Lenguador

Lenguador, 10 months ago

Worth a watch, there were some good jokes in there. This channel does videos on other programming languages and related things (I quite enjoyed this one on emacs, though maybe that's because I know nothing about emacs).

Lenguador, 10 months ago

The article does address this. 20% of unmarried 40-year-olds are cohabitating. So that still leaves 20% of all 40-year-olds as never-married and not currently cohabitating.

The stats on education level were surprising to me too, though it does make sense as education is correlated with wealth, health, and long-term planning.

Lenguador, 10 months ago (edited 10 months ago)

I asked the same question of GPT3.5 and got the response "The former chancellor of Germany has the book." And also: "The nurse has the book. In the scenario you described, the nurse is the one who grabs the book and gives it to the former chancellor of Germany." and a bunch of other variations.

Anyone doing these experiments who does not understand the concept of a "temperature" parameter for the model, and who is not controlling for that, is giving bad information.

Either you can say: At 0 temperature, the model outputs XYZ. Or, you can say that at a certain temperature value, the model's outputs follow some distribution (much harder to do).

Yes, there's a statistical bias in the training data that "nurses" are female. And at high temperatures, this prior is over-represented. I guess that's useful to know for people just blindly using the free chat tool from openAI. But it doesn't necessarily represent a problem with the model itself. And to say it "fails entirely" is just completely wrong.

Lenguador, 10 months ago

The greatest fix to all your pointer issues are to use references.

Lenguador, 10 months ago

If this is a real question, this talk explains the fundamental concepts of atomic operations in a couple of minutes.

The talk itself is over an hour long, as the use of atomic operations has a large number of pitfalls. The joke in the meme leans on a specific type of memory ordering guarantee, known as "relaxed" in C++ parlance, which can be a lot faster, but which is much more likely to violate the default assumptions a programmer may make about order of operations and visibility across threads.

Lenguador, 11 months ago

None of the related magazines in this community are listing this community in their related magazine section.

I'd advise expanding (and trimming) the related magazine section, plus messaging mods to ask if they'd like to add your magazine to their related section.

Lenguador, 11 months ago

Thanks for the advice, but I actually love code reviews.

Lenguador, 11 months ago

Looks like the same guys were doing publicity around 2019 https://www.abc.net.au/news/rural/2019-07-30/australia-joins-lab-grown-meat-industry/11360506
At the time, they claimed the cost to make a single hamburger was $30-$40, and now 4 years later, they claim to have gotten it down to $5-$6 per patty.
The article claims the first demonstration of a lab-grown hamburger was in 2013.

So 6 years from proof of concept to (probably) first capital raise, then 4 years to start regulatory approval, 1 year for approval to take place (target is March next year).

Lenguador, 11 months ago

I wonder what specifically they're interested in vs long deployments in Antarctica (people do 12 months rotations in some stations there).

I found this article discussing the psychology of placements in Australian antarctic stations: https://psychology.org.au/for-members/publications/inpsych/2021/february-march-issue-1/life-in-the-australian-antarctic-program.

The differences as I see them are:

1. Smaller crew
2. No unsuited outdoor time
3. Smaller space
4. Communication latency / outages
5. Personal belongings weight/volume limits
6. Dietary restrictions

Lenguador, 11 months ago

I am reminded of this comic by Randall Munroe https://xkcd.com/505

To make it easier to wrap my head around, I summarize each dot point:

o World can be simulated
o Substrate independent
o Simulate with pebbles
o Feels real to simulated beings
o Simulator can be turing machine
o Deterministic
o Independent of obsever
o Aside - different interpretations give different worlds?
o Can symbolically represent patterns
o pi is an example. Digits of pi could represent a world
o Sequences are independent of symbols
o Infinite sequences with infinite interpretations gives infinite possible worlds - thus, we're infinitely likely to live in a simulated world, even if a physical world is required to give rise to all those permutations

Then summarize the argument:

1. All worlds can be simulated
2. All simulations can be formulated as a sequence
3. All sequences exist
4. Therefore, all worlds exist

I think plenty of people would argue about (1), though I do not. (2) follows from (1).

I think (3) is not necessarily a given. We can conceptualize of a sequence, we can even write down a representation of a sequence, but in what manner does a sequence actually exist? Similar to the concept of a triangle. No such thing exists. We can only create representations of a triangle, or posit statements based on the supposed properties of a triangle, were it to exist.

Finally:

First you say that a sequence can represent a world, given an interpreter.
Then you say that such a sequence can represent a world, independent of an interpreter.
But you give no justification for that.

Consider an alternate formulation:

1. All worlds can be simulated
2. For the set of all interpreters, there exists at least 1 sequence which would be interpreted as the simulated world
3. All interpreters exist <- Doubtful
4. Therefore all worlds exist

You say that a sequence can be interpreted as a world, but is independent of that interpretation. In that case, sequences are unnecessary, because I could conceive of an interpreter which perceives our universe from any symbol or sequence. So it is not the existence of the sequence which is what causes the world to exist, but the interpreter.

Lenguador, 11 months ago

!literature should go in your list, it has more of a poetry slant.
!books has almost 3000 members

I'm sure you can find plenty more on kbin.social as well.