alt Hacker NewsThis is a nice and unexpected release, thanks for writing it. Getting a RPG endorsement is great. I just finished reading his foreword and skimming the table of contents and bibliography from the preview. I'd have liked to see a sample of a middle chapter to really see how technical and deep it gets (e.g. Land of Lisp gives its chapter 8 as a sample which I think is very representative for that book). But I plan to get this book regardless -- just not right now.
The back blurb hints that expert systems might be mentioned, but how much? No one ever seems to go much into their implementation or usage.[0] It also mentions writing some JS, which I guess is part of chapter 5, I wonder if that was a publisher request. (My favorite take on that subject in recent years is https://github.com/jart/sectorlisp)
Would it be fair to say this is mainly a history told through the lens of AI and PL research?
Amusingly I think part of me is already setting myself up for some disappointment -- it seems too short with too few references! But it's good to have a Lisp history book like this looks to be and I'm sure I'll learn things from it, and the promise of more RPG writings inside is enticing. Besides, any complete telling would take multiple books. (There's so much of historical interest locked up in proprietary applications and companies with their own histories, and so many papers published, there's also so much that can be dug through in the standardization mailing list (and other lists, like emacs) archives[1], the SAIL archives[2], the Xerox PARC archives[3], the CMU archives[4], and the many undigitized things sitting in boxes at the computer history museum...[5])
[0] Norvig's PAIP gives a small taste, one of the files: https://github.com/norvig/paip-lisp/blob/main/lisp/mycin-r.l... And a book about a particular system, MYCIN: https://www.shortliffe.net/Buchanan-Shortliffe-1984/MYCIN%20... And a short video: https://youtube.com/watch?v=a65uwr_O7mM
[1] http://ml.cddddr.org/ and http://cl-su-ai.lisp.se/
[3] The url I had before is down... I made a local copy but https://archive.org/details/2014.01.ftp.parc.xerox.com might be the same content
[4] https://www.cs.cmu.edu/afs/cs.cmu.edu/project/ai-repository/...
[5] Even in the earliest Lisp reports like https://www.researchgate.net/publication/42766480_Artificial... there are interesting things mentioned like a two-move checkmate program or "Other projects on which work continues include the Advice Taker, visual pattern recognition, and an artificial hand." Multiple times I've tried to track down those sorts of things mentioned in really old papers only to hit dead-ends on so many of them. Sometimes things were embellished, or were abandoned, or were just lost to time, and sometimes there's an undigitized box at the museum that might contain printouts etc. (There might be MYCIN source code, even.)
Yes, the book doesn't do Lisp justice, it is too short. But at the current volume, I had something at least passing muster; I'm toying with extending it but that depends on feedback/success/etc. There are a lot of Lisp implementations I haven't mentioned (or dealt with in the depth they deserve), there is a lot more to say about the sort of AI work that was (and, I think, is) done with Lisp, etc. And I have written it with a "general techie" audience in mind more than "I'm already a hardcore Lisper", I will probably disappoint the latter group with a lack of depth. I haven't aspired to LOL or PAIP or similar great works.
It's a history through a lens, but if there is one I'd say "MIT/Stanford" as a central axis rather than a field of reesarch.
And Javascript? My own choice. The amount of "language" I needed was very small and I actually like the very minimalistic (lisp-y?) sort of Javascript you can write these days if you just ignore most of its history. It's accessible, that was more important to me than anything else - one of the few concessions where I wanted to make things digestible to as wide an audience as possible in a language that was good for the problem at hand. Strangely enough, it worked very well (I think).
I heard your (and others') request for a better sample chapter than the intro that Amazon shows, I'll put it on the site as soon as possible.