Re: [edict-jmdict] Examples file duplicate id numbers

[Francis Bond <bond@********>]

G'day,

On 5 April 2010 09:36, Glenn Maynard <glenn@********> wrote:

 

On Sun, Apr 4, 2010 at 9:20 PM, Francis Bond <bond@********> wrote:
> tatoeba thinks of the English and Japanese as first class objects, whereas the tuples are not --- therefore it does not have IDs for the tuples (although I also wish it did).  So if we are importing them into a database, we need to assign our own ID to the tuples.

That's just a badly broken design. You can't just assign your own ID,
because the next time you update, there's no reliable way to assign
the same IDs to the same tuples in a way that survives edits.

Not a battle I feel like taking up right now, though...

I'm extremely grateful to Trang for building the infrastructure to edit the example sentences, as well as adding translations in new languages.   So I don't want to ask him for more unless absolutely necessary.  I am quite happy to make my own IDs, and  map them to the tuples --- the tuples allow us to remap IDs edits. 

On the other hand, he must have a table of links so we could ask him to add IDs, which should not be so much more work --- that would be enough for everyone would it not?

-- 
Francis Bond <http://www3.ntu.edu.sg/home/fcbond/>
Division of Linguistics and Multilingual Studies
Nanyang Technological University