5 ms·

> I just think it will slowly become less relevant. That's true for virtually anything. Do you think C/C++ or Java or JS will be mainstream 30 years from now.

> I just think it will slowly become less relevant. That's true for virtually anything. Do you think C/C++ or Java or JS will be mainstream 30 years from now. They will be in the same situation FORTRAN is today: important because of lots of legacy code, but not a viable choice when starting a new project. Of course some of them will never die. LISP comes to my mind. :-)

noelwelsh 14y ago
Sure. It's the time scale that's important. Irrelevant in 30 years has vastly different implications to irrelevant in 5.
coldtea 14y ago
>Of course some of them will never die. LISP comes to my mind. :-) Or, to put it another way, LISP was never alive to begin with. If Ruby/Python had a community and third party libraries the size of LISP we would consider them mostly irrelevant languages.
- rjbond3rd 14y ago
  > ...we would consider them mostly irrelevant languages. And apparently "we" are using "irrelevant" to mean not fashionable. Or does it mean, "Do not use for new IT projects" or something? What does it matter what other people do, especially those with a strong (if small) community, frameworks, libraries, books, articles, conferences... And where does Smalltalk fall? I hate to tell those people their language is "irrelevant" -- will their programs crash? And which LISP, by the way? This irrelevance property is dumber than a language debate because we don't even debate them, we condemn them from the outside by applying our own arbitrary criteria. Why do we need to belittle the work of others? Maybe we are not as sure as we appear.
  - coldtea 14y ago
    >And apparently "we" are using "irrelevant" to mean not fashionable. Or does it mean, "Do not use for new IT projects" or something? What does it matter what other people do, especially those with a strong (if small) community, frameworks, libraries, books, articles, conferences... It matters because there is not enough of "community, frameworks, libraries, books, articles, conferences". If I want to do something in LISP there are 2-3 libraries in different states of broken and/or abandoned. In Python I can usually find something with much more ease. In Java even more so. The same for hiring programmers AND getting hired yourself, etc etc. >And where does Smalltalk fall? I hate to tell those people their language is "irrelevant" -- will their programs crash? You don't have to tell them. They know it already. It has been ever since most Smalltalk shops closed and the bigger ones turned over to supporting Java. Now, people using Smalltalk to do hobby work, or even nice important work with paying customers, good for them. Their programs will not crash (unless their Smalltalk platform of choice gets abandoned to die with too few contributors or is proprietary and doesn't get support for newer platforms and architectures). But they wont be getting as much new toys as the other boys. >And which LISP, by the way? All of them, inc. Common Lisp. Clojure might have a fighting chance, but it still is too small for people who value ecosystem over cowboy style "let's build all the wheels ourselves". >This irrelevance property is dumber than a language debate because we don't even debate them, we condemn them from the outside by applying our own arbitrary criteria. Irrelevant? It's very pragmatic. Number of available employers, employees, number of documentation, vibrancy of the ecosystem, new libraries and continuous support for older libraries, etc. I would love to have a Smalltalk that is well supported to play with. One that is not aimed at kids and doesn't have a Fisher Price UI like Squeeze, and that can match Python or Java for ecosystem. Unfortunately none exists.
    - rjbond3rd 14y ago
      I appreciate that you write really well, but I think your argument boils down to "there's safety in numbers." And that is a totally valid strategy for IT careers, but I would quit programming if I had to follow that advice. It feels too much like, "Stick with the herd." Just one person's opinion.
PommeDeTerre 14y ago
C and C++ will very likely still be "mainstream" 30 years from now. Essentially every important piece of software today, from operating systems to programming language implementations to server software to applications, is written in one or both of them. All of these systems are not just going to go away. C is over 40 years old, and C++ around 30, yet in that time nothing has come along that can compete with them. Some other languages have tried, but their success has only ever been partial, and typically in just one or two specific domains.
- whalesalad 14y ago
  Go is making a strong go at it. No pun intended.
  - PommeDeTerre 14y ago
    Go hasn't really seen much success. It isn't as obscure as many other languages, and being supported by GCC does help it out, but we haven't seen any major adoption of it. Maybe that'll change at some point, but it doesn't look very likely.
army 14y ago
The rumours of Fortran's death are greatly exaggerated. People still write new code these days in Fortran for scientific computation. Fortran compilers are still markedly better at optimizing numerics, multi-dimensional arrays, etc, partially because of the effort invested in the compilers and partially because the language is easier to optimize. It's definitely a niche language.
- damoncali 14y ago
  I used FORTRAN at NASA as late as 2003. Id be surprised if they were not still using it. It was painful, though.
wes-exp 14y ago
Lisp has had keyword arguments since at least the 80's. Ruby 2.0 adds them and this is supposed to be the "new" hotness? Get off my lawn!
kamaal 14y ago
Lisp will likely remain alive, but never as a mainstream language which really defeats the whole purpose.

