Loading ...
Sorry, an error occurred while loading the content.

alternate implementations

Expand Messages
  • Steve Dekorte
    How do you guys feel about alternate implementations and very similar spin-off languages? Are they tend to be good for a language (faster implementations like
    Message 1 of 3 , Mar 12, 2007
    • 0 Attachment
      How do you guys feel about alternate implementations and very similar
      spin-off languages? Are they tend to be good for a language (faster
      implementations like IronPython) or do they dilute it (hundreds of
      LISPs)?

      - Steve
    • Christopher Diggins
      Multiple implementations will improve a language s popularity and usefulness if there is a clear and well-defined specification. Too many languages lack a
      Message 2 of 3 , Mar 12, 2007
      • 0 Attachment
        Multiple implementations will improve a language's popularity and
        usefulness if there is a clear and well-defined specification. Too
        many languages lack a proper specification, and the implementations
        end up implementing different languages. Consider how there are
        several dozen C implementations, which only contributes positively to
        the language.

        My two cents,
        Christopher Diggins
        http://www.cdiggins.com

        On 3/12/07, Steve Dekorte <steve@...> wrote:
        > How do you guys feel about alternate implementations and very similar
        > spin-off languages? Are they tend to be good for a language (faster
        > implementations like IronPython) or do they dilute it (hundreds of
        > LISPs)?
        >
        > - Steve
      • Chuck Esterbrook
        Nice question. I do feel a bit lost with some languages that have umpteen choices. With Python, I do not, but that could likely be due to any of these factors:
        Message 3 of 3 , Mar 12, 2007
        • 0 Attachment
          Nice question. I do feel a bit lost with some languages that have umpteen choices. With Python, I do not, but that could likely be due to any of these factors:
          * when I started with Python, there was only 1 (or there were 2, but I didn't know about Stackless)
          * there are not umpteen Pythons--there are just a handful
          * more than LISP, Eiffel, Ada, etc. there is a clear "canonical" or "standard" Python: the standalone one found at python.org

          I would say having a handful is good because of the choices offered. Python is a great example of that. There is a version for .NET and for Java, as well as Stackless and even Pyrex. This empowers people who choose or are forced onto a platform to still work Python into their project with relative ease. Or people may have special needs (Stackless, Pyrex) only met by having a special version of the language.

          If Io also ends up with a handful of implementations, I think it will do more good than harm. Especially if you document it on the main website as "here is the standard Io implementation... and here are alternate implementations with how they contrast and why you would choose them."

          Maybe it comes down to the ol' balance principle: not too much, not too little.

          -Chuck



          On 3/12/07, Steve Dekorte < steve@...> wrote:


          How do you guys feel about alternate implementations and very similar
          spin-off languages? Are they tend to be good for a language (faster
          implementations like IronPython) or do they dilute it (hundreds of
          LISPs)?

          - Steve


        Your message has been successfully submitted and would be delivered to recipients shortly.