Andrew Miloradovsky is a user on You can follow them or interact with them if you have an account anywhere in the fediverse. If you don't, you can sign up here.

Andrew Miloradovsky

Lua ought to have a symbol type (in the Lisp sense) imo; I really shouldn't be able to call 'string.reverse(coroutine.status(cr))'

Have any culture ever devised a proper #hexadecimal notation? I'm not speaking about utilizing the Latin #letters (a-f/A-F) as the #digits.
For instance, is there ever a #glyph which would mean "ten" in any context, just like 2 means "two", even if standing alone.

Social media — a platform for thinking aloud.

What I have on my mind here is something along the lines of a #p-adic #analysis of #algorithms.
#2-adic (#dyadic), more specifically, since base 2 is easier to interpret in logic.

The point is to (extensionally) identify programs with some numbers, and apply all sorts of "continuous" math to them.
And the p-adic expansion seem to be more appropriate for this approach than the usual "decimal" expansion.

OTOH, the #Godwin's law as whole seems to be meaningless — any stream of text, if it isn't algorithmically generated, has exactly #probability 1 of finding any given word or phrase or generally a pattern "eventually".

If the law were quantitative, p(n) = …; n is natural, p is monotonic, p(0) = 0, and limit p(n) = 1 for n → ∞.

If we have a #program, generating a #stream of symbols from an alphabet (say, 0's and 1's), then there is a (finite) natural #number of steps, after which the output will repeat?
And thus there is a #map from such programs onto the real numbers in [0, 1]? Not invertible, because e.g. 1.0… and 0.9… are mapped to the same real number.
May we introduce e.g. a meaningful #topology or #measure on the programs this way?

Any references, besides the whole theory of #computable #functions or #automata?

when they told you that computers are really dumb, because they only do exactly what you tell them to do, that was a lie

a lie of omission.

they do what you tell them to do, while also doing what a million other people over the past 40 years told them to do to

sometimes, those commands interfere with, or contradict each other

Regular reminder: "legal name" is a very boring sounding phrase.

It sounds WAY cooler if you call it your "government alias"

Ideals may be prime, primary, primal, primitive, and principal, to name a few. All of those are similar but different. — Good luck figuring out how exactly.

In case you've been wondering why relying on text messages to your phone for authentication sucks. I've just had my phone number stolen.

I'm thinking about giving a talk about how we went balls deep with extensible effects in our production code.

My #job is #hiring again in both #Montreal and #Edinburgh! The Canada position is for a Python and Django developer who will work on our #ecological #mapping platform, and the Scotland position is for a project manager for the same #environmental mapping platform.

#Graph #theory #terminology question: Let

• \( A \) be a type/set (of vertices)
• \( d: A^2 \rightarrow \mathbb {R_+} \) — a distance(s)
• \( f: A^2 \rightarrow \mathbb {R_+} \), defined as \( f (x, y) := \sum_{a \in A} d (a, x) d (a, y) \)
• and \( g: A \rightarrow \mathbb {R_+} \) — as \( g (x) := f (x, x) = \sum_{a \in A} d (a, x)^2 \)

What's the proper name for \( f \) and \( g \)?

Another great quote from the same man:

" is the [possible] between persons."

"They won't give you what you don't have, they don't have it themselves."

— Mikhail Zhvanetsky

To get bidirectional rendering to work correctly, I have to learn and implement UAX#9, Bidirectional Algorithm.

There is a lot of yak that remains to be shaved.

Latest progress on the Freetype renderer for McCLIM.