I do all my work in Rust now via o3-mini-high and convert to WASM... JS just for DOM and event handling. What is the point of building these CPU-bound functions in TS. Why not Rust->WASM?
What I'm asking really is what is the benefit of TS if we could use decent and improving coding AIs to help us code in Rust, even if we are new to it, or never used it, and compile to WASM. Much faster execution per my experience. I mean not even comparable.
It's nice for performance, but if that's not a bottleneck, Typescript is pretty convenient. It has better tooling for the web, and compiles almost instantly. Rust -> WASM can be frustratingly slow if you're exploring a new idea or prototyping.
After some discussion with some friends from the former "colonizer", it just occurred to me that apparently it's very hard for people from those countries to appreciate their countries' role for a lot of massacre, genocide or any man made disaster like this kind of famine.
They always find ways to deflect their countries responsibilities with some "reasons", although they generally agree that any kind of genocide is wrong.
I think this is what we witness today too, with some massacres and genocides going on. People from those colonizer countries just can't relate to the victims.
Maybe deep down they acknowledge that those genocides are good, or at least necessary, because those things are what brings them prosperity they enjoy today.
“The British” did not genetically engineer the potato blight.
Further, in the 19th century state capacity was small and massive modern style relief programmes were not possible. Despite this, Britain managed to spend a large degree of GDP on relief. Proportionately more than it did on covid response recently, for example.
The reason people deny british culpability for “genocide” is that there was no “genocide” and britain did what it was able to do, to an unprecedented degree in fact. If anything, we should be proud of britain’s response, especially knowing that it would never get aby kind of gratitude for it.
>If anything, we should be proud of britain’s response, especially knowing that it would never get aby kind of gratitude for it.
Some examples of Britain's response:
* Establishing soup kitchens for the starving, where, to acquire food one must renounce your religion, anglicise your name, and abandon your native tongue.
* Provide maize for the starving and destitute but not for free for fear it would generate a sense of self-importance amongst the millions who are dying of hunger
* Maintaining the exportation vast amounts of food to Britain throughout the Great Hunger
* Requiring the starving who couldn't afford to buy food from the British to build pointless walls in order to earn that food
* Forcibly evicting the starving and dying from their homes because they couldn't pay their rent for some reason
* Denying aid to anyone who owned more than a quarter-acre of land, forcing starving farmers to give up their land and become destitute in order to qualify for relief
So, on behalf of all those before me in Ireland; go raibh maith agat.
The entire reason there was a famine was due to absentee landlords who demanded absolutely everything but the bare minimum from farmers for the "right" to work on "their" land.
My view is that those who designed CSS suck when it comes to designing intuitive systems.
I would not advocate constraint solvers, in the future I hope ViTs are so cheap to run that they can infer the right layout of things at any orientation and size in single digit milliseconds, solving the layout problem for good =)
Organic growth leads to vestigial warts - that is the price of success and development. I'm really impressed with so much of CSS and what can be achieved with what it has become.
> My view is that those who designed CSS suck
I think that is a stink attitude because there is no need to malign people. Don't be a dick. Most successful things "suck" because it is easy to feel the compromises when you use something. It is incredible hard to see the warts before we build, and even harder to find consensus solutions. All too often I hear whingers, that lack the ability to deliver working solutions, who are often unrealistically idealistic and too quick to poopoo the work of others. They are the people with second-system syndrome - who often deliver a version 2.0 using technology B and get an outcome all too often far worse than v1.0.
Agreed that there is no need to be a jerk on this discussion. The points you make are generally true in the abstract, but we are talking about CSS here which really is one of the most convoluted and confusing pieces of UI software out there. And the problems are not cruft from success and time: the core parts of the spec are terrible and confusing. The basic idea of a Cascade is a bad idea for a variety of reasons and nobody uses this feature if they can avoid it. The over all layout spec is really convoluted and requires you to figure out the difference between absolute and fixed position and top vs bottom. (Making a horizontal row of images is shockingly difficult). There is also a whole system of units for "absolute length" that are defined in relation to the inch (including the terribly named pixel which is 1/96th of an inch (inches being defined for printers but not for screens). These are issues were all features of the very first release. The later updates have provided alternatives, but the core spec sets people up to fail.
It is so easy to look backwards and imagine a better world and then moan about imperfections. You can do that with absolutely any technology.
Most of your complaints have been solved, or there are best practices that avoid the downsides.
> Cascade is a bad idea for a variety of reasons
CSS is fucking genius. It is flexible enough that we have solutions that avoid the problems. I love the idea of an alternate universe where we do a redesign: but personally I'm very unfond of some of the (now rightly dead) systems designed to "fix" CSS.
> The over all layout spec is really convoluted
The domain is difficult: we just think it could be simplified because we are given CSS and so we think we could improve upon it.
> absolute
Don't use it. It's a bad design smell. Rarely needed in a few corner cases. A nightmare before getBoundingClientRect() was reliably usable (a decade ago)
> fixed position
Is damn fucking awesome! I developed butter smooth fixed/sticky positioning that worked in IE5.5 to IE11 using absolute instead. I developed a zero-jank fixed position feature in IE but there were insanely complex requirements (global order of events with DOM updates plus needs extremely complex techniques to deal with viewport scrollbars - only a stubborn idiot engineer like me would attempt to yak shave it but I solved it).
fixed positioning was hell in early iOS (broken zoom, broken events, broken inputs, jankyness, scrolled offscreen). Let's not talk about early Android browsers. Arrrgh fixed positioning (and everything else) was broken in Edge (Spartan/Legacy). Yet fixed positioning is a critical feature for mobile screens - a major new feature that CSS managed to embrace.
