• 0

Can someone explain the needs for new languages?


Question

Well after reading the F# thread, it makes me wonder, what exactly is the need for creating new languages? If there's a need for new features can't they just be implemented into an existing language?

Link to comment
Share on other sites

4 answers to this question

Recommended Posts

  • 0
from the business perspective, if you develop  a lanaguage that is widely used and successful = $$$$

584807954[/snapback]

Ah, but F# is by Microsoft and with the .NET framework, so chances are they won't charge for it.

I haven't looked over F#'s feature set, but if it's similar to C# (which I doubt it is), then I don't see the point. Microsoft tends to keep their code centered around C#.

Link to comment
Share on other sites

  • 0

Languages are tools. You use tools to solve problems. If you can develop a language that helps a developer better express a solution, it's mighty helpful. I have to say that LISP is quite good at helping you focus on the solution and not the syntax. I can see why F# is cool, though I doubt I'd use it simply because my company is solely a C# shop. Functional programming is a powerful language feature.

Remember that programming languages have two goals. First, to abstract the hardware, and second, to abstract human language. I don't think a perfect balance has been developed yet. I'm fine, however, with the C-type languages just through constant use. I think, ideally, scientists would like to develop a natural language or visual language that didn't sacrifice performance for simplicity.

I kind of think that a short coming of object oriented programming is that you have to change the way you think. People don't inherently think in terms of objects/interfaces, so it becomes a chore to begin to think like that. Thinking procedurally comes natural, since we all do procedural thinking every day(e.g. get up, shower, eat breakfast, get in car, drive to work, scream at people on the road, etc...) Behavioral thinking, core to object thinking, is relatively new and requires a good deal of examination of the "problem." This is why I tend to think that true OO programming isn't being done that often. I don't know if it's even possible to do true OO completely.

I digress. Adding features to an existing language has been done, not necessarily for the better. It can complicate things. Sometimes it's better just to start over and develop a clean language. For instance, look at D ( http://www.digitalmars.com/d/overview.html ). The author makes a good case for a new language.

Link to comment
Share on other sites

  • 0
Languages are tools. You use tools to solve problems. If you can develop a language that helps a developer better express a solution, it's mighty helpful. I have to say that LISP is quite good at helping you focus on the solution and not the syntax. I can see why F# is cool, though I doubt I'd use it simply because my company is solely a C# shop. Functional programming is a powerful language feature.

Remember that programming languages have two goals. First, to abstract the hardware, and second, to abstract human language. I don't think a perfect balance has been developed yet. I'm fine, however, with the C-type languages just through constant use. I think, ideally, scientists would like to develop a natural language or visual language that didn't sacrifice performance for simplicity.

I kind of think that a short coming of object oriented programming is that you have to change the way you think. People don't inherently think in terms of objects/interfaces, so it becomes a chore to begin to think like that. Thinking procedurally comes natural, since we all do procedural thinking every day(e.g. get up, shower, eat breakfast, get in car, drive to work, scream at people on the road, etc...) Behavioral thinking, core to object thinking, is relatively new and requires a good deal of examination of the "problem." This is why I tend to think that true OO programming isn't being done that often. I don't know if it's even possible to do true OO completely.

I digress. Adding features to an existing language has been done, not necessarily for the better. It can complicate things. Sometimes it's better just to start over and develop a clean language. For instance, look at D ( http://www.digitalmars.com/d/overview.html ). The author makes a good case for a new language.

584809509[/snapback]

yep, I like very much your explain and agree too, maybe I try D language for know it and see your functions more deep, thanks for the link (Y)

Link to comment
Share on other sites

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.