Silica uses femtosecond lasers for writing and supposedly mostly ordinary silicate glass as the medium. This would probably mean more costly drives and severely cheaper media (Folio claims $3/TB now, $1/TB eventually. I wager Silica would be at least 10x cheaper). If Silica keeps its promises, we might be looking at archival prices cratering. But it's been in development for over a decade, I wouldn't hold my breath.

Really interesting would be R/W speeds.

> All type parameters have an implicit bound of Sized. The special syntax ?Sized can be used to remove this bound if it’s not appropriate.

, which HashMap does not do, i.e. the keys and values have to have a statically known size.

This also applies to reading restaurant reviews before visiting. Or for that matter, reading any published book you outsource judgement to the publishing house (what books they pick up). Or, if you are easily convinced by jargon: Exploration has a non-zero opportunity cost.

This is HN where we discuss specific things such as articles, blog posts and tweets. There is often good insight in the comments but mostly it’s people talking about their opinions. I’d rather review the item myself first and then come for the extra nuggets that might be in the comments.

Likening this to restaurant reviews and publishing houses is disingenuous. One is nothing like this and the other is not helpful in that it’s such a generalization of influence as to be not useful. Everything is an influence doesn’t mean anything.

Read the article. The elaborator (where metaprogramming is evaluated) is not part of the trusted computing base. The security model is that bad proof terms get rejected, not that they're never generated.

There is one example that succeeded and became the most widely used programming paradigm on the planet: Spreadsheets

Though you might argue that the "programming language" part was rather accidental

And interestingly using Excel resembles pure functional programming in a lot of ways

And it has array programming features, logging, and (time-travel!) debugging all intuitively graspable for most non-technical people.

I sincerely think there are lots of unlearned lessons in studying Excel.

And that is without counting on Lambdas, VBA, Power Query.

It’s reactive programming in a functional style. I wouldn’t call it functional reactive programming though because that’s usually used to describe more robust languages like older versions of Elm before TEA.

> How's this any different than say, not bringing your wallet with you, because 95% of the time apple pay works

Apple Pay involves at least a third party (your bank) as a part of its function. This entity has to be reachable.

Unlocking my car needs me, and my car.

> when the payment terminal randomly decides to not accept apple pay?

I would be justified in being mad if, randomly, my BMW car would decide to not accept a BMW key (virtualized or otherwise).

> Not to mention there are plenty of other ways a phone can fail. It can [...]

"I will not accept this Euro bank note. You could have dropped it into the gutter on your way here, so it's strange for you to be mad at me if you don't have a backup."

The point isn't that it's fine for smartphone car keys to randomly fail, it's that there's plenty of other reasons they can fail that you're already living on the edge if all you're carrying is your phone. Therefore the "it's dangerous to assume ..." claim doesn't hold.

> there's plenty of other reasons they can fail that you're already living on the edge if all you're carrying is your phone

I see what you're arguing, but I just want to point out 2 things:

1. If I leave the house for a trip (especially a distant trip) with just my phone, I and anyone would understand intuitively my own responsibilities: namely, to keep my phone intact and its battery from draining completely. Since I already have a strong interest in both of these for other reasons, this is basically automatic anyway, and a backup plan for the latter is completely doable - I'd have to drop $20 on a gas station phone charger and hang out somewhere for 10 minutes while it goes from 0->6% or whatever. Anyway, if I understand my responsibilities like this, I'm able to make an informed decision on whether to trust it.

2. If we're victim-blaming here, saying "Serves him right, should have carried the keyfob as a backup" that renders it pointless to even have phone-as-key. The fob not having to be in your pocket or handbag is the only benefit. Unless, I suppose, you consider phone-as-key as the actual backup, for use only in the case you drop your keyfob into a storm drain. But that isn't how it's marketed at all. In my entire life, I've literally never lost my car keys, because they just stay in my pocket the whole time I'm away from home. So I'd value a 'phone-as-backup' at $0.

> Apple Pay involves at least a third party (your bank) as a part of its function. This entity has to be reachable.

No it doesn't. It has to have a recent token, and if it does not it will contact your bank.

Apple Pay even works if the phone/watch battery is "dead"

Only for Express Transit though, right? Not normal payment cards.

Once I start a conversation about Postgres query plans, maybe 95% of the knowledge will be almost certainly not needed, and for an inference provider there will be many more concurrent queries with reasonably large overlap. Maybe future architectures will be able to take advantage of this so not all parameters are needed in memory for almost all instances.

If we are talking about all knowledge, then I agree that the compression ratio is very impressive already.

Mixture of Experts is already used by pretty much all modern LLMs to address exactly this phenomenon.

Hopefully, future models can be trained to be even more aware of external knowledge, accessible through web search / RAG / whatever it will be then, and might not need to internalize much knowledge at all.

> future models can be trained to be even more aware of external knowledge

Then you need longer contexts, which is proving to a much more stubborn problem than general knowledge compression.

