[weasel]

zzo38, on Aug 9 2008, 04:24 PM, said:

No, I mean that any developers of MZX could look and see if my module is good, and then compile it in so that it will already be compiled when someone download MZX it will already be in.

But that means it'd have to be compiled for every port, though, right? Hence what Exo is saying about how something like this is unportable? (I hope I got that right this time.)

[zzo38]

weasel, on Aug 9 2008, 04:28 PM, said:

But that means it'd have to be compiled for every port, though, right? Hence what Exo is saying about how something like this is unportable? (I hope I got that right this time.)

Aren't all the modules in MZX compiled for every port anyways?

[Exophase]

zzo38, on Aug 9 2008, 07:24 PM, said:

No, I mean that any developers of MZX could look and see if my module is good, and then compile it in so that it will already be compiled when someone download MZX it will already be in.

Your "module" that does what? If you want to develop something for MZX then submit something to the maintainers, I am 100% failing to see what this has to do with Lua or Forth or anything.

[ajs]

Since MZXLua isn't finished and isn't atm mergeable, it's just too hypothetical to be discussing larger changes built on top of it at this stage. As the guys here have already stated, in my view Lua is enough already, and things like LuaForth are basically fine (if they remain in PURE Lua) but loadable C modules are completely unacceptable. Nightwatch even disabled the module loader in Lua for MZXLua at my request.

So IF Nightwatch reappears and/or somebody else takes the reins (I'm not going to) and MZXLua gets to a level where me, Exo, and the other guys are happy with the core changes, THEN we can discuss shipping support for other languages (implemented only in Lua or some small built-in code). Discussing anything further at the moment is just a waste of everybody's time.

--ajs.

[Frobozz]

zzo38 said:

Forth is a decent language, and very powerful, one of the best in my opinion, actually. The fact that it is old means nothing (some people don't like Forth, but a lot of people still like Forth today, I'm not the only one!). Forth is very good design and you can implement more control structures and stuff directly in Forth. Program languages such as Lua, JavaScript, C, BASIC, etc are not so versatile as that.

I'm not that great in programming, but last I knew the currently provided structures in most modern languages as well as C++ are more than enough. Who cares if you can define some off-the-wall new programming construct to add to a language? The problem you will run into is when you give the code to someone else. Just look how confused WildWeasel was. Granted I'm not sure about his programming talent, but he was confused by the code. That alone is reason enough to avoid stupid features like this. How would it be if anytime someone modified a Forth program, they had to analyze in detail every new "feature" someone else added so their program would work. It would be like learning a new language every time you wanted to look at a piece of source code.

zzo38 said:

You don't really need Lua, but Robotic is very limited. Adding something else, such as Forth or Lua would improve it (some people wanted to put Lua in MZX for this reason, I think, but please keep Robotic as well!).

If Lua isn't necessary, then Forth certainly isn't either. Obviously the community has made do with Robotic up to now and they will continue to make do with it. I'd imagine a good number may not even use Lua if it were added.

zzo38 said:

But still a lot of things are missing if you just do that, there are other things that should be improved in MZX as well, such as the ability to edit the internal structures, edit the pieces on the board in raw mode

Such a thing seems pointless, would require a major rewrite, and even after all that I'm not sure how you'd pull it off. It'd be like rewriting the source code each world. Even if it could be done, I don't see any benefit. We already have a lot of built-ins and yet most people only use a handful (customfloor, customblock, etc).

This post has been edited by Frobozz: 14 August 2008 - 06:55 PM

[RoSS]

I think we should just break down and port .NET over to mzx.

You said you wanted flexibility? You got it. You said you wanted robots to use their originally-intended language, VB.NET? You got it. You want MZX to eat other GCS's children for breakfast? You got it.