Notices by Tom Ritchford (tomswirly@toot.community)

@toni @futurebird Sadly, AWS counts: it's about 60% of Amazon's profits, it seems: https://www.cnbc.com/2024/10/31/amazons-cloud-unit-records-highest-profit-margin-in-at-least-a-decade.html

I personally use Linode precisely because I detest Amazon so much.

@Sarahw @tompearce49

I too am 62. I too got very sick from measles as a small child. 35 years later my dentist could detect this in my teeth!

My parents were strong on vaccinations, but the measles vaccine just didn't exist in the UK yet.

@cstross @futurebird @jannem @dx Wow, you're even more conservative on this than I am, and my theory is you've done more of the work than I have too, so your number is more accurate.

My go-to is the computer chip. Building a single chip requires an entire chemical industry able to produce thousands of specific molecules, many of which have to be at insanely high purities.

@tchambers @kamalaharrisforpresidentnews

I have very high hopes for the Harris/Wals administration.

But it would be better not to remind us of the failed promises of the Obama administration by repurposing old design styles. Obama's love of fracking and fossil fuels, of bankers, and of war, his fumbling of the Supreme Court, these are still affecting us negatively do this day.

We need something new, real and true, not empty slogans like "Hope", "Change" and now "Forward".

@peterbrown @gerrymcgovern In fact, it's a huge problem.

We needed to decarbonize our society by 90% in less than 10 years to avoid climate disaster. We won't hit that target, not even close.

Every watt of additional power that we pile on makes that target even further away.

Imagine CO2 was a monetary debt. We are in a huge amount of debt; every year we go further into debt.

Now people are proposing spending even more CO2/debt on data centers.

1/

@breadandcircuses Skeptical about the number of climate offsets that are are worthless.

To be honest, I think they're all worthless.

We need to decarbonize by 90% in a decade. Some plan that you can pay to emit CO2 as long as you have some other activity that supposedly consumes it, seems like an obvious loser to me.

@dalias @leighms @malcircuit

> Because these are the meaningful working definitions for intelligence.

I'm sorry, but I disagree.

Oxford defines intelligence as "the ability to acquire and apply knowledge and skills". MW: "the ability to learn or understand or to deal with new or trying situations : reason". I went through a half-a-dozen other definitions, all similar.

Your definition conflates three different ideas: "intelligence", "living", and "agency".

@malcircuit @dalias @leighms

That isn't very polite. 😕

You use "intelligence" in a way quite different from how regular people use it, how the dictionary uses it, and how scientists use it.

Even without the mockery, no useful conversation could happen. By your definition, intelligence implies being alive, so machines aren't intelligent, and what is there to discuss?

But this sheds no light on the actual question at all.

1/

@malcircuit @dalias @leighms As someone who loves words and learning, it makes me sad to see people mocking dictionaries. One of their key purposes is to make sure that communication is possible between people because they are using the same terms.

Many interesting debates about whether computer programs are or could become intelligent are possible, but not if one side redefines the word intelligent in a non-standard way to specifically exclude computer programs.

2/

@malcircuit @dalias @leighms And a note to two of you in this thread: using mockery and profanity against someone who has been scrupulously polite, if perhaps a bit ponderous and academic, is not just disrespectful, it shows a lack of confidence in your argument.

If I wanted to be mocked and insulted, I could go to Twitter.

Let's hope you cheer up and have a much better day!

@leighms @malcircuit > I think the test suggested by @malcircuit is a good one.

Because why? Here are some highlights

> True intelligence can only exist in living creatures.

Why?

> Intelligence is the thing living creatures use to improve their chances of survival.

Why?

> Intelligence requires agency in the physical world.

Why?

> An algorithm alone cannot be intelligent.

Why?

These aren't arguments: these are unsupported claims.

@malcircuit @dalias @leighms If you can redefine any word to have a different meaning from both how regular people use it, and how dictionaries define it, you can win any argument, but it makes everything rather meaningless.

Your argument is now this: "computers cannot be intelligent, because my definition of the word intelligence excludes computers."

It certainly ends the argument, but not in an interesting or useful way.

@SpaceJellyfish @simon_brooke You are proposing exactly the same strategy that environmentalists have been doing for the last 60 years, a strategy that has utterly and completely failed.

Civil disobedience has generations of success - it got women the vote, and got civil rights for people of color.

I don't think it's too much, I think it's not enough.

1/

@SpaceJellyfish @simon_brooke

Imagine this was an action movie, where a tiny number of psychopaths were systematically destroying the Earth, not even for goods and services, but simply to win a meaningless numerical game called The Money. These psychopaths control pretty well everything, so anyone opposing them gets crushed in horrible ways.

Is "starting an information portal on the web" really going to be what defeats these incredibly powerful villains?

@sidereal @sandlapper37 "The United States is also a one-party state but, with typical American extravagance, they have two of them."

@sandlapper37 Gosh, I heard things like that for thirty years. Thing is, there are only two bus lines in the US of A, and both of them go in the opposite direction from where I want to go. Sure, one of them goes straight to hell, and the other one goes there much slowly, but either way I get further from my goal.

@foone Come now.

Programming is an engineering discipline! There are very few happy little accidents. You want code that does exactly what you intended it to.

Everyone understandably loves Bob Ross, I do too, but he isn't a good example for programmers, engineers, surgeons, or airplane pilots to follow.

@pippin @foone Writing a program that doesn't actually work doesn't seem more frustrational than recreational to me! 😀

@pippin @foone Even in the demo scene, 98% of the accidents are errors that need to be corrected or else nothing will happen (I did write "very few happy accidents", not "no".)

And hacking on old machines is even more exacting. When people get unexpected results out of old gear, it's the result of a huge amount of systematic, exploratory trial and error, much closer to science than art.

1/

@pippin @foone

If a typical drawing is made up of 200 strokes, and each stroke has a 98% chance of being "wrong" and spoiling the drawing, then that would mean that 99.99999...% of drawings would be wrong.

(Nearly all drawings or paintings are built up of a large number of small gestures, and an error in any one of them isn't usually significant.)

I live with someone who's a trained draftsperson for art: this is a skill only peripherally related to what Bob Ross does.