> Making a horizontal row of images is shockingly difficult)
That's what a <table> was for. Many people searched for difficult solutions: no surprise that they had troubles of their own making. Now solved in more flexible ways with modern CSS layout features (there's that amazingly flexible capability of CSS to absorb new features again!)
> units for "absolute length" These are issues were all features of the very first release.
Which was used because we didn't have modern high resolution screens. Layout units are just extremely difficult (demonstrated by the variety of different units we now have - made available by more powerful computers).
It is just too easy to use 20/20 hindsight to criticise every historical decision.
I feel your pain: I dealt with every quirk in every browser for far too many years.
It is healthier to thank the people that worked on the specifications: awesome technology that is astonishingly flexible (and quite frankly amazingly backwards compatible). Yes, you need to learn how to use the features. No, that isn't a problem. Most anti-features can now be ignored when developing a greenfield site.
I'm extremely thankful that I never have to make a page work on IE (or many other now obsolete browsers) ever again.
There are a huge list of awful gremlins in CSS you didn't mention - luckily most developers can avoid the nastiest of them (zoom: 1; to fix layout?). There are also some truly inspiring CSS features: declarative transitions and animation seem like magic to me (complex, but deliver amazing results) with hidden risks of course: https://news.ycombinator.com/item?id=43259722
I guess CSS is the worst form of layout language except for all the others that have been tried.
This is all BS tbh. People don’t know how to use the current gen AI take to do very useful reasoning.
I keep posting our work as an example and NO ONE here (Old HN is dead) has managed to point out any reasoning issues (we redacted the in-between thinking most recently like the thinking traces that people were treating as the final answer)
I dare you to tell me this is not useful when we are signing up customers daily for trial:
You sound grumpy but I followed the link because I was curious. It’s super hard to use on mobile and I can’t really tell what it is even clicking and swiping a little. What is it? Congrats on the signups.
I clicked your link and have absolutely no idea what I am looking at. There is overlapping text, obscure words without definition, terribly confusing UX.
What is this?
Who is it for?
Why do you think this is demonstrates the OP is BS?
Read the reasoning after clicking in the stats or swipe to next page and click on the donuts. For in-house legal teams not for mobile users. Read the statements and reasoning. Tell me if the reasoning is wrong. One of the top 100 lawyers in US looked at it and told us he likes how it balances the intent of the law and no gripes with reasoning. Not using a reasoning model unde the hood. We built one for legal. It means that gen AI is useful and LLMs can be made to reason with inference time scaling. It’s not rocket science but also not easy. What hype? We think gen AI is actually under hyped.
What is this link supposed to illustrate? It doesn't render properly in Firefox - some CSS glitch, elements all on top of one another. Was this produced by AI?
it's a good thing for the creator that their dead reply is default hidden. I'd hate to work with someone who's response to my blunt, possibly rude comment is to call me stupid. whether or not I am, that's just not a good look.
I found Gaussian Processes with the right kernel to be very powerful with even just a few data points and a very small set of parameters. I don't know if I was using it correctly tbh, but it worked out great in predicting values that I could not predict so accurately. I used it as a predictable yet non-linear process to tweak the input in a computer vision task. The proof was literally in the pudding.
LLMs don't reason the way we do, but there are similarities at the cognitive pre-conscious level.
I made a challenge to various lawyers and the Stanford Codex (no one took the bait yet) to find critical mistakes in the "reasoning" of our Legal AI. One former attorney general told us that he likes how it balances the intent of the law. Sample output (scroll and click on stats and the donuts on the second slide):
I built the AI using an inference-time=scaling approach that I evolved over a year's time, and it is based on Llama for now, but could be replace with any major foundational model.
"One former attorney general told us that he likes how it balances the intent of the law."
In a common law system you generally want actionable legal advice based on predictions on how a judge would rule in a case not "balances the intent of the law" whatever the heck that means.
The sensitivity can be turned up or down. It's why we are asking for input. If you're talking about the Disney EULA, it has the context that it is a browsewrap agreement. The setting for material omission is very greedy right now, and we could find a happy middle.
A former attorney general is taking it for a spin, and has said great things about it so far. One of the top 100 lawyers in the US. HN has turned into a pit of hate. WTF all this hate for? People just really angry at AI, it seems. JFC, Grow up.
I built this AI tool to spot "bugs" in legal agreements, which is harder than spotting errors in research papers because the law is open textured and self contradicting in many places. But no one seems to care about it on HN, Gladly, our early trial customers are really blown away by it.
Part of me thinks that that before they gave the $130M they knew exactly how they'd get it back (thru some pre-arranged M&A) if it doesn't work out. Or at least that would be the smart thing to do.
I spent $10 on a bad apple and it got $0 back. Then I go and sell that bad apple for $10. Whomever bought it is giving me back my money. This is kindergarten level arithmetic.
I didn't mean that it's mathematically impossible. I meant that it's practically not gonna happen. Much as they'd love to, no VC gets a chance to invest in an early stage startup knowing that if the startup fails, there's a buyer ready to make the VC whole.
