Hacker News Reader

terminal_don PFAS can suppress white blood cells’ ability to destroy invaders

"The best nootropic is getting up in the morning and getting sunlight. Then you sleep at a reasonable hour and wake up fully rested."

terminal_don Seattle becomes first U.S. city to ban caste discrimination

North and South both screw each other over. And people of the same race / region / caste / creed screw over others belonging to the same group. This is universal and will never change. What this legislation does is line the pockets of the people putting their pens to paper for it while piggybacking on a divisive issue.

And yeah, I've had good friends from almost every state in India. Some of them were weird about it, most don't really care. So-called "nationalists" will screw you over in any region if you're a migrant.

Anyway, it's not a binary issue. It's just something that happens, and it's not possible to legislate.

terminal_don Seattle becomes first U.S. city to ban caste discrimination

Yup, because that works out really well in India...

terminal_don Seattle becomes first U.S. city to ban caste discrimination

Funny how the United States is speedrunning ""sensible"" legislation without considering what it entails in the country the practice originates from. And yep, once enforced it's never going away. I guess people can now look forward to ever-convoluted steps for mitigating issues caused by this, and it'll fill the coffers of anyone that puts his stake into it. Better for the lawmakers, crap for literally everyone else.

BTW, to anyone that hasn't lived in India: you don't understand how extensive """caste discrimination""" is or how it impacts society. You just don't.

terminal_don Software 2.0 (2017)

>Have you _used_ ChatGPT?

I have, and I didn't find it to be useful for anything I did. I can do what it does with a search engine and trusty C-f. Also, TTS exists.

It's 50% tech and 50% marketing (and I doubt it's 50% tech at that), it's not gonna upend anything. Except maybe increase the authenticity of online scams and make people get more degrees in machine learning. And yeah, make the people that rely on it bound as it degrades their skills.

It's basically the "internet is educationally useful" argument. At some point everyone's gotta use it but you can live without it just fine. And even though people tout its usefulness for everything good, the majority of data transferred is porno.

terminal_don Software 2.0 (2017)

The adoption of the so-called "Software 2.0" depends on how influential OpenAI is as a company. The majority of the use-cases of GPT-3 are only about "remixing" information, and the "notoriety" of the AI is (IMO) mostly a publicity stunt; I'm assuming that there are prompts for personalities / jokes / poems / etc that are never shown to the end-user and are hailed as an "organic outcome" of the model. Obviously, no one can check for these right now. Seems to me that microsoft has taken a few leaves out of the OpenAI playbook and applied them (rather chaotically).

So when you're looking at actually writing software that needs to be dependable / modifiable / bug free, you'd need a massive overhaul of whatever software stack is being used, so there's very little human-assisting "cruft", and instead you'd want a lot of supporting material for a model, which might look like something written in languages used for formal verification of programs.

The promise of GOFAI was about having a human-understandable bottom-to-top framework, and the current "AI" paradigm is at odds with it. The "formal verification" assumption, then, skews towards GOFAI. But since there has to be some human support for the current not-there-yet AI to write software, we might see yet another abstraction layer based on NN / something newer in the years to come.

terminal_don Ask HN: How do AI chatbots develop personalities?

They don't. It's you who gives it meaning. Also, it's trained on trillions (gazillions?) of man-hours of data, so even if it's an antiquated "algorithm" it manages to be authentic. Also, no one factors in the small adjustments (and likely deliberate prompt / text insertions) that the parent company would undertake to make the ""AI"" seem more real. It's 50% tech, 50% marketing.

terminal_don People can read their manager's mind (2015)

Companies either want you to be technically proficient or to support the hierarchy. You need to suss out what kind of company you work for and act accordingly.

terminal_don Consultancies know less than they claim and cost more than they seem to

They exist to propagate good feelings amongst management and to ensure a fat paycheck keeps coming though, even if it means sabotaging the company.

terminal_don Consultancies know less than they claim and cost more than they seem to

IT Consultants =/= Standard Consultants from Big4 and the like

terminal_don Consultancies know less than they claim and cost more than they seem to

No, that's just the business tagline. Consultancies purity spiral on Goodhart's law. They're worse than useless, and it's their "cult"-like behavior that is the basis of their entire business.

You HAVE to believe they bring in value because... well, you gotta believe.

terminal_don Ask HN: Things that have made you a better programmer

Thinking hard and pausing before writing: If you can suss out the pitfalls for some program before you write a significant amount of it, it'd save you a lot of time. The way you've learnt something can lead you to be stubborn about a certain process, which means you might have to "hack" your way through it -- I ""prototype"" code in an easier language before I start writing it. It doesn't have to run in that language. I use lua for this purpose.

Take diligent notes: My system (in emacs) helps me out here. Taking a few seconds to look for a solution in your notes trumps going through a wild goose chase on the internet only to encounter links that you immediately half-recognize as something you tried six months ago when you had $issue. "Quantified life" stuff helps out here. Here's a big protip: never delete your bash history.

terminal_don Ask HN: Are you tired of reading ChatGPT headlines?

