4 ms·

It's interesting how history rhymes with itself. This pretty much sums up how I felt about Java in 2000. There had been a healthy evolution and interchange bet

It's interesting how history rhymes with itself. This pretty much sums up how I felt about Java in 2000. There had been a healthy evolution and interchange between different language ecosystems up until the late 90's. And then Java took over. One pundit opined "In 20 years, we will not talk about Java for any groundbreaking technical choices, but business schools will talk about how it was the first widely marketed programming language and the effect it had." I'm sure this will rhyme again.

TimTheTinker 4y ago
He/she turned out to be somewhat right, though the JVM is still quite a marvel of engineering, not to mention Truffle. And today Java is a far more capable language than in 2000. I think the major difference between them and now for developers is that MIT/BSD/Apache-licensed, non-marketed languages and toolsets have gone mainstream. In other words, developers today no longer are paying through the nose for the tools they use, unless those tools are cloud-based.
- travisgriggs 4y ago
  > though the JVM is still quite a marvel of engineering I'm aware that some of those "marvels" were technologies harvested from Self/Smalltalk. But I suspect that there have been some innovations in the JVM that I just have not kept up with. When you say it's a "marvel of engineering", which types of things are you thinking of?
  - TimTheTinker 4y ago
    I'm thinking specifically of the runtime performance optimizations. Hotspot is incredibly good at JIT generation of fast machine code.
hyperman1 4y ago
You can add C++ to that list. It seems the pattern goes like this (with examples for C++, Java, Javascript) 1) There is an unmet need in the current technology ( e.g. C can't organize huge complexity, native app crashing is not tolerable on the web, web 2.0 is incompatible with full server side html generation). 2) A current techical powerhouse withbopen enough culture has a solution fixing it, while leaving current knowledge mostly intact ( C++ is C with classes, Java is safe C++, javascript is webified Java) 3) The powerhouse tech is used everywhere because it's voice is heard louder than the small playser, and also everyone wants to feel like a big player.( at&t, sun, google/facebook) 4) A cambrian explosion pops up full of components and frameworks. Most of them dies, nobody knows what tobbet on long term. The new good tech is both used and abused on every possible problem. 5) Things settle down, we learn what to use the tech for. 6) Circumstances change, we start fighting our tools, and the cycle starts again. The new group will not understand why the old guard is still loving their old klunkers (C illegal behaviour, Java XML), even if they have to construct a strawman first. In fact, both Cobol and C might be exampled of this, but they were before my time. There are plenty of smaller examples out there.

TimTheTinker 4y ago

He/she turned out to be somewhat right, though the JVM is still quite a marvel of engineering, not to mention Truffle. And today Java is a far more capable language than in 2000. I think the major difference between them and now for developers is that MIT/BSD/Apache-licensed, non-marketed languages and toolsets have gone mainstream. In other words, developers today no longer are paying through the nose for the tools they use, unless those tools are cloud-based.

travisgriggs 4y ago
> though the JVM is still quite a marvel of engineering I'm aware that some of those "marvels" were technologies harvested from Self/Smalltalk. But I suspect that there have been some innovations in the JVM that I just have not kept up with. When you say it's a "marvel of engineering", which types of things are you thinking of?
- TimTheTinker 4y ago
  I'm thinking specifically of the runtime performance optimizations. Hotspot is incredibly good at JIT generation of fast machine code.

hyperman1 4y ago

You can add C++ to that list. It seems the pattern goes like this (with examples for C++, Java, Javascript) 1) There is an unmet need in the current technology ( e.g. C can't organize huge complexity, native app crashing is not tolerable on the web, web 2.0 is incompatible with full server side html generation). 2) A current techical powerhouse withbopen enough culture has a solution fixing it, while leaving current knowledge mostly intact ( C++ is C with classes, Java is safe C++, javascript is webified Java) 3) The powerhouse tech is used everywhere because it's voice is heard louder than the small playser, and also everyone wants to feel like a big player.( at&t, sun, google/facebook) 4) A cambrian explosion pops up full of components and frameworks. Most of them dies, nobody knows what tobbet on long term. The new good tech is both used and abused on every possible problem. 5) Things settle down, we learn what to use the tech for. 6) Circumstances change, we start fighting our tools, and the cycle starts again. The new group will not understand why the old guard is still loving their old klunkers (C illegal behaviour, Java XML), even if they have to construct a strawman first. In fact, both Cobol and C might be exampled of this, but they were before my time. There are plenty of smaller examples out there.