3 ms·

Smalltalk ruined my life

2 points by horrido 10y ago

gk1 10y ago
From the author's LinkedIn profile: > Leading a marketing and PR campaign to promote and popularize the Smalltalk programming language.
- horrido 10y ago
  I've never made any bones about this. Of course, I'm an evangelist.
SlipperySlope 10y ago
But where are the libraries? Compare the size of Smalltalk's ecosystem to Java's. Not to mention the various competing flavors of Smalltalk, whose code bases may be incompatible with one another.
- horrido 10y ago
  The size of the Smalltalk user base isn't large enough to support a healthy ecosystem. Hence, the reason for my evangelism to grow the community! Another issue is fragmentation: there are too many dialects. That's why I generally promote Pharo, which is the largest open source Smalltalk. If I can get developers to rally around Pharo, then we will have the best chance of developing a strong ecosystem.
prodigal_erik 10y ago
"Provably the most productive" seems to boil down to estimating that Smalltalk is more terse in statements per function point. Incidentally, by that metric Python is only half as productive as Visual Basic, which doesn't seem plausible. No comment about extra test coverage and prod bugs caused by lack of static typing. "Image-based persistence" means your production-ready code and dinking around during prototyping (in a language with frictionless monkey-patching!) are all buried in one big mutable blob. It's often argued the impedance mismatch with version control is what killed Smalltalk.
- horrido 10y ago
  "It's often argued the impedance mismatch with version control is what killed Smalltalk." Wrong argument. Here's the truth: https://medium.com/smalltalk-talk/why-smalltalk-failed-to-dominate-the-world-93e7e4195039 https://medium.com/smalltalk-talk/why-smalltalk-failed-to-do... Smalltalk does have version control. It's just not the same version control software that everybody else uses (eg, Git, Subversion, etc.).

3 ms·

Smalltalk ruined my life

2 points by horrido 10y ago

gk1 10y ago
From the author's LinkedIn profile: > Leading a marketing and PR campaign to promote and popularize the Smalltalk programming language.
- horrido 10y ago
  I've never made any bones about this. Of course, I'm an evangelist.
SlipperySlope 10y ago
But where are the libraries? Compare the size of Smalltalk's ecosystem to Java's. Not to mention the various competing flavors of Smalltalk, whose code bases may be incompatible with one another.
- horrido 10y ago
  The size of the Smalltalk user base isn't large enough to support a healthy ecosystem. Hence, the reason for my evangelism to grow the community! Another issue is fragmentation: there are too many dialects. That's why I generally promote Pharo, which is the largest open source Smalltalk. If I can get developers to rally around Pharo, then we will have the best chance of developing a strong ecosystem.
prodigal_erik 10y ago
"Provably the most productive" seems to boil down to estimating that Smalltalk is more terse in statements per function point. Incidentally, by that metric Python is only half as productive as Visual Basic, which doesn't seem plausible. No comment about extra test coverage and prod bugs caused by lack of static typing. "Image-based persistence" means your production-ready code and dinking around during prototyping (in a language with frictionless monkey-patching!) are all buried in one big mutable blob. It's often argued the impedance mismatch with version control is what killed Smalltalk.
- horrido 10y ago
  "It's often argued the impedance mismatch with version control is what killed Smalltalk." Wrong argument. Here's the truth: https://medium.com/smalltalk-talk/why-smalltalk-failed-to-dominate-the-world-93e7e4195039 https://medium.com/smalltalk-talk/why-smalltalk-failed-to-do... Smalltalk does have version control. It's just not the same version control software that everybody else uses (eg, Git, Subversion, etc.).