I suppose the right approach would be to run queries like the ones you want your model to be able to do, sort weights/experts by usage frequency and then reareange the weight so that all but the most frequently used can stay on disk. Tricky though, it could be that a e.g. database question uses 90% of the network at some point or another

Edit: I guess at the moment this is just having an LRU cache of experts

Note that MoE is a sparsification mechanism and doesn’t actually have to do with expertise or what would commonly be considered areas of expertise.

I wonder if the EU will someday figure out a patch for this "Deteriorate everyone's user experience in a way that is technically compliant and blame the EU" behavioural exploit...

I'm blaming Google for this, but most people just don't know why it happens. Probably they'll end up mandating that Google must show a notice with the reason and it will be implemented in the worst possible way.

All of these hissy fits Google and Apple are throwing are just making me wanting to use them less and less and I'm doing my best to explain how tech companies are doing malicious compliance to others.

How about EU starts legistlating towards a competitive market and creates its own web search Tech Giant ?

So Google is forced to take competitor to account or will be pushed out of the market ?

No? Better to complain and do nothing positive? Welcome to EU Parliment.

How about Google obeys the law and stops abusing their position?

The corrective force should be market competition, but Google is a monopoly not a monopoly with so many moats it might as well be a monopoly.

They also buy up the competition before it gets a chance as is the way now.

Why should be? I think it is abundantly clear by now that market competition alone will not fix this. Though that does not stop people from claiming "Yes that failed and people suffered/died, but that was not real capitalism! If only we had a real free market, it would have been better!"

The European Commission is way too pro-business for that, even when the said businesses are parasitic actors who don't even pay taxes in the EU.

What does it has to do with the topic? And saying “governments do not build” is a crazy antihistorical take: many things, from the space industry to the telco one were built by governments, as well as the energy and transport sectors in many countries.

For the past 50-ish years the West convinced itself that government should not build, and as a result we're suffering from a slow collapse of basic infrastructures. Meanwhile, in China and the Gulf, government kept building and the world's center of gravity steadily keeps moving eastwards.

"What have the Romans ever done for us?"

I merely illustrate that the EU commission certainly has the means to replace Google in the EU. Build a new Youtube. Build the services that society, frankly, depends on now ... which is quite a few of them.

And they're not doing that. They are only making it harder for existing services to exist, in order to change the situation, and in order to make money for themselves. But they're not helping in the sense of building anything.

Yeah, it's very overpowered and should be nerfed - they still haven't patched the cookie consent dialogs out.

To expand upon the silly comment a bit, not sure about this particular initiative, but to me it very much feels like something that should be configurable at browser level, instead of half-assed dark-pattern-filled JS dialogs: https://killthecookiebanner.eu/

It definitely should be, but this just goes to show that regulations are necessary if we don't want companies to unnecessarily screw with users. It would have been better that way for every user, but it would be worse for people who want to sell everyone's personal information.

Oh, and please ignore that this surveillance apparatus is now being used to expell people, all of which already has resulted in many missing people, and even some deaths. Any historical parallels are entirely coincidental.

It's fine + obligation to fix, and daily fine until fixed, and higher fine if you do it again.

I know that sounds like it comes from "it will never work", but that's how active directory, office file format, etc ... got opened, and in a case more similar to here that's how windows n and decoupling of media and internet component from windows internals happened.

The issue is, yes, it takes time.

Another issue, purely PR, is that yes the EU gets attacked repeatedly by people trying to match "respect the law of the market or leave" with "the EU tax US companies because they can't compete".

> It's fine + obligation to fix, and daily fine until fixed, and higher fine if you do it again.

The important proviso is that the penalties need to (on some non-geological timescale) get high enough that the offender is no longer capable of operating. That could mean the fines reach $500 billion, or it could mean the company is barred from operating in the EU, or it could mean people get arrested and assets are seized. But unless the penalties become crippling, it won't matter. It needs to reach a point where the downsides of noncompliance are actually greater than the benefits.

NIS2 allows for the arrest of managers in case of cybersecurity incidents resulting from negligence. I think it's a step in the right direction, but we'll have to see how it plays out.

In the EU they like the "daily fine until fixed". It's how Microsoft bowed down.

But unlike the story some on HN and in the US like to think, US tech companies don't ever get there except rare exception, they know the game.

The solution is to mandate that websites are built for browsers/devices as much as they are for humans. That way you never see a cookie banner because you've set up your device to never accept marketing cookies; A Google Maps link clicks through to Google Maps because you told your browser Google Maps was your preferred online mapping tool.

Once once they are rejected by charities, they can be destroyed just like before this law. EU regulators are flawed humans, but you can be assured that any argument you can come up with in 10 minutes has been considered.

Fair enough, but now I would "solve for the equilibrium" and assume that any large clothing company will have a friendly charity to reject its clothes for it.

... than by surprise, when you find your integer column contains "[object Object]"? Jokes aside, breaking backwards compatibility is obviously a non-starter.