleted, cannot display properly" tooltip