#lispyGopherClimate okay Mastofriends, it is Sunday-Morning-In-Europe ( #peertube #live ) time.

https://toobnix.org/w/ueLph1idzd7HDrz9cPDSu2

My ansi #commonLisp #conditionSystem | #ontology blog here : https://lispy-gopher-show.itch.io/leonardo-calculus/devlog/1451887/my-ansi-common-lisp-condition-ontology-eg or what it is like on my machine, anyway!

Failed attempt to verbally skim Kent Pitman, Ramin, Crew, DM, et al.'s thread: https://climatejustice.social/@kentpitman/116167936268937333 though it is weeks of reading material.

Edit: Ignore every mention of the very nice mastodon, founder of Java please. My brain was not working.

Public Lispy Gopher Climate sunday Morning in Europe stream - 3/8/2026, 7:48:59 AM

PeerTube
@screwlisp
I'm sure we were all vastly surprised by the sudden brief appearance of *Gosling* in this thread.
@dougmerritt Hang on, what terrible mistake did I make
@dougmerritt oh yes I attributed an emacs to gosling when he was of course the founder of *java*. I had my brain wires completely crossed last night.
@screwlisp @dougmerritt He is both of those things. He's a floor wax (wrote Gosling Emacs) and dessert topping (wrote Java).

@mdhughes @screwlisp
Gosling Emacs had an intelligent display update algorithm, which optimized screen rendering, which was important back in the day of slow connections (e.g. 9600 baud serial multiplexers for timeshare or individual modems).

My introduction to NP Completeness was Bill Joy explaining to me (late 1970s) why it arose in the context of optimized display update algorithms, with the conclusion that one shouldn't hope to have a provably optimal algorithm for such.

Eventually of course it became feasible to just rewrite the whole damn screen every time, and not worry about differential updates.

And sometimes similarly with small instances of data structures.

@dougmerritt @screwlisp If I remember right (dubious), my own editor (MRED, Mark's Reliable) on the Atari 8-bit kept a dirty line list, and just redrew those lines; scanning a line out of store and poking into display memory was not quite fast enough to do full-screen redraw.

Scrolling up or down had 2-3 lines of buffer before I had to redraw.

@mdhughes @screwlisp
Reasonable, especially in that precise circumstance.
@mdhughes @dougmerritt @screwlisp On my TRS-80 I converted a cursor gap editor from Byte magazine or such to the Trash-80 and had it redraw the whole screen. It worked just fine. But THAT screen was memory-mapped SRAM and the code was handwritten 8080/Z80.
@mdhughes @dougmerritt @screwlisp My TRS-80 had a piggybacked SRAM chip for lowercase. That was a cool hack.

@chemoelectric
I have nightmares about past uppercase-only devices, and when I wake up, I say to myself "we have lower case too, it's all ok, we have lower case, don't fret, there there."

@mdhughes @screwlisp

@dougmerritt @mdhughes @screwlisp the "a" and the "p" were not on the baseline, but I lived with that.

@mdhughes @dougmerritt @screwlisp

I am sure trying to prove optimum algorithms for screen motion is a waste of time. But poor screen performance is still a big problem in virtual machines.

@chemoelectric
Which virtual machines are we talking about here?

@mdhughes @screwlisp

@dougmerritt @mdhughes @screwlisp Oh virtualbox. I mean, I know it sucks, but I am used to it.