I think it's better to create a specific fields to contain data representations of quotations.
The way it is now (with quotations being recorded in the titles of their entries):
So my proposal would be some like
Text version: String, Language
Shomoa, I see you've made a couple of comments today about strings (names, quotations, etc) in other languages. As it happens, we *can* store multiple languages for things already. Freebase's backend is inherently multilingual, and when anything is entered it is tagged with the language used. At the moment our UI only supports English, so we only use English strings here, but if you go to any topic (I like coffee for demonstration purposes) and then hit F8 for the dev toolbar and then go to "explore", you can see that we have keys for most topics in a variety of languages. Or check out this translation demo that one of our people put together. As you can see, we have lots of languages already when it comes to names for things.
The problem, of course, is exposing all this via the user interface on the website. This is a way off yet, but it is on our radar, and it's becoming increasingly important to us as we get more and more users from other parts of the world.
Hi skud, thanks for the reply. Sorry for not knowing about all the translated stuff but it feels kind of hidden for the moment (as you say).
I don't really get what you say when you talk about a dev toolbar and hitting F8, so I'll ask you directly: are "Kaffee", "Kava", "Cafè" etc stored as different topics in the database or are all part of a relation with the topic "Coffee", say in a table named <TopicNames>?
Yes, those are all stored on the *same* topic, the Coffee topic.
(As an aside, we don't use tables. We use a graph database, so it would be better to say they are stored as nodes connected to the "coffee" topic.)
Here's a direct link to the explore view, where you can see them: explore.
OK.
So I guess I'll have to wait for the translations being exposed via a newer interface before translating quotations.
Anyway, I think you'll agree that actual works such as books and movies should be recorded as different "topics" altogether. They represent different works and there may be several of them (for instance more than one translation for the same novel).
I think LibraryThing.com had this right and I see that in this database too books translations are new entries.
But what about movies? Should localizations of movies be added as title translations only? Is that really the right choice? You Americans (assuming you are from the US) usually don't dub foreign movies (apart from some animations) but here in (most countries of) Europe there is a solid dubbing tradition. Isn't dubbing equatable to books translations, although almost always movies get only one "translation" per country?
I wish to know what you and the staff in general think about this.
/OT
It sounds very interesting you didn't opt for a more mainstreamed relational approach.
Is there any performance with graphs when it comes to large amounts of data like this? I'm really interested.
P.S.
You were right about we posting in very different timings. It's kinda late in the night here and now I'm leaving. I'll be posting again tomorrow.
OK, with relation to our graph database, take a look at this blog post about graphd. One of the most compelling reasons we don't use an RDBMS is that our schema is changing constantly, but that article gives some other background and explains in a lot more detail.
I'm not an American, I'm Australian, though I do work at the Metaweb office in San Francisco these days, so I hope I'm fairly sensitive to international issues, even if English is my first language. I will wholeheartedly agree that Freebase is rather US-centric at the moment and that this really needs to change if we want to involve a broader community of contributors.
There are a number of steps we need to take to support international use. I see the following as the major areas:
I get involved in about six data modeling discussions every day and I'd say that the last point there, keeping our data models non-US-centric, is something that comes up with extreme frequency. All I can say in general terms is that the more people who stay involved in these issues, and advocate for an international point of view, the more it will occur. So please, keep raising these points!
Jeff will probably have opinions on the actual modeling for books, and I think some other people might pipe up with regard to your questions about Film. Unfortunately neither of those are my field of expertise. Perhaps you could raise the subject in each of the relevant domains?
You might also find our data-modeling mailing list to be of interest.
So a very popular trend in music in the past decade has been artists reworking pop/rock standards (like Paul Anka's swinging big band version of Smells Like Teen Spirit) as cover songs (tributes).
I would love to capture the relationship between source and covered version (sort of like adaptation in the media commons (should it be part of that type? Though I'd like to have a property that characterized the type of cover this is: identical?, revised, radical, language (english version of a foriegn language version), etc.
I guess another related property to consider would could be sampled song for samples as used in dance, rap/hip hop, mashups.
I hadn't seen Chris's post when I wrote my last one. I wonder if it's possible, though, with the current composition/arrangement types, to assert "cover-songness". A cover song, I as understand the term, is necessarily of a pop song (otherwise, a majority of classical, stage musical, opera, and folk recordings, and a great many jazz recordings, would be considered covers, which they aren't). Or in other words, I don't think that Jimi Hendrix's recording of the Star Spangled Banner would really be considered a cover, any more than one by Kate Smith be.
There was some modeling being done in June 2007 for such video products like Home Video (videotape/dvd/online/downloadable content), Direct-to-DVD/Videotape, OAV/OVA etc.
I guess it got erased on sandbox and was wondering if any new/further efforts to classify such things as the direct to dvd sequals of movies, episodic programming like Star Trek: Hidden Frontiers and other fan-created shows that are not for sale, the big Anime market for Videotape/Laserdisc/DVDs of non-broadcasted/non-theatical works.
Then there is the new internet-online only shows which aren't really podcasts but attempts to do close to if not network-quality production to be watched streamed on broadband connections.
