But it's the way it is. This project is developed by users in their free time. People follow their interests (e.g. Florian improving the optimization, Jonas implementing the LLVM backend, me playing around in the parser).
And me being a pain in the collective backside of the core developers for the last 15 years or so :-)
Seriously: it's not just that writing a compiler is a big job, it's that /particularly/ when it extends an existing language it's not something that can really be done in one fell swoop: it has to grow organically and build on what came before it.
It's easy enough to build a language and compiler from scratch: that's exactly what Go does. Similarly, it's easy enough to build an operating system from scratch and to offer limited backward compatibility: that's what Windows NT did (and it took another ten years before a sufficient proportion of developers had been taught the error of their ways that it could go mainstream).
But to build something on a venerable language like Pascal that doesn't break backward compatibility is a massive job. And while I criticise Pascal and regret a couple of things left out of it (when compared with ALGOL) and now derided as "too C-like" I do have to defend FPC's core developers since by now their "gut feeling" that some change would either break the language or would require sufficient reimplementation that there would almost certainly be serious bugs- and some of us do actually use this compiler for serious work- has to be respected.
@hunghung: Don't feel bad. Get involved, learn the ropes, learn to distinguish between what's easy and what's not so easy. And just to put things in context, PascalDragon's port of the compiler so that it can use the low-level Windows NT API (normally available for device drivers etc.) is one of the simpler things he and his colleagues have done.
MarkMLl