Is it even worth trying to improve LSL

The comments on Mono in the recent bytecode thread make me wonder if it's even worth trying to improve LSL.

Adding elif to LSL seems straightforward. Changing the grammar is easy, but I haven't yet studied the code generation. Still, it ought to be a relatively easy change.

But if the move to Mono means that LSL may someday be replaced by something closer to C# or Python, then it may not be worth the effort. If all of parsing is moved to the server-side, then it may not be worth the effort. Finally, if LL is reluctant, as a matter of policy, to accept user-submitted changes to the grammar, then it again wouldn't be worth the effort.

So, is enhancing LSL worthwhile? Or a waste of time?

given LL's unknown deadlines for actually getting mono on the grid, I'd say 'yes, probably'

Even if LL do release a c# variant some years from now, LSL will not go away.
Some of us have just learnt to code in LSL, if we were suddenly told that you now have to learn and yet another, and arguably more complex, language, a lot of us will become unproductive for ages as we struggle to learn the language in the spare time that we would normally spend scripting. That is the few that would even make the effort to learn it; most would just walk away if faced with that prospect. So LSL is here to stay, C# or other languages that can spit out mono byte code may be supported, but will only be used a top few scripter’s, at least initially. LSL will continue to be the platform of choice for beginners, or people who want to do quick and easy functions in SL.

granted LSL is a very simplistic language, but it contains most of the key concepts, even assuming LL gave us another input language (mono won't change what we write with just make it faster on the server...mostly) the only thing that changes is the words, the structure remains basically the same, learning the new syntax is slightly harder than comparing a new function to the old familiar one.... your second programming language is always the easiest.... unless it's assembly... they call those people masochists...

Count me in on that one It is my favourite language, I still think in 6809 byte code almost 30 years after I last had to program an embedded controller. I prefer to have direct control of the virtual machine, so even C#, in fact especially C#, I see as just more restrictions between my ideas and the implementation.

Not that I would ever want to attempt to control the physics engine with assembler though so I tend to see LSL as a set of library functions to handle the higher level math. It is the lack of control over the memory and its access that gets to me.

But now that I have got used to it, I am beginning to regard LSL as one of the beter languages I have worked with, at least in terms of fun per line of code.

... its taken them years to get mono (a backend ONLY project) there isnt enugh time in the world for them to switch to python or whatever

and nevermind its been stated a billion times that LSL is not going to be replaced by anything just because they switched the interpreter to mono

If you can't learn C# in a weekend, you are in the wrong field.

(Yes, it is that easy, and yes, I am a professional programmer.)

It may be a purely academic question. LL doesn't want to change LSL until after Mono is done, and then the compilation is going to be done server-side, so changing it will probably fall more along the lines of begging LL to find the time, resources, and motivation to do it internally rather than submitting a proposed change to them for approval and integration. Cynical, I know. :-/

And I am an electronics engineer, not a programmer. But yes I do know C, and Java and C++ and C#, I just avoid them out of preference. Apart from which, out side of SL, I don’t do my own coding, I generate the algorithm, hand it over and then test the finished design.

The learning C# over the weekend comment is absolutely true though. For anticipation of SL possibly supporting the language down the road, and because it's about time I brush up on a universal programming language -- the demand for C# programmers is growing fast -- I picked up a few books. What I quickly realized is, I already know this language. Of course, a few things are different, but I've already done a lot of work with LSL and with PHP, so most of the concepts are already familiar. This isn't like jumping from Spanish to Russian.

Admittedly, support for C# would probably make competing more difficult for anybody who chooses to stick to LSL since C# is much more powerful, but why should this hold back the more ambitious scripters? A more versatile language and .NET support could usher in a new generation of in-world applications, something Second Life sorely needs.

I think we put up with LSL because there is no alternative. The lack of associative arrays annoys me no end. The whole thing feels like a hack.

I love LSL.

@void:

Mono is closer to release than you may think it is.

Also, they *have* gotten C# scripts to run on Mono. There is just not a user friendly way to do it yet. They had to employ a number of tricks for it to work.

I'd be happy enough with LSL if they added stuff like switch/case and break/continue. Arrays would also be nice but their lack hasn't annoyed me nearly as much as the lack of some basic control statements.

<- Writes in VB6, VBA, C, VC++, Java, and is randomly learning LSL for the LULZ. (^_^)

From what I see... Mono is not getting rid of LSL, but, giving the option to write in different languages in parallel to allow for higher performance scripting for more advanced scriptors. This is very normal for an open environment and really shouldn't be something to be afraid of. (=_=)

Just like sculpted prims, they're a great alternative to regular prim builds in ways of creating detail, reducing prim count, or simply making things impossible with the default prim collection... But, that's only if you know how to make them and have gotten past that learning curve. (^_^)

Where is the wiki/FAQ/transcript that clearly states that Mono will mean instant death for all LSL scripting? (o.o)

You think like my Dad =)

The 6502/10 was his specialty. Cross-platform game development in the olden days was a piece of cake he reckons.

Damn, now I feel old Never did work on Games mostly boring stuff like what feed mixture a cow gets in the feed booth according to the amount of milk produced ..... Hell come to think of it, in those days it was all so new it was all a game

=)

He tried to teach me assembly once. My program went:

LDA "H"
JSR $FFD2
LDA "E"
JSR $FFD2
LDA "L"
JSR $FFD2
LDA "L"
JSR $FFD2
LDA "O"
JSR $FFD2

I was quite pleased with myself. He just rubbed his head, sighed, went out and bought me a computer with Windows on it and said he'd teach me Visual Basic instead. I was even more pleased with myself =)