Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site browngr.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!unc!mcnc!philabs!cmcl2!seismo!harvard!talcott!wjh12!foxvax1!brunix!browngr!dk
From: dk@browngr.UUCP (David Kantrowitz)
Newsgroups: net.ai
Subject: Re: Thus spake the DoD...
Message-ID: <10056@browngr.UUCP>
Date: Tue, 12-Mar-85 23:56:21 EST
Article-I.D.: browngr.10056
Posted: Tue Mar 12 23:56:21 1985
Date-Received: Fri, 15-Mar-85 02:11:32 EST
References: <313@wdl1.UUCP> <215@rtech.ARPA> <215@bu-cs.UUCP>
Reply-To: dk@sluggo.UUCP (David Kantrowitz)
Organization: Brown University Computer Science
Lines: 9
Summary: 


I don't understand. I thought the idea was to
  make a lisp archive file and whenever you want to create an executable
  lisp program, link in the appropriate routines from the library.  What
  difference should it make that you are using 70% of the routines in the
  library?  Actually, this way you can put many more things into the library
  than the average lisp (things that the Lisp user now has to specifically
  'load to get).  It's not as if linking takes very long, especially in
  comparison to loading compiled code *into* lisp.