#MetaBrainz Meeting 2016-05-02
New Search Servers?
right, last week.
there was lots of office shenanigans.
the saga alone to get the water turned on… straight out of Brazil.
tuesday I flew to London and starting meeting with a pile of people.
I had one meeting that was rather interesting.
if I were not under a self imposed embargo for creating new sites, I would be creating a new site now.
one that would crowd source the collection and validation of an artist MBID <-> IBAN mapping.
wanna give money to an artist? sure, here is the IBAN, go for it.
wednesday I was on a panel and talked about MB.
it was really nice to say: how many of you have used MB? 1/3 of the hands go up.
how many of you have done a search for music on google? all the hands go up.
great, you’ve used MusicBrainz!
that got reactions.
I posted a link to the summary of the panel, which was good, but a bit drawn out.
made some changes to MBS so asterix doesn’t explode when someone requests an audiobook
investigated some issues with the CAA and missing index.json files.
@reosarevok pinged the IA about connection timeouts there, not sure what the status is
@reosarevok: Reported deeper into the IA, no more news yet, is the status finally fixed the last reported issue with the new AC editor on the beta server
submitted both schema change PRs assigned to me, just the SQL parts
the UI might be delayed a couple weeks since I had to work on other stuff
Over the last week, I apparently didn’t do much other than sitting around in IRC and on forums, went to a festival to perform, and wrote an e-mail to all our accepted GSoC students.
Support (including some grumbling by Bryan Adams’s team today!) mostly. Some talking about autoeditor elections, more on that later. I didn’t get much time to look at the open style ticket, so that’s for this week. And some talking with the IA about errors as
extracted common functionality from our Python apps into a separate package:
implemented oauth for meb
and done a bunch of other improvements there
started writing that webapp for the office without electricity
fixed mbspotify, which was broken for some reason
ported memcached wrapper to python 3
made progress on defining our needs for NewHost, still a lot of open questions, extinguished fires of different sizes and intensity, some minor Picard works, moved almost all remaining domains to Gandi, nerfed few ws abusers, initiated a new rate limiter discussion, and usual bunch of upgrades/fixes/improvements
we had a good discussion in bookbrainz meeting regarding plans for my project. Aim for this week is to set up a blog to keep track of my progress.
long term aim before the end of the community bonding period is to nail down a plan for my gamification system the other BookBrainz devs are happy with.
We are nearly there, but levels is something LeftMost takes issue with and after the meeting I somewhat agree, going forward we are currently thinking a title system may be an acceptable compromise that doesn’t make users feel anothers contribution is worth more than their own.
finished college work
agenda for coming week:
Complete the PR stuff, its been a while i worked on them. They have been open for a while now.
Discuss the project idea with
@Rob, and plan things before coding period starts
discussed project with
@gentlecat and received some good advices on how to improve te proposal started to look into the jira system and identified somo issues that are related to datasets in AcousticBrainz
working with the AcousticBrainz server and started creating my development environment
Hi, I’m Goran from Croatia but people call me cetko (tsetko). I’ll be working on AcousticBrainz BigQuery data exploration this summer with alastairp as mentor. I’m in the middle of my midterm exams this week so my focus has mostly been on surviving that. Once that’s done I’ll discuss the account setup with alastair for BigQuery to play around and start working on the db schema design.
finished off with my exams last week
this week i plan to get
AB-63 merged, after which we will have a nice page on acousticbrainz which will show a list of all public datasets also finalize everything about the project with
Lots of work on the underpinnings of the BB site code this week. A few wins as far as user experience, but it will also make it much easier to improve things going forward. Going to try to release those changes in the next couple days.
I mostly procrastinated on my schema change ticket, as I’m finding it a bit frustrating to find all the places that barcodes are pulled in, but I think I just need to make another run through and fix up tests and the WS and then it’s review time.
not much wrt AB server this week, I reviewed a few of Gentlecat’s patches, will do more later this week
now that we have SoC students, there’s a bit of planning of things to do
some people saw filipe around last week, he’s a developer advocate at google and was in Barcelona, so he and ruaok and Gentlecat hung out for a few hours
basically, it took us about 20 minutes to import all of acousticbrainz into google bigquery
and now we can do queries like this: http://demo.redash.io/embed/query/1571/visualization/2103?api_key=58158fa71d00028f2e5d1843a23986d47d367591
which is key vs release year
we also got some really cool correlation plots between things like loudness and bpm grouped by genre
takes about 4 seconds to do a query, but in the current way we have it set up, eats through about 80GB of data processing it
first job for @gcetusic is to write a schema which will get that down to <50MB for a similar query SSO
I don’t have much concrete to say about it at this point, but I wanted to direct people to
Single sign-on brainstorm (or, SSO SOS) for some brainstorming on what it should look like.
I suspect it will be schema change-y, so I’d like to gather a bunch of feedback and then I might start working on a prototype to see what’s feasible and what isn’t. If anyone else wants to help out, that’d be much appreciated. New search servers
@weeksio dropped out of this year GSoC, @Zas wanted to know what our plan was regarding the new servers.
@Rob stated that he “might just need to bit the bullet and take on this project myself.” @Bitmap stated that once MBS-8736 had been submitted, he’d have more capacity to help out with search server issues as well. @Rob affirmed that the new SOLR search server needs to be up for our move to NewHost. Autoeditor elections
@reosarevok and @Freso talked a bit over the weekend about the current state of affairs with regards to autoeditor elections and came up with a temporary workflow until it can get automated again.
This was approved by the meeting and is explained in more details in this blog post: