With the permission and assistance of the author himself, Bernie Cosell, I have added the original Lisp Eliza to the Eliza Generations collection. Cosell wrote this Lisp version of Eliza at BBN in the mid-late 1960s. (Weizenbaum’s original was written in about 1966 in SLIP, a long-dead Fortran-based symbol processing package.)
Thanks to Peter Seibel for connecting me with Bernie Cosell.
!!YOU CAN HELP!!
One way that you can help is by writing to Deborah Cotton (firstname.lastname@example.org
) at the ACM permissions office and encourage them to open source Weizenbaum’s paper, which is still inaccessible under copyright protections.
Second, I’m hoping to create a “perfect” OCR of this code and then macrify it to run in a modern Lisp with as little modification as possible. You can help create the codebase for this by choosing a single page at random from among the 48 TIFFs, manually entering the code as precisely as possible (including indentation) into a text file, and then emailing it to me: email@example.com
. If you decided to do this, here are a few details worth attending to: So that we get good coverage, please really choose a TIFF file at random, e.g., via the moral equivalent of (1+ (random 48)). Please don’t just OCR the TIFF file; I’ve already tried this with very high end OCR tools, and they make terrible encodings of this sort of thing! It would help me do the reconstruction if you put the name of the TIFF file in a leading comment. Finally, if you would like to be explicitly acknowledged for your efforts, please include a comment line for yourself as well. The codebase will be released on github or some such public location. Then you’ll be able to help actually hack it!
Finally, if you know of open source Elizas, in any language, roaming around the net, please send me links to them so that I can update the “more recent” section of the page.