[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [edict-jmdict] One or two things.



Regarding those words with suffixes, I am not against having them, but could we include a link to its root :

XX的 linked to XX (and maybe to 的, but it would lead to many reverse links for the most popular suffixes, so I am not sure about this one)

The same could apply to words with prefixes.

JL

Le 2012-09-18 05:57, Jim Breen a écrit :
On 18 September 2012 01:18, René Malenfant <rene_malenfant@hotmail.com> wrote:
On 2012-09-15, at 2:51 AM, Jim Breen <jimbreen@gmail.com> wrote:
One thing I would like to raise for discussion is the topic of
having words with the 的 suffix. As we know, this is is
highly productive, and I have been down on it it bit for that
reason. We've tended only to add it if a major reference has it too.
Something like 宗教的, which Marcus has proposed, is one such.
It's in GG5, but not in most 国語s (日本国語大辞典 has it, but it
has everything.) I'm wondering if the time has come to relax our
grip a bit and let more XX的 forms in. I can see a particular
reason with reverse lookups - at present you'd be hard put to
find a straight adjective meaning "religious/spiritual".

I don't have a problem with allowing them in as long as
there is some sort of consistent rule applied to 的 and to
other productive suffixes.  (e.g., ~さ, ~み, etc.)  However,
since they're all easily derived from their base form, in the
end, allowing all of them in may be a lot of effort for little
reward.  Perhaps there's a way to grab a bunch of them from
gg5, etc. and add them automatically as a batch?

A consistent rule that won't lead to open slather will be hard.
Possibly something that combines:
- already in a major dictionary;
- reasonably high-frequency
- useful for reverse lookups.

I had a quick look at GG5. It has about 400 headwords
of the XX的 variety, of which about 300 are already in JMdict.
It also has about 300 subheadwords of that sort and we
have around half of them already. I'll see if I can create
a list that can be worked from (can't just copy the glosses.)

For -さ/-み it's a similar story. For -さ GG5 has 80 headwords
and ~650 subheadwords, and for -み it reverses with about 500
headwords and 120 subheadwords. The -み headwords are things
like 足搦み/家並み/家並み, which we mostly have already.

I'll see what I can do with them.

Jim

--
Jim Breen
Adjunct Snr Research Fellow, Japanese Studies Centre, Monash University


------------------------------------

Yahoo! Groups Links