noelwelsh 14y ago

Sure. It's the time scale that's important. Irrelevant in 30 years has vastly different implications to irrelevant in 5.

coldtea 14y ago

>Of course some of them will never die. LISP comes to my mind. :-) Or, to put it another way, LISP was never alive to begin with. If Ruby/Python had a community and third party libraries the size of LISP we would consider them mostly irrelevant languages.

rjbond3rd 14y ago
> ...we would consider them mostly irrelevant languages. And apparently "we" are using "irrelevant" to mean not fashionable. Or does it mean, "Do not use for new IT projects" or something? What does it matter what other people do, especially those with a strong (if small) community, frameworks, libraries, books, articles, conferences... And where does Smalltalk fall? I hate to tell those people their language is "irrelevant" -- will their programs crash? And which LISP, by the way? This irrelevance property is dumber than a language debate because we don't even debate them, we condemn them from the outside by applying our own arbitrary criteria. Why do we need to belittle the work of others? Maybe we are not as sure as we appear.
- coldtea 14y ago
  >And apparently "we" are using "irrelevant" to mean not fashionable. Or does it mean, "Do not use for new IT projects" or something? What does it matter what other people do, especially those with a strong (if small) community, frameworks, libraries, books, articles, conferences... It matters because there is not enough of "community, frameworks, libraries, books, articles, conferences". If I want to do something in LISP there are 2-3 libraries in different states of broken and/or abandoned. In Python I can usually find something with much more ease. In Java even more so. The same for hiring programmers AND getting hired yourself, etc etc. >And where does Smalltalk fall? I hate to tell those people their language is "irrelevant" -- will their programs crash? You don't have to tell them. They know it already. It has been ever since most Smalltalk shops closed and the bigger ones turned over to supporting Java. Now, people using Smalltalk to do hobby work, or even nice important work with paying customers, good for them. Their programs will not crash (unless their Smalltalk platform of choice gets abandoned to die with too few contributors or is proprietary and doesn't get support for newer platforms and architectures). But they wont be getting as much new toys as the other boys. >And which LISP, by the way? All of them, inc. Common Lisp. Clojure might have a fighting chance, but it still is too small for people who value ecosystem over cowboy style "let's build all the wheels ourselves". >This irrelevance property is dumber than a language debate because we don't even debate them, we condemn them from the outside by applying our own arbitrary criteria. Irrelevant? It's very pragmatic. Number of available employers, employees, number of documentation, vibrancy of the ecosystem, new libraries and continuous support for older libraries, etc. I would love to have a Smalltalk that is well supported to play with. One that is not aimed at kids and doesn't have a Fisher Price UI like Squeeze, and that can match Python or Java for ecosystem. Unfortunately none exists.
  - rjbond3rd 14y ago
    I appreciate that you write really well, but I think your argument boils down to "there's safety in numbers." And that is a totally valid strategy for IT careers, but I would quit programming if I had to follow that advice. It feels too much like, "Stick with the herd." Just one person's opinion.

PommeDeTerre 14y ago

C and C++ will very likely still be "mainstream" 30 years from now. Essentially every important piece of software today, from operating systems to programming language implementations to server software to applications, is written in one or both of them. All of these systems are not just going to go away. C is over 40 years old, and C++ around 30, yet in that time nothing has come along that can compete with them. Some other languages have tried, but their success has only ever been partial, and typically in just one or two specific domains.

whalesalad 14y ago
Go is making a strong go at it. No pun intended.
- PommeDeTerre 14y ago
  Go hasn't really seen much success. It isn't as obscure as many other languages, and being supported by GCC does help it out, but we haven't seen any major adoption of it. Maybe that'll change at some point, but it doesn't look very likely.

army 14y ago

The rumours of Fortran's death are greatly exaggerated. People still write new code these days in Fortran for scientific computation. Fortran compilers are still markedly better at optimizing numerics, multi-dimensional arrays, etc, partially because of the effort invested in the compilers and partially because the language is easier to optimize. It's definitely a niche language.

damoncali 14y ago
I used FORTRAN at NASA as late as 2003. Id be surprised if they were not still using it. It was painful, though.

wes-exp 14y ago

Lisp has had keyword arguments since at least the 80's. Ruby 2.0 adds them and this is supposed to be the "new" hotness? Get off my lawn!

kamaal 14y ago

Lisp will likely remain alive, but never as a mainstream language which really defeats the whole purpose.