Links to the three unlinked papers:<p><i>Compiler Optimization Based on Viewing LAMBDA as RENAME + GOTO</i><p>ftp://publications.ai.mit.edu/ai-publications/pdf/AITR-474.pdf<p><i>Debunking the "Expensive Procedure Call" Myth, or Procedure Call Implementations Considered Harmful, or LAMBDA, the Ultimate GOTO</i><p><a href="http://dspace.mit.edu/bitstream/handle/1721.1/5753/AIM-443.pdf?sequence=2" rel="nofollow">http://dspace.mit.edu/bitstream/handle/1721.1/5753/AIM-443.p...</a><p><i>Design of a Lisp-based Processor</i><p><a href="http://dspace.mit.edu/bitstream/handle/1721.1/5731/AIM-514.pdf" rel="nofollow">http://dspace.mit.edu/bitstream/handle/1721.1/5731/AIM-514.p...</a>
SICP now free online too (and has been for quite some time)... <a href="http://mitpress.mit.edu/sicp/full-text/book/book.html" rel="nofollow">http://mitpress.mit.edu/sicp/full-text/book/book.html</a>
Previous discussion about the relative merit of these papers versus the relative number of times a link to them floats about: <a href="https://news.ycombinator.com/item?id=599306" rel="nofollow">https://news.ycombinator.com/item?id=599306</a>