From dex@SYSLINK.MCS.COM Fri May 28 21:38:31 1999 X-Digest-Num: 152 Message-ID: <44114.152.905.959273824@eGroups.com> Date: Fri, 28 May 99 21:38:31 From: dex@SYSLINK.MCS.COM Subject: (no subject) Have you considered Forth? You can easily redefine anything in it. In fact, writing a program is just defining a word. That word is then available for use in defining other words; and so your dictionary grows. It combines the best features of compilers and interpreters, and when used skillfully it can be more compact than assembly language and run faster. There is a safeguard against redefining things too deep into the structure of the language. When you know how, you can give a simple command to move the safeguard deeper and then redefine it anyway. Alternatively, you can move it higher and ensure that your earlier work won't be trashed by mistakes in your later work; or clear everything down to the current level of the barrier, and load different modules on top. Now, that's mutable. There are commercial versions with all sorts of custom add-ons, but FIG Forth has been public domain since 1978, and was implemented on just about every processor family out there, mainframe and micro, with full source available on any of them from FIG (the Forth Interest Group). It contains its own Operating System, but has often been ported to run under others. I'd give better contact info, but I haven't so done much programming since the 1980s, and lost touch with the community. I later lost my archival Forth stuff in a flood. Somebody ought to be able to track it down. The latest standard version of Forth the last time I was active in that world was Forth-83. > From: William Tanksley > > A guy named Sassenrath is presenting a new computer language named > Rebol (www.rebol.org). It's an interesting looking language, but > I was especially facinated by the fact that it's very mutable -- > you can easily replace any and all of the syntactical elements > with ones of your own choosing. > > It seems to me -- although I don't know either Rebol or Lojban > very well -- that Rebol might serve well as a test vehicle for > a Lojbanic computer language. Its parser is just a little bit > too unsophisticated (we'd want a full spaces-irrelevant parser > for the real thing), and its source isn't available (but it > doesn't cost money). > > If an enhanced Rebol turns out to look interesting, perhaps it'll > provide a guide towards implementing a real Lojban computer > programming system. > > -- > -William "Billy" Tanksley > "But you shall not escape my iambics." > -- Gaius Valerius Catullus