Thread

Web-based IBM 1620 emulator and IPL-V from 1963

48 points · 14 comments · abrax3141

Loading the complete thread in the background. This saved snapshot is available now. Refresh

abrax3141 · 1790121985 · focus · HN ↗
Paul Kimpel has built a lovely web-based IBM 1620 emulator, and was able to compile and run a 1963 IPL-V compiler and runtime from U Oregon. Run it all from your browser! Execution details and project history linked from Paul’s repo: <a href="https://github.com/pkimpel/retro-1620" rel="nofollow">https://github.com/pkimpel/retro-1620
rzzzwilson · 1790128710 · focus · HN ↗
A blast from the past! An IBM 1620 was the very first computer I programmed, way back in 1969, likely a model 2. As part of a second year Applied Mathematics semester we were given a block of coding forms, a McCracken book on FORTRAN II and a list of problems to solve. There was no instruction on how to do this beyond reading the book, as far as I can remember. Someone punched cards from our coding forms and put them into the machine. We never actually got to touch the hardware, but we could look at it through glass windows.
That was a life-changing experience as it led to me dropping out of my Physics degree and eventually going into a computing career.
kragen · 1790177153 · focus · HN ↗
Currently the title says "Web-based IBM 1620 emulator and IPL-V from 1963" until HN's title policy inevitably vandalizes it to remove the informative part. But, to give further context on IPL-V, IPL (I think actually IPL-IV) was key to some early AI research, but the main reason it's interesting in 02026 is that it's what inspired Lisp. It's a terrible programming language by today's standards, and the manual is a horrible slog, but you can see the seed of almost every programming language popular today in it.
1. nxobject · 1790202824 · focus · HN ↗
  Reading through a CACM publication, I'm struck by how they call the whole system the "IPL Computer" -- they were specifying a bytecode VM. It looks like early IPLs dated to the 1950s, so they'll get a pass for the actual language...
2. agumonkey · 1790285023 · focus · HN ↗
  I didn't realize IPL lasted that long since indeed it's mentioned by mccarthy as a strong influence.. somehow I always assumed it meant pre-1959.
  btw, for people who were there at the time, even considering the bare metal assembly nature of IPL, were people already approaching solutions similarly to lispers in the 60s ? aka dynamic generic lists, tree traversals etc
  1. abrax3141 · 1790313541 · focus · HN ↗
    Yes. All of those were written in IPL well before Lisp. There were a dozen IPL-V implementations across almost every common computer of tat time. (Of course, there weren’t that many common computers! :-)
    agumonkey · 1790319665 · focus · HN ↗
    I need to find readings about the various versions and projects made on them.
    abrax3141 · 1790346878 · focus · HN ↗
    The 1964 manual has a list of implementations on pdf pg 3: <a href="https://archive.org/details/bitsavers_randiplInfanguageVSecondEdition1964_15001417" rel="nofollow">https://archive.org/details/bitsavers_randiplInfanguageVSeco...
    agumonkey · 1790361704 · focus · HN ↗
    Oh thanks. The first pages are already very interesting.
3. abrax3141 · 1790313325 · focus · HN ↗
  True enough that IPL inspired Lisp (along with Church’s Lambda Calculus), but, to my mind, what is interesting about IPL-* is that the first true AIs and Cognitive Models were written in it/them, esp The Logic Theorist, which was the first heuristic search program, and which actually improved upon proofs by some of the most famous logicians of the time. David Moews has made several of the original IPL Logic Theorist programs run again (although not on this 1620 platfor’. . . Yet!) :: <a href="https://news.ycombinator.com/item?id=48116935">https://news.ycombinator.com/item?id=48116935
shevy-java · 1790184147 · focus · HN ↗
Has it been ported to TempleOS yet?
fernly · 1790185726 · focus · HN ↗
I have heard that among older IBMers the 1620 (model 1) was known as CADET, for "Can't Add Doesn't Even Try".
1. OhMeadhbh · 1790189213 · focus · HN ↗
  LOL. I came here to make the same comment. I think the deal was they used a lookup table for addition instead of a modern ALU.
  1. TedDoesntTalk · 1790199521 · focus · HN ↗
    That’s what the GitHub page says, too. Lookup tables for multiplication and division, too.
2. vincent-manis · 1790208122 · focus · HN ↗
  The reason for that is that it did arithmetic via table lookup. As I heard from someone who was there, the University of British Columbia wanted to show off its brand new 1620, so they wrote a demo program, and with a host of onlookers, they pressed Memory Clear, then loaded the program, and it failed. Memory clear had erased the addition and multiplication tables.
rbanffy · 1790195652 · focus · HN ↗
Shameless plug (and hopeful suggestion): every esoteric/ancient computer system deserves a page like this one (that I’ll eventually finish).
<a href="https://rbanffy.github.io/fun-with-big-computers/fun-with-vm370/" rel="nofollow">https://rbanffy.github.io/fun-with-big-computers/fun-with-vm...
BTW, I have updated the container images for VM/370 Sixpack and VM/370 CE today.
jonjacky · 1790221203 · focus · HN ↗
Most people programmed the IBM 1620 like in the rzzzwilson comment here: FORTRAN on punch cards, running out of reach in a closed machine room. But the 1620 did have a console terminal so it could be used interactively.
Somebody at IBM built an interactive APL interpreter that ran on the 1620, but it seems this was never released:
Experimental personalized array translator system <a href="https://dl.acm.org/doi/10.1145/364520.364573" rel="nofollow">https://dl.acm.org/doi/10.1145/364520.364573

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.