4 ms·
It failed because it was too expensive. 25 years ago I was looking at job in an elite project that ran smalltalk. I didn't get it but liked the language enough
by x87678r 6y ago
It failed because it was too expensive. 25 years ago I was looking at job in an elite project that ran smalltalk. I didn't get it but liked the language enough to investigate using it in my next job. The cost was crazy, like 1-2 months wages which was hard to justify when hardly anyone else used it. If it were the same price as C++ environments/libraries (yeah remember when you had to pay for container libraries) it would have been a hugely successful language.
- leoc 6y agoAnd then the buyer likely had the cost of extra RAM on top of that...
- protomyth 6y agoYeah, try to be a small developer in the 90's and deal with the Smalltalk companies. If you made it in the door now deal with runtime fees.
- Rochus 6y agoI don't think it's the main reason. I rather share these views: http://www.wirfs-brock.com/allen/posts/914 http://www.wirfs-brock.com/allen/posts/914. Smalltalk beyond that suffers from the same issues as any dynamically typed language. You quickly lose track when projects get bigger and many problems that a compiler could easily find are discovered much later. I used ST in the nineties and then switched completely to C++ and other languages. Recently, I started using ST again, when I built the two Smalltalk-80 interpreters; from my point of view, it feels quite similar to JS or Lua, but is even more inefficient, since even control structures and loops are dynamic objects.