Received: from SCRC-STONY-BROOK.ARPA by MIT-MC.ARPA 19 Jun 85 18:01:52 EST Received: from RIO-DE-JANEIRO.SCRC.Symbolics.COM by SCRC-STONY-BROOK.ARPA via CHAOS with CHAOS-MAIL id 258264; Wed 19-Jun-85 11:35:30-EDT Date: Wed, 19 Jun 85 11:35 EDT From: Kent M Pitman Subject: (SETQ PRIN1 'SPRIN1) To: GUMBY@MIT-MC.ARPA, FREDERIC@SU-CSLI.ARPA cc: MACLISP-QUERY@MIT-MC.ARPA In-Reply-To: <[MIT-MC.ARPA].549557.850619.GUMBY> References: The message of Fri 7 Jun 85 11:37-PDT from FREDERIC at SU-CSLI.ARPA Message-ID: <850619113535.6.KMP@RIO-DE-JANEIRO.SCRC.Symbolics.COM> Date: Wed, 19 Jun 85 04:38:49 EDT From: David Vinayak Wallace Subject: I can't find a maclisp manual; can you answer this To: KMP@MIT-MC.ARPA Message-ID: <[MIT-MC.ARPA].549557.850619.GUMBY> I doubt this was detailed adequately in the manual anyway. Date: Fri, 7 Jun 1985 11:37 PDT From: FREDERIC at SU-CSLI.ARPA To: gumby at SU-CSLI.ARPA cc: frederic at SU-CSLI.ARPA Re: maclisp Is there something I can do to get maclisp to always pretty print the things it outputs ? I tried (setq prin1 'sprin1), 'sprinter, setq'ing print, prin1, printer, and all sorts of combinations, but no result. Thanks, -freddie. Of curiosity, do you know what is defined to happen when an irresistable force meets an invincible object? Some people call PRIN1 because they want the standard printer, whatever the standard printer is. Others call it because they know what the standard printer does and they want exactly that regardless of any `enhancements' that others claim to have put in. Some people are forever coming up with code which has been carefully thought out and for which they can say "this piece of code is completely upward compatible with this other piece of code and will help me more so should simply replace all former uses of this other piece of code." While they're busily working this out, another group of people is always carefully working out pieces of code which work only if they can absolutely depend on the semantics of certain other pieces of code never changing at all. When these two groups of people try to share code, all kinds of problems have been known to arise. The PRIN1 variable, therefore just controls what happens when Lisp does toplevel read-eval-print. That is, it works roughly like: (DEFUN REPL (X) ;; Highly oversimplified. Ignores assignment of variables like +, *, etc ;; and calls PRIN1 rather than simulating a more PRINT-like behavior... (FUNCALL (OR PRIN1 'PRIN1) (EVAL (READ)))) This is the only thing (except maybe TRACE) which respects the binding of the PRIN1 variable. If you think for a minute, you'll realize that if -every- place that used PRIN1 could be made to get SPRIN1 instead, then SPRIN1 would really be in a bind (pardon the pun) because SPRIN1 uses PRIN1 as a subroutine and would thus infinitely recurse -- unless you introduced ultra-hairy special control variables and made sure they got properly rebound in break loops and ... The point is that Maclisp just wasn't designed with the idea in mind that you should be able to control things in the way you want to control them and that given the built-in assumptions that people have (reasonably) inferred from its original semantics, it's not the sort of change you can easily retrofit. Common Lisp fixes the problem by letting you just set *PRINT-PRETTY* to true. (SETQ PRIN1 'SPRIN1) is about as close as you're reasonably going to come. I suppose if you got in early enough before anything was loaded, you might be able to do something dangerous like: (LOAD (GET 'SPRIN1 'AUTOLOAD)) (DEFVAR *PRIN1* 'PRIN1) (REMOB 'PRIN1) (DEFUN PRIN1 (&REST ARGS) (APPLY 'SPRIN1 ARGS)) But I won't guarantee this won't somehow cause you grief way down the line. None of this solves your problem gracefully, but maybe it will put your mind more at ease if you decide to continue to search in vain for a magic solution...  Date: Tue, 23 Apr 85 03:45:12 EST From: Pandora B. Berman Subject: [KDO: Anybody there] To: ALAN@MIT-MC, kdo@SRI-KL cc: CENT@MIT-MC, MACLISP-QUERY@MIT-MC Message-ID: <[MIT-MC].465748.850423.CENT> Date: Tue, 23 Apr 85 03:38:57 EST From: Alan Bawden Subject: [KDO: Anybody there] To: CENT@MIT-MC Did I forward you his question originally? yes you did, and i'm pretty sure i answered it. however, just in case i missed, here goes again: Date: Mon 22 Apr 85 19:38:37-PST From: Ken Olum To: maclisp-query%MIT-MC at SRI-KL Re: Anybody there I sent a message asking how to get a new copy of TOPS-20 maclisp but received no response. hello out there. i distribute twenex maclisp. to get one (current distribution version is maclisp 2141 with complr 1128) you send me a 2400' tape. in a couple of months or so i send it back to you with maclisp written on it. default format is twenex dumper; if you want something else please ask. my address is: Maclisp Distribution NE43-906 545 Technology Sq. Cambridge, MA 02139  Received: from SRI-KL.ARPA by MIT-MC.ARPA; 23 APR 85 00:35:10 EST Date: Mon 22 Apr 85 19:38:37-PST From: Ken Olum Subject: Anybody there To: maclisp-query%MIT-MC@SRI-KL I sent a message asking how to get a new copy of TOPS-20 maclisp but received no response. Ken -------  Received: from SRI-KL.ARPA by MIT-MC.ARPA; 10 APR 85 21:59:44 EST Date: Wed 10 Apr 85 13:29:15-PST From: Ken Olum Subject: New version of TOPS-20 MACLISP To: maclisp-query%MIT-MC@SRI-KL I'd like to get a new version of TOPS-20 MACLISP, preferably on tape. Who do I talk to about this? Ken Olum Schlumberger Palo Alto Research 3340 Hillview Palo Alto, CA 94304 -------