Hacker Newsnew | past | comments | ask | show | jobs | submit | starlevel003's commentslogin


answer: use Trio (or AnyIO if you're stuck in the asyncio land) and get sane cancel mechanics built-in.


> That RFC is accepted, and this is starting to happen.

> Progress has been disappointingly slow,

I don't think there's ever been a more concise summary of Rust.


LOL some people complain that Rust is moving too slow, and others that Rust is moving too fast and adds too many features. Can never make everyone happy...


Depending on the scenario, I don’t hear mutual exclusion there.

It can be moving too slow on important things, and too fast on unimportant or volatile things.


The problem is the things people think are important or unimportant is not universal.


Good projects with bad leadership die ever day.

Hopefully Rust has people with vision and drive in places to let it escape its niche status.


I just filled out the Rust survey [1] and may have done both—iirc there were checkboxes for missing features and concerns the language is getting too complicated. It's a hard balance to find.

[1] https://blog.rust-lang.org/2023/12/18/survey-launch.html


This is HN, so the answer to that question should be very obvious.


> we lived thousands of years fine without the FDA

For those thousands of years our only medicine was "ground up unicorn toe".


I did not know that medecine was a 20th century science and that we lived in dark ages until the year 1899


We effectively did.

Look around where you are. Unless you're in the middle of a forest, pretty much every single thing you see is the result of scientific and technological advancements of the last 150 years. Half of it probably owes its existence to the oft forgotten chemical and petrochemical revolutions that started about 100 years ago.

Before that? Before the tail end of 19th century, we knew fuck all about anything. Medicine, in particular, graduated from voodoo and aforementioned "ground up unicorn toe" into a proper science and profession, also about 100 years ago.

It's really hard to overstate how different 20th century science and technology is from anything that came before. We've crossed a qualitative threshold there.


> Before that? Before the tail end of 19th century, we knew fuck all about anything. Medicine, in particular, graduated from voodoo and aforementioned "ground up unicorn toe" into a proper science and profession, also about 100 years ago.

Nevermind the thousands of plants with verified therapeutical effects across the world for centuries.

But yeah, old societies were completely clueless and a bunch of idiots and surely we have learnt nothing from them.


Yes we should absolutely go back to keeping our humors balanced, bleeding people, treating syphilis with mercury, dying of an infection from a minor injury and insane level of mother and infant mortality.


You're setting up a false equivalence when you imply that our position is that older societies were completely clueless.

Nobody is saying that. What we're saying is that the application of science to medicine and nutrition had an absolutely enormous effect, and before that, we didn't really have explanations- just empirical observations- of the effect of medicines.

https://en.wikipedia.org/wiki/Spontaneous_generation


As late as 1899, taking out a living person's heart couldn't be anything besides a gruesome execution. There wasn't the understanding to allow it to be survivable, and even the anaesthesia at the time was not really good enough for anything like that.

Now? Though still rare, it's just another thing you can have transplanted if you need it.


> But yeah, old societies were completely clueless and a bunch of idiots and surely we have learnt nothing from them.

Yes, they were, in comparison. They had neither predictive theoretical models, nor the tools to confirm and refine them - hell, for most history, they didn't even understand they need those things in the first place. Please appreciate that the period starting in late 19th century brought all the pieces together - the theoretical foundations, the observations, communications, precision manufacturing, the philosophy of science - they all added up to exponential growth across all scientific disciplines, and all occupations.

That's not to say we didn't learn anything from them. They did accumulate knowledge. But the sum of all knowledge humanity accumulated since beginning of recorded history is just a rounding error compared to what humanity accumulated in the last 100 years. That's the exponential growth at work.


Pretty close, maybe more like mid-19th century as far as the "chemistry side" is concerned.


I'd push things back to around 1800, when the study of medicine became more scientific.


You should try reading up on it then. Like how life expectancy absolutly sky rocketed from the end of the 19th century.


I guess this is sarcastic? But yes, things have gotten a lot better since we invented antibiotics, modern surgery, routine childbirth, and other lifesaving medical techniques...


Man, I didn't realise HTTP alternatives were so serious.


Python does not use semantic versioning and never has.


Hopefully all that money goes into making the GTK file picker better


I think I remember at some point reading that the end goal is to have file picker just be "a mode" in Nautilus instead of a separate thing

Edit: found the discussion I think I saw https://discourse.gnome.org/t/nautilus-as-file-picker/11605


The proliferation of requirements.txt is one of the key reasons why Python packaging sucks so much.


Right what we need is a requirements.yaml (better yet, create an entirely new markup language for this particular project) and another new package manager for it. One day (one day!) I will start a project without python. One can hope.


> Right what we need is a requirements.yaml

It already exists, it's called pyproject.toml. It already existed for years in the form of setup.py. Requirements.txt means that projects can't be automatically installed which contributes massively to the difficulty of getting packages to work.


Pyproject.toml is a right step afaict but man is it complicated: "Please note that some of these configurations are deprecated, obsolete or at least discouraged, but they are made available to ensure portability." Core vs setuptools-specific etc. See https://setuptools.pypa.io/en/latest/userguide/pyproject_con... and https://packaging.python.org/en/latest/specifications/declar...


That was a Simpsons episode. S9E22 Trash of the Titans.


Ah thanks, damn. Mixed up my adult animation. I swear I can picture the entire scene as Family Guy characters in my head though.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: