There are many dictionaries around and there is no real central place for them because they are under different licenses and some are proprietary.
There is an approach to so called authoritative databases which get connected to OmegaWiki, but in some way that is problematic since you don't see the whole stuff "together" - you need to go from the community "place" to the other one and vice versa to see if there are contents.
Like always when I do things in the house I have ideas and often can see things like a picture in my mind.
Bèrto is changing the architecture of the Database and we will have a new version, that is OmegaWiki 2.0.
The database design I saw is really easy to understand and so it is also easy to "imagine" new stuff.
Now the data within the community database is licensed under CC-BY and GFDL license - Francis (Apertium) just told me in the chat that there is a really good Friulan dictionary under GPL license. So why not include this into OW? I already hear you know: but that is GPL and with that not compatible ....
Well if I look at an expression page right now we have the lemma + the definition and sometimes various different definitions that are defined meanings. IMHO it would make sense to be able to see the contents that refer to that very same lemma from other dictionaries. The only thing needed is "tagging the entries as being part of this or that license" and therefore they are shown under a different header. You would have some thing:
Contents available under CC-BY and GFDL double license
Contents available under GFDL license
Contents available under GPL license
As long as we don't allow for the export of mixed licensed lists I don't see a problem, because the contents show up one after the other, like book titles. I am wondering if this fits in the actual database without the need to have separate ones where you then need to connect to ...
It is just a thought not really thought out well ... but there is something in there
...
Subscribe to:
Post Comments (Atom)
Khalil Gibran über die Musik
Die Musik wirkt wie die Sonne, die alle Blumen des Feldes mit ihrem Strahlen zum Leben erweckt. ( Khalil Gibran ) Image by Pete Linforth fr...
-
It is approximately a year ago when there was the first translated article on Wikipedia that was paid for. The idea then was to create a t...
-
Deutsch Vor Jahren, Ende 2009, habe ich, als ich noch in Italien lebte, über die Viedothek des Bayerischen Rundfunks eine Sendung der F... -
Weiß jemand, wie diese Pflanze heißt? - Antwort gefunden - Dank dem Kräuterallerlei-Forum. Weitere Info und Links siehe unter den Fotos. ...
2 comments:
A traditional dictionary is either in one language or in two languages. OmegaWiki is oriented in a different way. A concept in a Friulan dictionary needs to be connected with a DefinedMeaning or it forms the basis of a new DefinedMeaning. In either case not attributing it to a license that allows for re-use is problematic.
Yes, you can map and mash content but the holy grail is to integrate information. This is not achieved by a different data design, it is achieved by an agreement that content that is common is available under the specified licenses.
The license for common data in OmegaWiki is either CC-by or GFDL. This allows for re-use.
When you keep content separate, you have exactly the same situation as we currently have. Also without being able to link data, or to add the Friulan translations to the common data, I wonder what the benefit would be.
Thanks,
GerardM
I've emailed you a reply in private, but I'd like to re-iterate here that I think this is a good idea.
I thought I'd also supply a link to the Friulian dictionary you mentioned, which is licensed under the GPL here: http://www.cfl2000.net/cfl2000/
Note that this is nothing to do with Apertium (yet!), but we're planning to adapt the data.
As I explain in the email, I'd probably advocate for a triple-licence under CC-BY, GFDL, GPL... so that the content can be used in:
* Paper dictionaries, grammars, educational materials etc. (CC-BY, GFDL)
* Spell-checkers, grammar-checkers, machine translation software, information retrieval etc. (GPL)
Nothing is lost by triple-licensing, yet so much is gained! :)
Fran
Post a Comment