From: "T. Onoma" <transami@...>
Date: 2004-01-24T11:13:35+09:00
Subject: Re: [ANN] Ruby/Extensions v0.3 released

On Saturday 24 January 2004 12:58 am, Gavin Sinclair wrote:

> Those methods make sense only in very specific cases.  They assume you
> want +nil+ to be like a valid string, integer, array, etc.  That's an
> assumption I (almost) never want to make in my code.

do you think NilClass' to_a and to_i methods should be removed then?

> One of the stated goals of the project is that methods be general
> purpose.  That's a pretty vague aim, but it's pretty clear that the
> methods above are not general purpose.

hmm... to me they seem even more general purpose then most, since they are so 
very basic. the differnce is whether one views the NilClass as a 
representation of "emptiness" vs. "notness" (for lack of better words). my 
additional methods thus adding in the way on the point of view of 
"emptiness". in my experience (and honestly a bit to my suprise) i have never 
run into any difficulties by including these methods and putting to use these 
two different conceptions of nil; though sometimes i do think it might be 
nice to separate the "notness" out into a separate NackClass, all the same.

anyway, i think you've answered my question. Ruby/Extensions is intended as a 
restricted set of approved inclusions rather than an open collection of 
acceptable inclusions. that's fine. i just wanted to be clear on the goals of 
the project.

thanks gavin,
T.