Interesting model of the brain https://archive.org/details/byte-magazine-1979-08/page/n69/m...
Proving that scarcity forces you to think ahead.
Could be very useful for AI/ML models today.
Byte Magazine: LISP (1979)
21–30 of 109 posts
Re: Byte Magazine: LISP (1979)
#22The first article in this issue of BYTE has a very interesting characterization of Lisp that I have not come across before. I mean, famous quotes like "Lisp is a programmable programming language" by John Foderaro and "The greatest single programming language ever designed" by Alan Kay are often mentioned in articles about Lisp. But in this issue of BYTE, the article "An Overview of LISP" by John Allen at page 10 has…
This exactly matches my thoughts. It seems that machine language and LISP are the only two languages (that i know anyway) where code and data are fundamentally the same kind of thing.
Prolog and Clojure (effectively a LISP dialect) probably also qualify, and more modern: I think that in a way JSON also qualifies.
Re: Byte Magazine: LISP (1979)
#23Earlier quoted context omitted.
This exactly matches my thoughts. It seems that machine language and LISP are the only two languages (that i know anyway) where code and data are fundamentally the same kind of thing.
There are many languages that are dynamic and let you generate code and run it inside the program. For instance in Java you can generate bytecode for a class and run one of the methods. In FORTH you can write new "words" (roughly functions) that are the same as the built-in words. It is mainly languages built around AOT compilation like C, Pascal, FORTRAN and such that completely separate code and data. (I guess thou…
Code is data isn't just about emitting and running code. It's also traversing your code as the data structure it is and changing it as you see fit.
Re: Byte Magazine: LISP (1979)
#24Just this week while going through the book _Ideas That Changed the World_ I read the original paper on LISP[1]. I have to say this magazine article reads a bit cleaner. But that's kind of the fun of the book -- learning to decode the excited yet formal language of computational innovation over time.
The book traces out what are effectively conversations across decades (or centuries if you include Aristotle). So when you see McCarthy name drop Church's lambda calculus, you know how it ties into a conversation about Hilbert's decidability problem, and that self-reference ("recursion") is a fundamental tool.
And from the perspective of now, you see that this is the first encounter with automatic garbage collection, describing a simple mark and sweep algo. The BYTE magazine covers a more advanced garbage collector, Lambdino, which assumes a massive amount of familiarity with LISP and internals than the previous article comparing LISP and LOGO.
Re: Byte Magazine: LISP (1979)
#25Starting on page 154 a now-familiar new product is introduced, and compared to Lisp as such:
> Returning to the LISP theme of our current issue, Visi- Calc is an example of a tree-oriented parallel data struc- turing problem for which LISP is a most appropriate lan- guage of expression. Due to a lack of availability of LISP as a software development tool for personal computing hardware, its authors did not use LISP. They also had to make a number of compromises and tradeoffs as a result of the small size (eg: 16 K to 48 K bytes) of the main memory of personal computers. But they did use many of the tree concepts of artificial intelligence research. This provides us with the ultimate example of the relevance of LISP-like languages and approaches to personal computing: one of the most generally useful new user software tools for small machines, Visi-Calc, tackles just the sort of problem for which LISP is an appropriate tool of expression.
Re: Byte Magazine: LISP (1979)
#26Earlier quoted context omitted.
This exactly matches my thoughts. It seems that machine language and LISP are the only two languages (that i know anyway) where code and data are fundamentally the same kind of thing.
There are many languages that are dynamic and let you generate code and run it inside the program. For instance in Java you can generate bytecode for a class and run one of the methods. In FORTH you can write new "words" (roughly functions) that are the same as the built-in words. It is mainly languages built around AOT compilation like C, Pascal, FORTRAN and such that completely separate code and data. (I guess thou…
Even "languages built around AOT compilation" are increasingly blurring the phase distinction between "ahead-of-time" and "at runtime", via increased use of metaprogramming and constant evaluation facilities. We may end up with a state of things where even some "AOT-language" programs are written to do much of their work at compile time, perhaps adding more and more of those facilities that were formerly allowed only at "runtime".
Re: Byte Magazine: LISP (1979)
#27The first article in this issue of BYTE has a very interesting characterization of Lisp that I have not come across before. I mean, famous quotes like "Lisp is a programmable programming language" by John Foderaro and "The greatest single programming language ever designed" by Alan Kay are often mentioned in articles about Lisp. But in this issue of BYTE, the article "An Overview of LISP" by John Allen at page 10 has…
This exactly matches my thoughts. It seems that machine language and LISP are the only two languages (that i know anyway) where code and data are fundamentally the same kind of thing.
Forth is the exact same thing.
Re: Byte Magazine: LISP (1979)
#28A
Re: Byte Magazine: LISP (1979)
#29Love these ads.
There were almost as many computer companies back then as there are crypto currencies now.
For a mere $1595, you got a 16k computer with microsoft basic but the best part is that it comes with 2 Z80 chips. Not sure how much a Z80 chip cost but surely the user would prefer an extra 8k or whatever. Best part is that you get to learn how the computer works by building it.
You can see why user testing is so important.
Re: Byte Magazine: LISP (1979)
#30Love these ads. There were almost as many computer companies back then as there are crypto currencies now. For a mere $1595, you got a 16k computer with microsoft basic but the best part is that it comes with 2 Z80 chips. Not sure how much a Z80 chip cost but surely the user would prefer an extra 8k or whatever. Best part is that you get to learn how the computer works by building it. You can see why user testing is…
There are things I don't miss about the era -- good riddance to floppy disks and hard drive crashes -- but the PC world pre-IBM had so much latent possibility and sometimes just plain weird stuff:
http://www.landley.net/history/mirror/cpm/IMSAI_pre-history....
> A major advance was announced in a press release dated October 25, 1975 in which the young specialty firm offered a relatively new and promising concept. The Hypercube was advertised as a four dimensional arrangement of dual 8080 processor "nodes" configured in 2x2x2x2, 3x3x3x3, and 4x4x4x4 arrays, with each node capable of communicating, via shared memory, with 8 adjacent nodes. This arrangement provided for the first processor in each node to handle system overhead and communications tasks while the second was left free to execute user code. The operating code was to be stored in ROM, and the total system promised unparalleled processing power at a fraction of the cost and overhead of mainframe machines from IBM, Honeywell, Boroughs, and other giants of the period. The advertised price of these three offerings was $80,000 for the Hypercube II, $400,000 for the Hypercube III (about 1/10th the cost of an IBM 370-168), and $1,280,000 for the Hypercube IV which was to be released in the second quarter of 1976. The concept was legitimized by publication in the December 11, 1975 issue of ELECTRONICS magazine. Ultimately, the U.S. Navy ordered a Hypercube II for installation in Huntsville, Alabama.