-- BotBot disconnected, possible missing messages --
-- BotBot disconnected, possible missing messages --
BrainzBot joined the channel
ZaphodBeeblebrox is now known as CatQuest
CatQuest is now known as dusjekatt
void09
flawed 30 million items db, or "perfect" 50k items db. hmm :)
No, I am not complaining about anything. Just trying to figure out what's the plan here
dusjekatt is now known as CatQuest
CatQuest
honestly bb is young? we've only really had the ability for easy addition for a few years now. since monkey took over as main bb dev. with help from several gsocers and such.
this is what, the 4th, 5th? year we're in gsoc?
we didt have much of a community until bookogs caved. and that was what, 2 years? considering all this, that bb isn't even a finished database, it's stil in beta very much
think 50k is pretty good
musicbrainz is comming up on 20 years now. it has many many more items. yo ucan't compare it with a database that is barely 5 years old as it is.
furthermore we are improving every day, the database entry and database model will be become even better. and people will come, I'm sure
so I suggest instead of com^^^"trying to figure out" why there are so few items?
add some, add everything. hlep improve BB
also, quality over quantity
this is why mb is today the go-to database for music metadata
it wasn't so whne it was new. not so even 10 yearsago
trickle addition is better than mass mport of data
because we're in this for th long haul. yes
so at first, there is little. not much. very lacking
and thne some people will come, add theri shit
so more peopel will come say "this is shit data I will improve it!" and more peopel will come
tracktion develops
more and more
and suddenly those "30 million flawed items" in that db is peanuts
will it take time?
yes
is that a bad thing?
no.
void09
Yes, makes perfect sense what you say. Although I want to know how openlib is flawed and how BB is better in that regard.
monkey
Hi void09, sorry for the delay! As has been mentioned by other people, do keep in mind BookBrainz is in its infancy; as such it will suffer from comparisons with other more established projects (including some that have since closed shop).
OpenLibrary is not flawed, nor do I think we are really competing in the space. BookBrainz benefits from the infrastructure, experience, knowledge and community of MusicBrainz, while the OpenLibrary benefits from the same resources at the Internet Archive.
Our goals are mostly aligned, with these main differences: OpenLibrary is geared towards digitized content and lending of digital books, while BookBrainz focuses on metadata only and the stability, perennity and curation of the identifiers.
void09
hi monkey
monkey
In my opinion, having similar projects with mostly aligned goals is beneficial for the longevity of both projects.
void09
digitized/lending content is an addition to metadata. could have the same thing on bb if you just added some extra fields in the database I guess
monkey
Soon we will open collaboration channels between OL and BB with the aim to improve the quality of data on both sides, although exactly how remains to be seens (and programmed :) )
void09
Oh that's good
CatQuest
ah that open liberary, i confused it with a dfferent project, sos
monkey
void09: We can't legally do that. That's why MusicBrainz hosts all its cover art (similar copyright issues) with the Internet Archive with the joint project Cover Art Archive https://coverartarchive.org/
void09
I arrived here looking for a self hosted platform to aid in the collaboration of a group towards digitizing some paper media collections (old books, magazines and such)
Of course you can't (well you can for public domain/opensource content)
I was just saying works metadata is independent of actual content
monkey
Nice! Well, we certainly can't help with the digitization aspect, but I'm keen on improving how BookBrainz stores magazines and other such publications. Currently it's not very well adapted IMO
CatQuest
+1 o that
void09
Yes, neither is openlibrary. i wanted to have something ready to go, not having to hire someone to do mods, for a limited scope project
monkey
I see. I can't think of any other open source project that would be more adapted to magazines
void09
By the way, are hashes of content legal to store/distribute ? Or does that depend on the country you're in, or too unpredictable to risk ?
monkey
That's a very good question; the general answer for the MetaBrainz foundation is always "when in doubt, don't"
The Internet Archive is legally set up as a library so there's a lot of differences in what they're allowed to do, but MetaBrainz on the whole steers away from any possible copyright issues.
CatQuest
ah it was worldcat I misstook it for, sorry :(
monkey
We've had the same questions recently regarding fine-grained analyses of audio data: now that AIs are getting good enough, one could reconstitute a song from fine-grained spectrum analyses, and so that is a barrier for us.
void09
That book lending thing of internet archive is pretty cool. Even DRM-ed, it's actually possible to circumvent it. I wonder what the point is in keeping 80's magazines copyrighted. They can't make any more money off them
monkey
The gods of copyright are irrational…
Yeah, beats me.
I think they probably see everything in terms of "demand". If you want them catalogued for posterity, that counts as "demand", and so the remote possibility of money.
"Who knows what else people can do with 80's magazines if we let go of the copyright! We might be missing on some sweet cash !"
void09
It's pretty funny how lending of digital content works, like simulating lending of a physical copy. First I thought it was a joke
monkey
Me too :D But that's the deal the IA had to strike
void09
They seem to be doing it right, whatever they're doing. They're huge, they are sustainable, and they still exist :D
monkey
And even then, they are under legal battle with publishers and authors associations
void09
Oh I remember they pulled some unnecessary stunt during covid, to allow unlimited lending. Is that what you are talking about ?
monkey
Yep. We aim for the same goals; we'll be there in 20 years, and by then we'll also be huge :)
Yes, we did consider FRBR, but it's a serious departure from the existing model
That being said, perhaps some of the concepts could find their way into the BB model
For the most part, in BookBrainz the idea of an "item" doesn't make much sense (it makes a lot of sense for libraries that have specific physical or digital items to lend)
The other concepts should map fairly well to the BookBrainz schema if I'm not mistaken
(It's been a while since I looked at FRBR in details)
void09
Item as in a specific instance of a published edition (eg. 1 piece unique physical book) ?
vs another such piece that might have a dick drawn on the first page :d
monkey
Yes, exactly.
void09
Yes that is completely unnecessary, as long as BB can be used as a parent database and items implemented as a extra db layer
monkey
Well, all in all that is the goal: to provide curated metadata with stable identifiers as a base for other projects
Oh, and all of that open source, of course :p Otherwise, ISBNs, VIAF ids, etc. would do the trick I suppose.