MisterKent

We're entering an era where we are being forced to ask fundamental questions about:

- what does it mean to be human?

- what is he purpose of education?

- (for us) what is the true value add of an engineer?

As we reach what I would call the Apex of the information age, information (knowledge) will become slowly worthless. But intuition, reasoning, and extrapolation seem to be worth more. Being able to ask the right questions, and apply the world's knowledge are (and were) invaluable skills.

This has been the trend of this era:

Encyclopedia Encarta made general knowledge more available, then Wikipedia, and Google indices. The bulk of human information became available. With comment sections in social media, the world's thoughts became available. With YouTube, even how-tos became universal.

If you view an LLM as a compressed archive of all human data (which it "sort of" is), you start to really separate knowledge from application.

Reasoning models take that one step further.

It will be years or decades probably until we have culturally / spiritually / philosophically caught up.

Nothing is fun anymore because:

- job market cooled

- everything is AI focused

- covid killed in office working where people could get properly excited

This is days old, but edgedb does allow generating a fully typesafe query builder package that is schema aware. At least for typescript, I haven't tried it with other languages yet.

Edgedb ends up being an extremely performant ORM if used that way. And since your database is fully aware of your object graph, it's just as or more type safe than graphql.

Definitely true, probably of almost everything.

Rarely the case, and rarely acknowledged, but almost everything is a tool. If everything followed the Unix principle rather than trying to constantly grow and pivot, we would have a lot more solutions that just work.

The more often people think about things as tools under the Unix principle, the (generally) less user hostile they are.

What an ignorant take.

https://en.m.wikipedia.org/wiki/Baseball_color_line

https://operations.nfl.com/inside-football-ops/players-legen...

Not only does that not need to be the case, he only needs to subvert a few swing states to interfere successfully.

I'm assuming this runs on your servers, otherwise you'd be bragging about it running locally?

Does that mean this will become a future paid service?

Edit: I'll add that, if this was local it would be the feature that would cause me to switch away from vscode (as a primarily typescript dev)

Your rebuttals are all agreeing with him...

I don't think you're grasping the nuance he's putting out there. Instead you're choosing to be overly pedantic over the word essential.

I've been loving edgedb in my typescript side projects for a number of years now. I have always hated ORMs, mostly for performance and a little bit for elegance / ergonomics. Curious if you have thoughts on a solution like that?

Malicious Compliance is not the right word. But, I thought about it and am slightly hopeful.

Their goal could be obedience such that

A) the Musk cronies don't come through and truly delete the data (vs, just taking it offline)

B) they don't get fired and can take a stand when "bigger" issues happen. I.e another pandemic.

Or y'know,

C) they want to keep their jobs and decided to just obey.

That's not pareto. That's just finding a niche of people who would prefer more focused products.

Tax prep software for simple returns only is an entire product. Adding support for the other 20% would lose your initial base's interest.

Tax software that aims to solve all problems whose MVP is it handles 80% of people's tax returns is the pareto the author is talking about. But the real complexity is the other 20%.

Pareto as a minimalism process for focused product development is not engineering (good or bad).

Forgetting pareto and believing (or lying) that you are truly 80% of the way there is a big problem in engineering and funding. The author is correct in that.

Allowed or encouraged?

This is the problem.

We can't be certain that a foreign actor couldn't destabilize our faith in our government by pushing pro-palestinian content.

A small push on a platform can snowball since creators take the stances that don't get them cancelled or want to mimic the popular opinion

I don't agree with the how, but the conclusion is true.

I'll elaborate a bit more though...

In a capitalist world, it will always seek to maximize profits. Human suffering, animal suffering, morality, and so on are basically irrelevant.

Someone willing to exploit human happiness for an extra 2% gain, will wipe out the competition if they are unwilling to do the same. So everything is guaranteed to spiral in a race to the bottom.

So with a new tool, as every time before, it will be a race to the bottom to see how can most unethically use AI to exploit value.

Whether we end up as slaves, or starving on the street remains to be seen. But the odds of humanity being better off without a sharp course correction and regulation by the government, are very low.

This is a really odd way to test capabilities of an LLM. First, most photos of clocks are 10:10, since the training data for watches are usually set to 10:10 (in order to better sell watches etc).

Second, I don't think the photo generation aspect of chat gpt is being marketed or presented as a problem solving AI.

Remarkable seems to be a pretty hackable device? They give SSH ability and have stated they're not removing it. I think it's a very good balance of letting the community do weird things while they focus on their core product.

I say this as someone who has not purchased one, but is considering it.

Jesus Christ, this blog is absolute trash, peddling a massive number of conspiracy theories, from vaccines causing autism to FDA doing massive coverups. Even going as far as to blame the Biden administration for things going on during the COVID vaccine period (when Trump was in power).

Honestly, just _fuck_ off with this nonsense. "I have a treasure trove of data that shows vaccines cause autism, and it can be verified. But I won't release it, but trust me."

I'm sorry the world isn't more interesting and there's no shadow government in place pulling the strings. The simple fact is people have always been autistic, and just went undiagnosed. I could even see arguments for it being less apparent when those children didn't have the comfort of the Internet to avoid developing their social skills etc.

/Rant.

Search tiananmen on any app and then on TikTok. Regardless of their testimony, it doesn't take much to show that it isn't 100% of the truth.

Wow, that's gotta be the first release that I've ever said "nothing interesting here" to.

I say this as a big typescript fan, who leverages tons of advanced features and typings.

Although, I have felt for a while that typescript is approaching "complete" in what anyone would want to do. There's some awkward ergonomics when doing awkward things, but awkward things should probably feel awkward.

