I can see that ST companies might have missed an opportunity, but Sun was a BIG presence in the dot-com era and was in a position to effectively support and promote Java (including adding networking).

It seems to me that the next major opportunity, and maybe a more serious failing by ST companies, was web development tools/framework as captured by Ruby/Rails. We are only now seeing effective commercial offerings like GLASS from Gemstone.

Clearly successful business models can be based on open source software. Commercial ST vendors don’t seem to have fully embraced the potential of that business model. This is unfortunate because part of what OSS-commercial vendors sell is access to innovation and ST is a great environment for innovation.

In addition to providing better tools for professionals, commercial ST vendors could help to promote ST/Squeak as an educational platform for kids/teens. Funding some kind of educational coordinator/evangelist would be a big help. ST/Squeak already has some traction in this area. Kids can be surprisingly innovative and once they start on traditional programming education with procedural languages it becomes MUCH more difficult to attract them to ST.

There was the big corp pull but Digitalk still had the independents who had bought at $99 - my company happened to be located a few floors up from Digitalk in the early days and I went downstairs and handed Jim Anderson a check - it didn’t feel so corporate then.

“Back in those days “web” and “corporate” were two different things.”
true

“As I understand it, Sun went to ParcPlace to get VW Smalltalk to be the “Oak” language for “Set Top Boxes” and such (remember, this was before Java was targeted at the web. Everyone was all about “Set Top Boxes”.) Apparently PP wanted to license it for more than an arm and a leg, and the rest is history. This is anecdotal but heard from more than one source.”
I’d heard this too and always felt it was at least somewhat accurate

@Before the internet took off, networking was secondary.
True, but there were several commercial Smalltalks that had every opportunity to jump on the bandwagon before Java ever got on the scene.

@It seems to me that ST’s business case is “making the difficult easy, and impossible possible”. The complexity of the solutions that programmers tackle keeps increasing. (Is there a law for this yet?) FORTRAN and COBOL were tools for fast computations and accounting with well defined inputs and outputs. Then C and soon C++/Java took over and allowed more complex systems to be modeled. ST has been head of its time.

If you check my other posts, you’ll see that’s not how it happened. ST did make the business case you mention in the late 80’s early 90’s and had clearly beaten C/C++ mindshare wise before Java ever arrived. IBM had blessed it and during that era $99 was like open source so a growing community of individuals had gathered around Digitalk as well. I agree that the commercial entities should support Squeak and GNU ST better.

Tech history seems to prove that incremental progress is valued over paradigm shifts. With “write once, run everywhere” Java offered to solve a real problem for developers and was innovative without being too different from what past procedural programmers were used to. Before the internet took off, networking was secondary.

It seems to me that ST’s business case is “making the difficult easy, and impossible possible”. The complexity of the solutions that programmers tackle keeps increasing. (Is there a law for this yet?) FORTRAN and COBOL were tools for fast computations and accounting with well defined inputs and outputs. Then C and soon C++/Java took over and allowed more complex systems to be modeled. ST has been head of its time.

This is somewhat borne out by some Java developers recent complaints about how bloated it is. Clearly, Java tries to be everything to everyone and there are Java efforts to incorporate Ruby and Rails frameworks (JRuby, Groovy, etc.)

It’s hard for any non-mainstream language to succeed. (Remember the complaints that Java was dog slow? and the endless “but we can do this in C++” demonstrations?) Java had the enduring support of Sun but its success was never a given. IMO Commercial ST vendors should support Squeak (better) so that they can take advantage of innovations like Seaside. (Surprisingly, they seem to be late to incorporate Seaside.)

Emacs will likely always be my window manager when not in a Smalltalk environment. And when I worked for Gemstone, Emacs *was* my Smalltalk environment. Kind of nice there too.

“my real point is that Smalltalk community didn’t strike the web while the iron was hot - before Java showed up and got traction.”

I agree. But thinking back on those days in the Smalltalk community there was a real “corporate” feel to Smalltalk. Squeak was not really out yet, everything was VW and Digitalk… multi $K USD products because they thought they were going to save IT from itself and make some scratch at the same time.

Back in those days “web” and “corporate” were two different things.

As I understand it, Sun went to ParcPlace to get VW Smalltalk to be the “Oak” language for “Set Top Boxes” and such (remember, this was before Java was targeted at the web. Everyone was all about “Set Top Boxes”.) Apparently PP wanted to license it for more than an arm and a leg, and the rest is history. This is anecdotal but heard from more than one source.

Regards
Friedrich