> The newer research papers coming out are where I would focus if I'm really worried. Great reply, could you expand on this?

terminal_don Ask HN: Are you tired of reading ChatGPT headlines?

I think that's more representative of tabloid writers than anything, haha. Understanding text is difficult, and scales with g. GPT-3 can make us believe that it can comprehend text that falls in the median of internet content, and I guess there would have to be some edge cases addressed by the devs, but it can't convince humans that is understands more difficult content, or even content that isn't in its db.

terminal_don The Business of Book Styling

Arranging books in shelves by color is the biggest pseud tell out there.

terminal_don Olympic medalists in art competitions

You know what would REALLY be a cool competition?

Who can build [X] the fastest, from scratch. Could be a house, clothes, food, etc. Known objective, and meritocratic.

terminal_don Microsoft to lay off 11k employees

Hey, you're right -- this is a bargaining chip for BigCo, but the "invisible" factor here is that the economy is clearly in freefall and no one is ready to call the emperor naked. Everyone's cutting their losses, and in sectors as important as tech and with companies as big as MSFT, there's always some government support and "insider insight" into these events. If BigTech comes out of this on the other side, I fully expect one big company to be absorbed / dependent upon another. If not, this is prime time for underdogs to really show what they are capable of.

terminal_don Ask HN: How to get better at spoken English?

Starting early trumps everything else.

You can still try and be competent in a year or two, but you'll sound weird saying certain words.

Read more, speak more, and try to not translate from your native language to english.

terminal_don Nikita Bier – Gas has been acquired by Discord

"A boy gassed you up"

I've never heard about this before. Seems like gen Z (gen alpha?) is increasingly tech illiterate and tech-dependent. Not ten years ago, most kids weren't given phones by their parents. I guess it would be impossible to graduate high school these days without a phone.

terminal_don Ask HN: What are some apps that feel like magic to you?

SQLite. Were it a corporation, it'd be a FAANG.

terminal_don NeevaAI

No, it's "PEBKAC", so we need multiple AI personalities even if they're connected to the same backend.

terminal_don Which CPU is best for long time use Intel or AMD?

What ice-cream flavor is best? Chocolate Chip or Strawberry?

terminal_don Fields where it matters, fields where you can thrive on BS alone, and in between

It's the managerial pipeline! Handle it with care because it will never be abolished.

terminal_don NeevaAI

Because google searches show results from many people, and so disregarding results (and people) is easy. This is ONE machine, in a take-it-or-leave-it scenario. You'll either like most of the results or leave because some didn't work for you.

terminal_don Things they didn’t teach you about software engineering

I love this comment. It dodges the point so well.

terminal_don Human gene linked to bigger brains was born from seemingly useless DNA

https://pumpkinperson.com/2019/01/04/what-is-the-correlation...

https://pumpkinperson.com/2019/08/15/increasing-u-s-head-siz...

Last link has a formula to calculate cranial volume.

Animal intelligence is vastly underplayed. Ants seem to be a very intelligent creature but they're little more than part of nature to us. We're similarly adapted to our environment, and those animals are adapted to our patterns and mentation by virtue of millennia of close contact / purposeful genetic pushes (eg, dogs)

terminal_don Google researcher, long out of math, cracks devilish problem about sets

Isn't that overcomplicating the solution?

The only requirement is to have a unique number of coins from the fake coin pouch. Since we don't know which one that is, every pouch gets a different number of coins, and not necessarily a sequence starting from 1. Then, you weigh them all assuming they're all real, but wait, they're not! So you take the excess weight, divide it by the excess weight per fake coin, and the bag from which you took out that many coins is the bag with the fake coins.

terminal_don Ask HN: Relatively less known but good blogs?

https://loper-os.org/

terminal_don Tell HN: Crypto/web3 grifters are Now AI/ML grifters

That's hilarious. Guy expects a 140IQ guy that's passionate and naive and will work for pennies so this Du Pont guy can inflate his ego. Crazy.

He'll end up with two college interns that he'll haggle with for the internship certificates / recommendations. That's what usually ends up happening.

terminal_don Tell HN: Crypto/web3 grifters are Now AI/ML grifters

> Most are no more than a saas template on the gpt3 api (or chatgpt unofficial as ‘prelaunch’) or one of the stable diffusion api’s, with a bit of lazy prompt priming showing a few cherry picked demos.

This is what really grinds my gears. No one seems to talk about how using ONE company's API for a paid service / product is VERY bad design. Using OpenAI's API to make your "product" (actually just a text template with an overdesigned website that doesn't even process data) is the height of grifting. It's similar to rolling "your own search engine" but actually passing all the queries through google. A lot of eyebrows would shoot up on that, but very few here because the marketing at OpenAI is really something. They're really great at keeping their mouths shut and letting their competitors and product take the heat. Very sneaky IMO.

Also, most of the AI offerings seem to target people that don't understand what it really is or what it does -- I'd expect a very hasty "seal of authentication" or something really soom, that'd be mandatory anywhere you'd want to "prove" your data is actually yours.