#musicbrainz-devel

How can I contribute to the GEORDI project and what pre-requisite knowledge should I have ?

hmm https://github.com/tfrce/thedaywefightback.js

something we should do?

it'd need hotfixing to make it to production before the 11th, but possibly?

it's slightly less interesting for non-US people too, of course

yeah.

geoip it in optionally?

It's very much pointless for anyone non-US, yeah

people are going to ask about this and we ought to have a plan or an answer as to why not...

Since your laws seem to mean "well whatever they're foreigners and have no rights" anyway :p

I mean, one of the things they're asking for *is* protections for non-US citizens, but :P

Sure, but they'd ignore non-US citizens asking for it :p

And "why not" would be "because it involves annoying users with a huge-ass banner for no real gain"

heh, yeah

But if it's only one day, I guess it doesn't hurt much

at least in germany there's been a lot of buzz about the whole surveillance thing

Mineo: still not enough.

Here I haven't seen much - but of course it's ex-USSR, this surveillance is soft for them :p

heh

even politicians were like 'OMG WTF BBQ' after it became public the nsa has been spying on our last two chancellors :P

at least the us' ambassador to the ukraine is helping keep EU <-> US relations icy. :)

What surprises me is that people are surprised?

Especially for the politician thing

anyway, I'm not opposed to putting it on, I don't think it'd be worth excluding the non-US people because there's not much sense devoting dev time to it

I mean, I'm fairly sure every spy service spies every other politician in the world? :p

(if they have enough money for it)

I mean, all the NSA stuff is pretty "well, yes?" to people who know things, but that's not really the point here so much as trying to help make things better :P

Heh

it was never "well, yes, and that seems totally fine" :P

Yeah, I meant the way most of the scandal seemed to be "omg and they're spying on our politicians" which is, you know, the only reasonable bit of the spying :p

(and the one those spied countries are almost certainly doing too, although without getting caught yet)

But anyway, even though it seems laughable to do the whole "call your politicians" as if we were talking of a democracy where people gave a shit about the voters' opinions, I imagine it can't hurt

heh, maybe it's an NSA plot to collect disidents' info! ;p

haha

yeah, I mean, they don't need to have a backdoor in the script for tat :P

that*

"Log in with Facebook to mail your politicians!"

(it would have been awesome)

lol

reosarevok has changed the topic to: Chicago week | Agenda: (empty), DR: MBS-3354, MBS-2518, MBS-4724, MBS-3537

ianmcorvidae has changed the topic to: Equatorial week (for a tiny bit longer) | Agenda: (empty), DR: MBS-3354, MBS-2518, MBS-4724, MBS-3537

ianmcorvidae: "Move MusicBrainz Search to SOLR" - Do you have some opinion about it?

of what sort? I think it'd be a good thing to do so we (as the project) don't have to maintain as much custom code

and ideally reworking would allow it to be live-updated instead of our current 3-hour delay, which would be a nice change

ianmcorvidae: And technical opinions? Should it one solr collection per entity or compined all to one collection? And should we have compatibility API for current musicbrainz xml search?

we would presumably want both per-entity indexes and an overall index

for searches where the entity type is known and where it isn't, respectively

the API coming out of the perl codebase would need to be compatible, of course, but I don't have strong opinions about the exact means of doing that

a combined index might work fine given that the entity type could be a field of the index anyway, of course

"

"luks: inspired your mbdata/mbslave solr things I created musicbrainz core to my solr and used solr's DataImportHandler for importing and solr's velocity for browsing... works pretty well (I also used only one index..)" https://chatlogs.musicbrainz.org/musicbrainz-de...

ianmcorvidae: Thanks.

i just saw the 2014 ideas list for GSoC, and i'd be interested in creating a website for picard.

i wolud like to know what exactly are the requirements?

anyone there?

Jozo: how does DataImportHandler deal with relationships and updates on them?

dufferzafar: there isn't really a set of requirements for it

the mbdata-based indexer I did two weeks ago is actually pretty nice in that regard

and I can generate update triggers without having to redefine the schema once again, thanks to the sqlalchemy models

nikki: The wiki page doesn't list any mentor either.

luks: Not figured that yet. After november I have not done anything with solr... I just written some queries that rely last_updated.

the mentors are just copied from last year's list, I think

Jozo: last_updated is not recursively updated, so that doesn't catch things like updating of an artist name on recordings

The wiki actually lists two sites, one for picard and the other one for its plugins.

luks: I know. But whit it it can update new entities pretty instantly.

do you know of someone i could talk to about them.

oh, the solr thing is mentioned on the gsoc page

it's not supposed to mean two sites, it's talking about if picard is a separate site, that site can be used for the plugins

well, whoever is doing that should take a look at my code

luks: That's why I asked more details. (And I'm not student...)

luks: yeah, your code is roughly the starting point (I think it's part of why rob listed python for the project)

and I'm not sure who to ask, perhaps try emailing the devel list?

I honestly have no idea of how GSoC works. I am in interested in this porject. Where do i go on from here?

the process of demonstrating yourself as a good GSoC candidate is basically showing that you can do and figure out things on your own as well as interact/communicate/take feedback well

How do i show that i am capable of doing this?

you might start making a bit of a plan on your own and working to understand what the constraints are

there isn't really a specific plan in mind, as mentioned, but the description lists some of the things it'd be nice for it to do

having a specific proposal (or at least the start of one) is a place to show you can communicate your ideas clearly, as well as providing a more concrete place for others to give you feedback

if you get a student on the solr project, I'd be happy to help them

ideally the result would be something that both you can me can use

sounds reasonable, indeed

I didn't know MB was interested in replacing the search server

would you want to be a mentor, per se, or just a sort of advisor?

(guessing the latter, but figured I should ask)

I don't know MB-specific goals, so probably wouldn't be a good choice for a mentor

probably not, no, unless you consider a human-readable vs. a computer-readable view of the same data to be two sites

One for Picard - The tool. The other one for its plugins. ?

the sentence that says "a separate site" refers to the fact that a dedicated site for Picard (as opposed to the current so-called site, which is just a wiki page) can have more functionality

Got it. My bad. Sorry.

Also, can i work on proposals for two projects - the other one being read-only site for MB.

you're free to, and there's a decent amount of time before GSoC, but generally it's better to focus your efforts on one project or the other

Does only a single student work on a project or a team of selected students?

single student

So, the design and the content of the website will be the students responsibility?

especially the design..

well, in both of those projects, the core of the content is either from MB or generated by other users, but yes

everything from turning the idea into a real plan to doing the actual implementation is what the student does

(with a mentor to advise)

Who is the mentor of the latter project?

The wiki page has no mention.

mentors aren't assigned at this stage; they're assigned to students/projects after the students are accepted

the people listed on that page are just potential mentors/vaguely related people

Okay.

Am i required to attach some code with the proposal?

there's a questionnaire that'll ask for things like links to github accounts/code samples (for the sake of seeing what you know how to do), but not necessarily any code for the project itself

once the mentoring organizations are actually chosen (we aren't officially in the program yet, of course :P) and when student applications are open it'll be more clear

Okay.

for now you'd be best to learn more, develop a plan, etc.

One last thing: I am still confused over it, i mean, i could design the front-end in a hundred different ways, some better than the other.