Notices by Weekend Editor (weekend_editor@mathstodon.xyz), page 2

@amszmidt

Years ago, I had a colleague from Estonia (Talinn). He kept telling me that your native language was probably the language in which you count money.

Couple years later, I was in the Netherlands. Guy in a shop spoke absolutely perfect English (probably RP accent from the BBC). But when I paid for something, he made change whispering the count to himself in Dutch.

@amszmidt

Got any basis to compare with Finnish?

My friends tell me it's fiendishly hard. But some of them are Finns, so they might be pulling my leg.

@symbolics @screwtape @remilia @notptr

RIght, there's no kernel. Everything runs in the same address space. It relies on hardware type-checking to make this *somewhat* safe; mostly you'd enter the debugger instead of crashing. (Or entering the cold load stream, which is more serious but still recoverable.)

Any function could be altered and recompiled directly to memory, not touching the file system. The next call to that function uses the new version. (This applies to recursion as well: if you're many stack frames deep into recursive calls, the active call will "return" into the new function!)

This is *very* useful for debugging user programs.

On the other hand, altering running system code is sometimes... highly inadvisable.

There were no guard rails telling you not to do that. There was no security model worth the mention, by today's standards (just the barest of user id and password at login).

Switching to another world load ("band" in CADR terminology because there used to be some effort to lay them out on a disk in a contiguous band) is rebooting.

@sigue

I know, right?

Like, writing a Unix shell script and people expect me to use sed? That looks like an editor I was using in 1974!

@ksaj @lispwizard @amszmidt @screwtape

There should be Special Compiler Warnings from Hell for people who do this.

I mean, sure, compile their program. But not without some verbal abuse.

@ksaj @lispwizard @amszmidt @screwtape

On the Symbolics system, the type tags were all encoded in the microcode source and recognized in the hardware (ok, really the microcode). The type tag names were constants whose names all began with "DTP-" (for "data type", probably).

There was a type calld DTP-NULL (see below, page 7, though I think that's for the Ivory chip).

It didn't *quite* mean WTF, but almost. :-)

(No, it was not used this way.)

https://bitsavers.org/pdf/symbolics/I_Machine/Lisp-Machine_Data_Types.pdf

@larsbrinkhoff @kentpitman @ldbeth @kim

I remember as an early Emacs user at MIT -- ca 1977 -- trying to figure out what TECO was all about.

It was utterly impenetrable to an outsider!

An ironic saying then was, "TECO: a moment of convenience, a lifetime of regret."

@amszmidt @kentpitman @runes @lispm @masinter @amoroso

To use that code directly, you'd need some 3600 microcode.

I think there's a CL version of it knocking around somewhere, though, if you're adventurous.

@kentpitman @runes @lispm @masinter @amoroso

It didn't *have* to be done that way, since there were other ways to do it. But we already had the microcode support for Prolog logic variables in place.

So that was the only time in my career when "microcode" and "straightforward path" were the same. :-)

@kentpitman @runes @lispm @masinter @amoroso

I keep thinking it would be a great retirement hobby project to make a successor to Joshua.

I'm thinking of embedding it as a domain-specific langauge extension of Haskell.

Partly that's just to get the project motto to be:

"Old-fashioned AI you don't need, in a language you don't know."

(Slightly seriously: every time in my adult life when I decided to learn a "radical" language, it led to major new opportunities opening up for me. Lisp and R being the 2 main examples.)

@larsbrinkhoff

But does it do CHAOS net? :-)

@amszmidt

Some of us have been using various Emacsen for a long time. (Since about 1978, for me.)

We're likely preserving old defaults that are too far down in finger memory to be worth changing.

@amszmidt

About half of us learned 40+ years ago, with Reagan.

The other half of us are incapable of learning, yet we must live with them.

@amszmidt

In the US right now, being happy is the incorrect response.

As we are all about to learn.

@amszmidt

Coulda been worse.

Frighteningly enough, we have other billionaires who are far worse.

@amszmidt

Or... perhaps we learned that Occupy Wall Street accomplished very little, and thus is not worth repeating?

@amszmidt

I think we have to invade ourselves?

I mean, if the Canadians want to take over the US government, I'm sure they'll be so polite about it we might not notice until we discover we all have health insurance.

@amszmidt

But if "nobody" agrees about software, then we all agree on *that* fact about software. And that's something about which we all agree, right?

Think ya got a Russell Paradox infestation, there.

@nuxi @amszmidt

If it's in Common Lisp, it can return a value, or return multiple values all at the same time, or even return no values.

@cutting @larsbrinkhoff @amszmidt @shapr

I had used a Dolphin in the early 80s, but never a Dorado.

My impression was they went crazy with ECL technology chips to implement it. But it still wasn't completely reliable because it was full of race conditions?

(Memory from the early 1980s, so not the most reliable memory. Also, I never liked Interlisp, so I may have some remaining prejudice in the matter.)