RE: bindings and assignments (was: Re: continuations)

["Gregor Kiczales" <address@hidden>]

I probably shouldn't get into this, but much of this conversation seems
to be a rehash of discussions from 20 years ago about why Lisp adoption
didn't happen, technical superiority and all that.

IMHO we've all seen a lot that we can learn from in that time. Just to name
a few stories, consider Lisp, Smalltalk, OO, Dylan, the Star, the Lisa, the
Mac, Windows, Objective C, C++, Next, Taligent, San Francisco...

One thing I've learned from all that is that if you want to change what
tools people use, you have to truly understand and accept why one set of
users has a different preference than seems right to you. Cynical stories
about people who don't use Lisp being stupid, or naïve don't cut it. They
don't give you the ability to really stand in the other person's shoes and
see it their way. And without that ability, you aren't going to be able to
speak to them in a way that makes them see why the points you see as important
might also matter to them.

Another thing I've learned is that there is less difference between technical
and social issues than we sometimes think. Some technical issues involve truly
objective facts. But  many technical issues are themselves social issues --
they are issues that one group of people, who self-identify as technical elites,
consider most important.

That's not to say that these so-called technical issues have no validity --
they do. It is just to say that their validity is socially constructed, and is
therefore on par with the kinds of issues other stakeholders find important.


My own favorite story about this dates back to a workshop that Peter Wegner
organized at MIT a while back (6 years ago maybe?).  In one of the sessions,
someone stood up and said something like "I don't understand how anyone can
think that Java is going to succeed with that type system it has".  He was
right, he didn't understand.