Re: Simple Example, How To Model It In RM

From: Neo <neo55592_at_hotmail.com>
Date: 29 Jun 2005 16:43:49 -0700
Message-ID: <1120088629.182557.161880_at_g49g2000cwa.googlegroups.com>


> "Because" is not a verb, it is a conjunction.

The classification of 'because' as a verb in the example is a simplification and the term is used loosely.

> You cannot (should not be able to) say "Sue because John",
> nor can you say "(Sue like John) friend (Sue like Mary)".

You are correct from an English/Language perspective (I'm not too concerned with constraints at this time).

> So you now have 5 "types":
> Person (e.g. "John")
> Verb (e.g. "likes")
> SimplePhrase (Person Verb Person e.g. "Mary likes John")
> Conjunction (e.g. "because")
> CompoundPhrase (SimplePhrase Conjunction SimplePhrase,
> e.g. "(Sue likes John) because (Mary isFriendOf John)")
> Do you agree?

I would agree that anything can have 0 to many classification (the exact classification of a thing isn't too important to me at this point).

> There is no point in trying to replicate a flawed design in RM!

Feel free to model with more appropriate classifications.

> Also, do you expect to be able to handle more complex statements like
> "((Sue likes John) and (Sue likes Mary)) because (Mary isFriendOf John)"?

Yes, and if you would like to accommodate for those in the initial RM schema/query, that would be OK. Received on Thu Jun 30 2005 - 01:43:49 CEST

Original text of this message