7 ms·

Yes , but the young contenders already exist. If Nim gets more traction , over time it may replace Python.

by jaco8 9y ago

analog31 9y ago
I think about that once in a while, since I've gone all-in on Python, and here's how I've come to grips with it. Even if Python eventually declines (as did Pascal, HyperCard, and Visual Basic), it will still have made me a better programmer, and the continual improvement of languages means that the young contenders have to prove themselves better than Python. So, bring 'em on.
alcalde 9y ago
Nim's designer is too enamored with Delphi and has chosen to borrow many of Delphi's bad ideas for nostalgic reasons. This adds more ambiguities, more Perl-like symbol use, more reliance on an intelligent IDE, etc. Python is about the importance of simplicity and readability. I don't see Python devs abandoning Python for Nim any time soon. The ones who would have been tempted by Nim have already been tempted away by Go.
- dom96 9y ago
  What bad ideas from Delphi did Nim borrow? Presumably you are referring to the way imports are done in Nim, i.e. the fact that all symbols are imported into the current module by default (the equivalent of `from module import *` in Python). Nim could certainly follow Python's lead, but that would limit the language: Nim supports UFCS and operator overloading which requires modules to be imported this way. You can of course feel free to use `from module import nil` in Nim to enforce module name prefixes for each procedure call. But keep in mind that there is no risk of ambiguities, the compiler statically checks everything and refuses to compile your code if there is an ambiguity (at which point you will need to resolve it by prefixing with the module name). I would be interested to hear more about other things that you think are a bad idea. Could you elaborate a bit?
goatlover 9y ago
Why would Nim be a better choice over Julia for data science stuff?
coldtea 9y ago
Don't hold your breath. Nim doesn't have the kind of ecosystem, and hasn't shown signs of having one.
- dom96 9y ago
  I'm curious what the ecosystem for Python looked like in ~1999. Did it show signs of having the ecosystem it has now in those years? I'm certain that almost everyone dismissed it due to its syntax (there are still people out there that dismiss it because of that).
  - nimmer 9y ago
    > what the ecosystem for Python looked like in ~1999 Very small. The language was not popular and many people claimed a community project could not overtake languages backed by companies. Same for the Linux kernel.
    - dom96 9y ago
      Thank you for supporting my point. However, based on your nickname I wouldn't be surprised if you were a little biased :) I would actually love to read some forum/mailing list posts about this. If anyone has any links handy (or knows where to find them) then please post them.
  - coldtea 9y ago
    I actually used Python commercially in 1999. The "killer app" then was an app server + document store called Zope. Things were smaller than today, but Python was already gaining popularity for scripting. Circa 2001 or so it also had good Gtk and Qt bindings -- there were even some KDE-inspired satellite commercial companies catering to Python/QT. The discussions in Python land then were about the mythical Python 3000 -- the next, re-imagined version of Python that would fix issues, etc. That was eventually Python 3. Perl was still king for scripting/admin stuff but losing momentum, and Ruby was mentioned as a nice alternative to Python, but with much smaller use at the time (and even today I think).
- deleted 9y ago
jeremyjh 9y ago
It had a flash a few years ago when it changed names but overall does not seem to be going anywhere hype-wise. https://trends.google.com/trends/explore?date=today%205-y&q=nim%20language,nimrod%20language https://trends.google.com/trends/explore?date=today%205-y&q=...

Yes , but the young contenders already exist. If Nim gets more traction , over time it may replace Python.

by jaco8 9y ago

Yes , but the young contenders already exist. If Nim gets more traction , over time it may replace Python.

I think about that once in a while, since I've gone all-in on Python, and here's how I've come to grips with it. Even if Python eventually declines (as did Pascal, HyperCard, and Visual Basic), it will still have made me a better programmer, and the continual improvement of languages means that the young contenders have to prove themselves better than Python. So, bring 'em on.

alcalde 9y ago

Nim's designer is too enamored with Delphi and has chosen to borrow many of Delphi's bad ideas for nostalgic reasons. This adds more ambiguities, more Perl-like symbol use, more reliance on an intelligent IDE, etc. Python is about the importance of simplicity and readability. I don't see Python devs abandoning Python for Nim any time soon. The ones who would have been tempted by Nim have already been tempted away by Go.

dom96 9y ago
What bad ideas from Delphi did Nim borrow? Presumably you are referring to the way imports are done in Nim, i.e. the fact that all symbols are imported into the current module by default (the equivalent of `from module import *` in Python). Nim could certainly follow Python's lead, but that would limit the language: Nim supports UFCS and operator overloading which requires modules to be imported this way. You can of course feel free to use `from module import nil` in Nim to enforce module name prefixes for each procedure call. But keep in mind that there is no risk of ambiguities, the compiler statically checks everything and refuses to compile your code if there is an ambiguity (at which point you will need to resolve it by prefixing with the module name). I would be interested to hear more about other things that you think are a bad idea. Could you elaborate a bit?

goatlover 9y ago

Why would Nim be a better choice over Julia for data science stuff?

coldtea 9y ago

Don't hold your breath. Nim doesn't have the kind of ecosystem, and hasn't shown signs of having one.

dom96 9y ago
I'm curious what the ecosystem for Python looked like in ~1999. Did it show signs of having the ecosystem it has now in those years? I'm certain that almost everyone dismissed it due to its syntax (there are still people out there that dismiss it because of that).
- nimmer 9y ago
  > what the ecosystem for Python looked like in ~1999 Very small. The language was not popular and many people claimed a community project could not overtake languages backed by companies. Same for the Linux kernel.
  - dom96 9y ago
    Thank you for supporting my point. However, based on your nickname I wouldn't be surprised if you were a little biased :) I would actually love to read some forum/mailing list posts about this. If anyone has any links handy (or knows where to find them) then please post them.
- coldtea 9y ago
  I actually used Python commercially in 1999. The "killer app" then was an app server + document store called Zope. Things were smaller than today, but Python was already gaining popularity for scripting. Circa 2001 or so it also had good Gtk and Qt bindings -- there were even some KDE-inspired satellite commercial companies catering to Python/QT. The discussions in Python land then were about the mythical Python 3000 -- the next, re-imagined version of Python that would fix issues, etc. That was eventually Python 3. Perl was still king for scripting/admin stuff but losing momentum, and Ruby was mentioned as a nice alternative to Python, but with much smaller use at the time (and even today I think).
deleted 9y ago

jeremyjh 9y ago

It had a flash a few years ago when it changed names but overall does not seem to be going anywhere hype-wise. https://trends.google.com/trends/explore?date=today%205-y&q=nim%20language,nimrod%20language https://trends.google.com/trends/explore?date=today%205-y&q=...