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

It's worse because many humans don't know they are.

I see a lot of outrage around fake posts already. People want to believe bad things from the other tribes.

And we are going to feed them with it, endlessly.


Did you think the same thing when photoshop came out?

It's relatively trivial to photoshop misinformation in a really powerful and undetectable way- but I don't see (legitimate) instances of groundbreaking news over a fake photo of the president or a CEO etc doing something nefarious. Why is AI different just because it's audio/video?


"AI" is different because it's low-effort and easily automated, making it easy to absolutely flood public spaces. Quantity has a quality all its own.


I did.

And it's not the grounbreaking the problem, it's the little constant lies.

Last week a photoshopped Musk tweet was going around, people getting all up in arms against it despite the fact it was very easy to spot as a fabricated one.

People didn't care, they hate the guy, they just wanted to fuel their hate more.

The whole planet run on fake content, magazin covers, food packaging, instagram pics of places that never looks that way...

And now, with AI, you can automate it and scale it up.

People are not ready. And in fact, they don't want to be.


The people disliking the language are very vocal about it, but there is a huge amount of silent people that loves it and an even bigger amount that just like it as much as alternatives. It's mainstream now, not trending like 10 years ago, so there is no hype about it anymore. We just use it to do stuff.

Add to that the existing excellent ecosystem, the strong culture of scientific stacks and a very good story for providing c-extentions (actually the best one in all scripting languages because of things like cibuildwheel).

It's only in small tech bubbles like HN that devs find it surprising.


Python has many issues that are quite clear when you operate at some kind of scale and need proper multiprocessing/multithreading support. And its not just the GIL, you get very unexpected behaviors when dealing with exit handlers and signal handlers in edge cases. Having seen what other languages look like it just doesn’t feel like a language that was designed well for running at scale.

The tooling has markedly improved though. Things like typing and compile time checks, great. But its also funny to me that some of the fastest tools for python are being built in rust (eg uv).


I’ve always found Python to be sort of loved on HN. Not by everyone or course but I guess it depends on each of our experiences on here. I’m usually rather surprised when I meet people who genuinely dislike Python, because that seems like such an odd occurrence. Even if people don’t “love” the language most people seem to have had rather fond experiences or memories of it. Usually criticism comes down to its inefficiencies, but those aren’t exactly unreasonable critiques.

As I said it’s anecdotal, but in my experience Python gets a lot of love compared to something like Java or C#. Both of which are often met with real harshness. Hell I’ve ranted unseemly about C# myself.


I mean, even on HN, I'd say if there's derision, it's mostly one uttered with a yawn rather than genuine hate. And that's almost justified; while I spend a lot of my time with lots of different languages (I can't think of a single one I outright hate btw), Python is the one that pays for my things and... Well, there's not much drama there is there (now that we're lost 2->3 anyway)? It's a glue language that's easy to learn, but offers tons of depth should you want it. My primary annoyance at Python used to be the typing, but type annotations have made this less of an issue. It's a nice language and you can do almost everything with it. It's a bit boring, but I guess that's a good thing.


My only big critism is the CPython folks resistance to any kind of performance improvements, and the way PyPy efforts have been largely ignored, making Python the last major dynamic language to finally start caring about performance and having a JIT in the box.

Finally thanks to data science, and people getting fed up with always writing bindings, this is changing, and Python can join Common Lisp, Scheme, Smalltalk, SELF, JavaScript, Ruby, Lua, Dylan, Julia, BASIC club.


To be fair, Ruby YJIT can still somehow be slower than Python 3.11+ at times. Lua JIT is a bit of a separate effort and the ones that truly try the hardest here are Julia (which is not necessarily interpreted, it's really difficult to call it a pure scripting language at this point) and JavaScript engine implementations. I can see pretty good numbers for SBCL on BenchmarksGame too.


Yes, but it isn't the only one, the oldest Ruby JIT goes back to Ruby Motion, still being sold.

I would not put Julia and Common Lisp on the scripting basket yep.


and elisp. Let's not forget that elisp got a JIT (well, actually a static compiler) before CPython did!


Indeed, I only listed the languages I know more about, it has been a long time since XEmacs was my poor man's IDE on UNIX systems.


The people disliking the language might also be the people that love it. I know I am in that camp.

It is a great language in many ways, that's why its shortcomings are so painful.


No he just made a mistake.

As usual, readers reading too much into something.


I bet you're fun at parties.


I know it's unrelated because this has an actual explanation, but I had an ex that stopped watches simply by wearing them for a while.

The model didn't matter, give them a few weeks, and they would stop. Put them aside for a while, they started back.

Never could figure out why, no particular behavior emerged, she changed jobs and houses and I couldn't see a pattern.

Fun that life is still full of weird stuff like this.


