Path: csiph.com!usenet.pasdenom.info!news.albasani.net!news2.arglkargh.de!news.theremailer.net!frell.theremailer.net!anonymous From: Fritz Wuehler Comments: This message did not originate from the Sender address above. It was remailed automatically by anonymizing remailer software. Please report problems or inappropriate use to the remailer administrator at . Identifying the real sender is technically impossible. Newsgroups: comp.lang.forth Subject: Re: Which standard? References: <7x1uokcsig.fsf@ruckus.brouhaha.com> Message-ID: <52114e1d4e0aabf7481a64843ea0bb8a@msgid.frell.theremailer.net> Precedence: anon Date: Fri, 23 Mar 2012 19:20:15 +0100 Mail-To-News-Contact: abuse@frell.theremailer.net Organization: Frell Anonymous Remailer Xref: csiph.com comp.lang.forth:10370 Paul Rubin wrote: > Nomen Nescio writes: > >> Do not underestimate the work to write a decent code generator. > > > > I haven't estimated it, much less underestimated it. The best part is I can > > always give up ;-) I don't think the code generation is going to be > > difficult. I think making a usable Forth that has practical value on the > > targets is going to be the hard part, it will require a lot of system > > interface and extensions and that's a lot of code. > > Forth implementations tend to be quite small, so by definition they > can't require a lot of code. What about the implementations that don't tend to be small? What kind of runtime support do you need to make a Forth usable for what you want to use it for, on the platform you use it? I don't have libc on my platform and I wouldn't use it anyway. I need to supply a complete runtime with storage management, I/O, program management, recovery, and much more to make a Forth that I could use on my target. Without you knowing the details of what I want to accomplish, and probably not knowing anything about the platform or environment, I think your generalization is not useful. > The traditional code generation technique going back to the 1970's has > been threaded interpreters, which are easy to implement and which (not by > coincidence) map very naturally to the structure of Forth. I intend to look into this but if I write anything the first round is going to be written in assembler with an assembler runtime and will be written to support multitasking and other facilities relevant to the platform.