Well, for me personally, do I want Smalltalk to be mainstream like C#
and Java?  Never.  Becoming mainstream currently means getting flooded
by a bunch of people who should not be programming.  C# and Java are
both charging down the path of complexity as fast as they can, and
this is what I would most want to avoid in Smalltalk.  When you go
mainstream then you have people with some stupid reason they need just
one more operator, precedence level and so on.  People came to Java to
avoid the complexity of C++, but all the people that caused that
complexity came along too and soon Java will be worse then C++ ever
was.

Do I want Smalltalk to be obscure?  No of course not.  I think a good
language to look up to is Haskell.  They manage to create for
themselves a reputation for simply being too hard to grasp, so mostly
only smart people came.  This is what Smalltalk needs.  More smart
people.  We need better, faster VM's, libraries, solutions to hard
problems.  We don't need the masses from C++ et al coming here
"solving" these issues the same terrible ways they solved them in
their languages.

I understand Tim's (and countless others) point about career
opportunities, but if Smalltalk became the language of choice for
sitting in a cubicle in a huge organization grinding out the same
meaningless, boring code year after year, would that be a victory?

On Nov 6, 2007 5:07 AM, Colin Putney <cputney at wiresong.ca> wrote:
>
> On 5-Nov-07, at 2:21 PM, Chris Cunnington wrote:
>
> > Smalltalkers talk about wider adoption, but they don't really want
> > it. They
> > like community to be as small and cozy as an English smoking lounge.
> > I was
> > talking to a developer at Smalltalk Solutions of a large product,
> > and he
> > said he was afraid of too many people using their product, because
> > then
> > people would blame their own deficiencies on the product.
>
> I occasionally get into arguments with Smalltalk advocates for exactly
> this reason. I don't want wider adoption of Smalltalk; I think the
> community is starting to get too big as it is. There are two things I
> do want, though.
>
> One is for Smalltalk to be "respectable." If you've got a problem that
> would be nicely solved in Smalltalk, it's a real shame to have to use
> some other language because the client or manager succumbs to FUD.
>
> The other is for the community to be effective. Using Smalltalk is
> only viable if libraries get written and maintained, tough questions
> can be answered, VM technology progresses, new platforms are supported
> as they emerge, and new language concepts can be integrated. As long
> as the community is big enough to let Smalltalkers remain
> Smalltalkers, it's big enough.
>
> To some degree, the community needs to be a certain size to
> effectively support its members, but I think that minimum size is
> shrinking. The "dark ages" for Smalltalk were that period when it had
> already become clear that Smalltalk would not become the standard
> language for enterprise development, but before the Internet had
> reached the level of social sophistication that we see now.
>
> Back to the original article. Shirky talks about the need for
> structure to all the group to protect it's principles against both
> newcomers and it's own group tendencies. I think the Smalltalk
> community would benefit by focussing on mutual support rather than
> evangelism.
>
> Colin
>
>