[rfk-dev] I'll make a new vanilla NKI list

[Peter A. H. Peterson]

Quoting Leonard Richardson:
> As the original author of robotfindskitten I'd like to offer my
> services in creating a new NKI list for vanilla. 

I think this is a great idea -- thank you for volunteering!

> I would like your opinion on approximately how many NKIs should be in
> the list. There are about 740 in the big vanilla list. There are about
> 200 that were added to the Inform version, but spot checks indicate
> most of those are also present in vanilla. Should we aim for more like
> 700, or more like 1000? Should we go big or prune it down?

I don't actually have an opinion about size. I think we should have
that discussion if things grow considerably larger, although even with
rfk gaining notoriety over the last 15 years that has not happened. So
I think at this point I think you should just take all the ones that
you think "make the cut" from the extant lists and go from there.

Also, FYI -- The Big List (messages.h-731) is not actually the
"vanilla list". To get the vanilla list, check out the new git repo
from Sourceforge. This list actually doesn't include *any* of the
Inform NKI. (This is because the Grand Unified NKI List that I made in
anticipation of the new Alexey version did not make it into the last
revision as made by Ryan Finnie.)

> IMO robot and kitten should never take definite articles.

Absolutely agreed. And they are not proper nouns. And there should
never be spaces in the name.

> 2. Regardless of how the NKIs are merged, I'd like to omit the "Clang"
> NKI from vanilla. The NKIs are deliberately ambiguous and presented in
> a context-free way. I don't want a kid thinking rfk is calling them
> the same name their bullies use.

I agree. This may seem overly cautious, but I think that is
appropriate in cases where the ambiguity could be seen as being
cutting.

pedro

-- 
Peter A. H. Peterson
Graduate Student Researcher
Laboratory for Advanced Systems Research
University of California, Los Angeles