Reminds me of the Pauli effect https://en.wikipedia.org/wiki/Pauli_effect


> the supposed tendency of technical equipment to encounter critical failure in the presence of certain people

Is there an inverse of this?

Often times, people ask me to help them do something on the computer or with other computer related or electronic things. I walk over, touch the equipment without actually doing anything and their problem is instantly solved. Could be a printer that won’t print, even!

I don’t see any links in the See Also section about any “effect” like this.


That would be "Demonstration Effect" or "IT Support Effect". Happens every other support call.

I bet this is why half the older companies still have a tech person on staff. :-)


Fascinating. Did she work in medical / hard science? I could imagine some exposure to radiation / magnetization could do this.


Nope. Hotel, then airplanes.


Maybe using their watch hand to magnetize the hotel key cards?


I thought about it but their colleagues would have had the same problem, yet didn't. Plus becomming flight attendant removed that activity.

Not to mention it happened during the hollidays too.

It's probably more life style related, but I was there during said life style and couldn't figure it out.


Maybe some sort of magnet phone mount.


Even if the companies took security seriously (which they don't since it rarely affects their bottom line, even in the event of a breach), it's one of the hardest topics in IT, and there are very few people actually competent in it.

With the explosion of the number of those boxes, we added more surface of attack, but the number of good security people didn't increase.



Awesome, thanks.

I think @flexterra (aka gcollazo) should also add `"check_same_thread": False` to the recommended OPTIONS, right? Unless they left it out intentionally?


Well, you gotta balance all that tax evasion with something.


Once it ends, we will buy more GPUs because any small website will want its own.

I know that as soon as I can output 100req/s on the cheap on a llama-level model I will put it EVERYWHERE. And my clients too.

DMCA handling? Content flagging alerts? Fuzzy categorization? Natural UI for end user complex queries?

All LLM baby.

And much, much more.


Swift won't be able to compete with rust until it has a great scripting story.

Rust is currently the defactor choice if you want to make something with js/python bindings, of id you want to speed up a dynamic bottleneck because it makes it super easy.

Swift doesn't even have a good ffi.


Swift can import C (and C++) modules directly, and it's fairly straightforward to wrap existing C APIs with callbacks to be usable from structured concurrency. Here's an example I just worked on. [1]

[1] https://github.com/PADL/NetLinkSwift


This is not what holding swift.

What's holding swift is not having a good story to call a swift compiled extension from Python/JS/Ruby/PHP.

Python's community is huge, very active, Python is everywhere, and it needs compiled extensions. And tooling to make them BFF, like pyo3 and maturin, is great

And that's how we got cryptography, pydantic core, polar... Which motives even tooling to be written in rust as a side effect, and how we have uv and ruff.

The scripting community is the most active, if you got them on your side, you get access to a huge pool of devs.


I'm not sure scripting-language interop is holding back Swift, but ...

> call a swift compiled extension from Python/JS/Ruby/PHP

The story there is excellent.

Here's a Swift function callable from C

    @cdecl("myfunc")
    public static func f(){...}
C/C++ types are mostly handled automatically (including sharing memory management), but it can be tricky.

To build a library to load and call from any language:

    swiftc mine.swift -emit-library -o libmine.dylib 
E.g., calling from java 22 FFI:

https://foojay.io/today/java-panama-polyglot-swift-part-2/

But most people instead use a slower json server network interface because (sadly) json is the lingua franca for scripting and Swift Codable makes it trivial to map the json to a swift type.


What Python needs is to catch up to SELF, Smalltalk, Common Lisp, and either embrace PyPy, or finally get a mature JIT into CPython.

Thankfully, Facebook and Microsoft are making it happen.


Yes, but it's a completely unrelated matter: even with a JIT you'll want compiled extensions since you can't beat hand crafted SIMD for the scientific stacks or machine learning.


Ever heard of intrinsics?

Maybe spend some time reading about other JIT languages expose SIMD to developers, without requiring folks to go down to C.

Which incidentally is what all those Python DSLs for GPGPU APIs also do, with the caveat of not being that useful for general purpose programming outside machine learning algorithms.


Try to make Rust interop with C++, C and Objective-C as easy as Swift does it.


This is a big deal. In reality, most of the code that powers your systems (phones, tablets, any computers) is built using C and C++, regardless what OS you use.


Agreed, and rust-c interop is as good as any other I've used. The c++ interop was clunky but getting much better with crates like zngur. That is what was holding back its use on chrome for a while. Oddly its python interop is way better with pyo3 and maturin making it super easy.


C++ is (IME) an order of magnitude more annoying from Rust than C or Objective-C. I've dealt with interop of the latter two on a number of projects and frankly found them to be fine.

(Though I wouldn't fault anyone for making an argument that Swift is still more ergonomic here)


swift-wasm exists


Most rubists like the optional parenthesis, but I suspect must devs don't.


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

Search: