Re: .NET Compiler for Interactive Fiction

"Marco van de Voort" <marcov@toad.stack.nl> wrote in message
> > I have thoroughly looked at the mud world and found it lacking in
> > flexibility and extensibility.
>
> Most are extensible via code in the central C code.


C is hardly a syntax that I would thrust at a potential writer. I'm
looking to make life easier. But the main point about muds is that
they have an interface that is sort of a soft multiuser
environment. I'd have to explain in detail what I envision for
multi-user IF, but that's another topic altogether and not appropriate
for comp.compilers.


> > One of my design goals is to maintain the ease of development that the
> > current Interactive Fiction languages provide.
>
> I'm just curious why that has anything to do with a .NET or JVM
> backend, or why the MUD systems don't suffice. It seems you have
> already made up your mind to target the fashionable platforms.


The .NET framework, to me, is infinitely more accessible than C or any
of the standard C libraries. And it also has an enormous support
mechanism built in.


> >>From there, I want to open the genre to extensions like database, socket
> > programming, xml, and other types of features that platforms like Java
> > and ..NET have.
>
> What do they have? Relative unportability (compared to the plain C mud
> systems)


I have already determined that portability is a lower goal although
Mono may provide it in the future. The .NET framework has an enormous
amount of pre-written code. So I build a system for IF and then
someone comes along and wants to add some feature. They're not limited
to the system I've defined nor the language I've chosen. I'd probably
build all of my work in C#, but someone else may choose to use
Perl.NET or VB.NET. There's such a huge upside to giving people that
sort of flexibility.


> > I wouldn't so much mind Java except that I want to create my own
language
> > syntax and although I could target the JVM, the .NET CLR and Framework
seem
> > so much more refined to me.
>
> Why? Do you need to sell it, and have to acronym-overload some manager
> with fashionable systems?


Actually, I don't care to sell the compiler I intend to build, nor any
of the related class libraries that would make up the .NET IF
system. That would all be open-source. But I would have the ability to
then make very sophisticated games and sell them. This is of course
directed at a Windows world, but I'm okay with that. The current IF
platforms are great for strict traditional IF games. Some have added
limited functionality for html, graphics, sound, but they still
prevent anyone from truly expanding their models. Something build
properly in .NET wouldn't prevent anyone from doing anything.


It sounds like there's a bit of a platform-wars type conversation
starting here and I'd rather table that or take it somewhere else. I'm
okay arguing about the good and bad of .NET/Windows, but not
here. Suffice it to say that this is my chosen method and yes, I have
looked very thoroughly at muds, C, and other languages like Python. My
primary goal is to build an open-ended IF compiler that is extensible
via the .NET platform.


David C.

Return to the comp.compilers page.
Search the comp.compilers archives again.