modrobert @modrobert@qoto.org

After the most recent shenanigans by Github, I really want to move my projects off it. At the same time I'm lazy, and it takes some effort figuring out how to do it this.

Ideally I'd like to self-host, that means I have yet another system to keep up to date. Then there is Gitlab, but they are also a US corporation, and while they have done less stupid things in the past, the drama around the telemetry is still remembered.

Then there is Sourcehut which is good but has some issues. On the technical side it's that it has sometimes been quite slow when accessed from Asia.

If you're not following @RoofSlapping why are you even here?

Been working some more on my programming language. Performance is not great, but acceptable in most cases (but since people seem to be happy with Python, perhaps I shouldn't worry about making it performant).

The code base is only 17k lines of Kotlin. That's much less than expected.

It not builds for the JVM, Natively as well as to Javascript, so it's even possible to run it in the browser.

@modrobert @loke The GEOS KERNAL defines the macro (and also uses it). You can change it in my reverse-engineered source: https://github.com/mist64/geos/blob/master/inc/geosmac.inc :)

A photo of Murray the Moray Eel from one of my dives in the caribbean.

#scuba #scubadiving #MorayEel #diving #USVI #Science

1969 Tektronix white paper explains how a mouse works. #retrocomputing https://archive.org/details/bitsavers_tektronixcormationDisplayConceptsJan69_5521470/page/n69/mode/2up

Oh I so love this comment from Reddit:

Computer Science had fantastically approachable textbooks with clear, simple diagrams, as well as practical examples that you could really implement yourself to solve real problems. I noticed however that while the research papers written in the 60s and 70s had a clear, easily readable style, around the late 90s they become riddled with made-up pseudocode unique to each paper and at least 50% Greek letters, because deep in ours hearts we all recognise that we're not proper computer scientists until we learn to program in ancient Greek like the great philosophers did thousands of years ago.

Found in a discussion here: https://www.reddit.com/r/AskPhysics/comments/fek5dd/why_are_textbooks_written_so_badly/fjpyxwu/

I don't think he deserved to be downvoted.

Latest progress on my programming language: Some basic APL functionality has been implemented (such as ⍴, ⍳, ¨).

The main thing that sets my language apart from APL is that it's lazy evaluated. That is essential when working with large arrays. An operation such as ⍉ or two-argument ⍴ are very cheap in terms of CPU usage.

I've also started building a simple user interface for it.

So #Trump just posted his new logo for the "United States Space Force"... apparently we are now officially in the United Federation of Planets, just without all that non-intervention equality nonsense.

#uspol #SpaceForce #StarTrek #Trek #SciFi