That Graham guy you made me think of.. he did the reverse. He came up with the problem first and then solved it leveraging the power of the ultimate tool.
I think you misunderstand the article. The problem wasn't "What are the jobs to which the tool is suited?" It was "What are the work environments that will let us use the tool that is suited to the job?"
One of those was a day job where they understood the politics well enough to work around institutional bias against non-Blub languages. Later it was their own company where they could make technical decisions without having to justify them to non-technical people.
Bear in mind that the article was written by a product evangelist for a Smalltalk vendor. It's not surprising it has the perspective of assuming the tool and looking for the application.
"Bear in mind that the article was written by a product evangelist for a Smalltalk vendor."
Indeed. From the post,
"They have two good Smalltalkers with a couple of laptops. They use available WiFi and open source code, and they deploy on cheap, hosted Linux boxes. [ed. I'll interject at this point - you could use commercial software all the way down, like Cincom Smalltalk, and not pay a dime until large amounts of revenue arrives]."
