[E-Lang] Re: inheritance, delegation, mental models for beginners

[Chip Morningstar]

>Chip, I believe you're saying that we should simply stop here and say "this 
>is inheritance".  Dean, I think you think you're agreeing with Chip, but I'm 
>not at all sure you'd agree with what I think Chip is saying (whether Chip 
>is actually saying it or not.  Chip?)

I believe the linguistic issues we are discussing on this list are of vital
importance to the success of E among the teeming millions, but I also believe
that 98% of the intended audience for E would find the content of these
discussions utterly and forever incomprehisible. Many of the issues which seem
of looming importance to many of us will seem subtle and inconsequential to
most of the rest of the world. However, to succeed it is not necessary to
persuade the rest of the world that these things really are of looming
importance but merely necessary to get the rest of the world to accept
our resolution of these questions. I think it is possible to fail by explaining
too much, because so doing makes things seem difficult and arcane.

The point I was making in my earlier post is that we should separate the
rhetorical and marketing issues from the syntactic and semantic ones. I believe
it is true that a large number of C++ programmers will have issues with the
language if it doesn't have inheritence, but vanishingly few of them are smart
enough to even tell you what inheritence *is*.  What I'm saying is you should
implement whatever inheritence/delegation mechanism seems best given the
consensus of the community here, and then label this mechanism "inheritance"
without attempting in the marketing of E to explain the nuances of what makes
whatever E does in this respect different from what C++, Java or other
languages do. You want to conserve your extremely limited budget for
unusual-language-concepts-that-need-to-be-explained to use on really big and
radical concepts like capabilities and promises.

Chip