Still, glad to see they're still improving on catching all the "weird" js things people can do.

Interesting, and agrees with the anecdotal trend I've seen playing out the last few years.

Summarized as: you must appeal to a niche (cheapest, healthiest, best tasting). Brands matter less.

I remember a few years ago reading that Tide detergent is one of the most recession proof items in the store. People are due hard loyal to it.

What's shocking to me, but that this point shouldn't be, why did it take Kraft decades to try different flavors of Mac and cheese? Was the landscape and desire to be more competitive really so weak?

Store brands are made in the same factories, with the same ingredients. And they're still profitable at 1/2 the price. Makes you wonder how much you've been spending on just brand names.

Fitness foods are interesting because there's so many trends and niches: protein, gluten free, low carb, dairy free. And people pay through the nose for them.

Is it own feature good? Let's look at our data that we're releasing on our blog.

Also: functionality is _not_ a good metric of higher quality code. Any interviewer can tell you that.

We just use LaunchDarkly for config and S3 behind a CDN for assets. S3s Ux is serviceable.

LD (and the other feature flag services), allow flighting, population management etc. as well.

It's unclear to me, where the value is here as a "config" solution vs the feature flag solution they offer.

Agree with the above, except:

Sharing solutions in slack works until stuff becomes impossible to find.

Using confluence or some type of team documentation feature can help with that stuff. Better yet, keep it very low effort to add docs there, so people can copy paste important (long lived) messages there.

As an audiophile, I could see why some would see this as a horrible thing. But I'm not really worried.

McIntosh has been a higher end Bose for a while now. They live on their brand name and their blue glow, but amplifiers and dacs, really most electronics, have been a solved problem. And that solution has been slowly dropping in price for a while now. McIntosh's gear has barely moved, which is because they're way more fashion over function.

Bose, if they're smart, will leave the brand intact and use their connections with car manufacturers etc to make them more prevalent. But, there's already such a huge margin on their products, I think it'll be a while before they try and cannibalize that for profit.

Afford doesn't mean the same thing as able to pay for in cash right now. Otherwise you'd say people can't afford houses, when you really mean people can't afford / are unable to get a mortgage that will cover their house.

Nobody forced these people to take our loans. The auto industry bends to trends, it doesn't create them.

The fact is, with or without marketing, people want nice things. And people have historically been willing to buy them on credit/loans. At least for our current capitalist society, that seems to be ab inherent feature.

People in this thread seem to somehow think marketing is pushing people to want xyz.

That _cannot_ be the singular or even primary driver of a movement. Car manufacturers don't just make a car in a vacuum. They make what the market is asking for, and _then_ marketing steps in. Very few companies, like apple and Tesla, can foist weird stuff on consumers and have them like it. And even in those cases, they are really just building something they believe to be the faster horse.

There's a reason Tesla wanted to build and SUV and a model 3. There's a key target price point in the US for a car. No amount of marketing can make people buy a car they simply cannot afford.

Personally, I don't see a huge upsell over JSON for interop. Typing is great, but largely mitigated by things like typescript, graphql generators etc etc. edit: and json schema

The biggest issue I see with JSON today is the inability to self reference values. Data duplication overhead is high, and deserializing doesn't work as expected if you want two objects to be reference equal.

Also, nitpick: that syntax would be a grind to write manually vs JSON.

Why even comment if you're not going to read TFA which actively disputes your "isn't it just" assertion in its title?

It's very simple: what sells on the showroom floor and what works at home aren't necessarily the same thing.

There's another comment referring to Elon. I'll leave that aside, but I think the same analysis can be done for him.

Clearly the population is split on wether or not intent matters when it comes to philanthropy. Some people say, it is only respectable if you're doing it out of the goodness of your heart, not for some ulterior motive (like repairing/boosting your image). On the other side, you could argue that the majority of philanthropy is done for image improvement, and without that benefit millions of people would be left without help. Hoping for "true altruists" to save us is probably never going to work.

True altruists that can gain enough power and resources to make a difference is a paradox. I need 10 billion dollars before meaningful impact happens (random number, but the point stands _a lot of money_). That's more money that any altruistic person would ever get to, since they'd start donating at the 100million or less mark.

The closest we have to a true altruistic person with power is someone like Mackenzie Bezos. No derision here, but she didn't have "what it takes" to get there, but she does have the desire to help. Which is also why we don't hear much about her donations, compared I Mr. Beast, despite her giving orders of magnitude more money away. And more correctly.

Then, we could look one level higher, how do these people acquire their wealth? Clearly selling garbage food to kids, creating fake giveaway videos, and other things isn't exactly helping the world along the way. Making electric cars cool / hip? That's pretty good. Embrace, extend, extinguish? Terrible.

Holistically, Mr. Beast is earning his money off of children and "legal scams". Then, fixing his reputation by donation pennies on the dollar of that back to the world. Overall, probably not a great person. Couple that with how conscious he is of what he's doing based on various leaks, and is say definitely not someone I'd categorize as deserving of any adulation. Obviously not actively evil, but definitely overall harmful to society in his pursuit of fame and wealth.

Aside: how isn't this titled "MrBeast's Faux-lanthropy"?

People here are incredibly hard to please. Very clearly a packaging issue that got blown out of proportion.

They've done largely the right things for _years_ in terms of security. They've operated pretty transparently in terms of open sourcing. They've allowed vaultwarden to exist, and eventually created a self hostable version as well.

But one bad release with a license screw up and nobody is willing to give them an inch?

I will continue to use bitwarden, and am willing to give them the benefit of the doubt. Especially considering this action above. They are a company that is perfectly toeing the free/oss and commercial line.