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

An “aural” deaf school? This seems like a fairly harmful approach. I know that approaches to deaf education are quite fraught, but pushing students to communicate orally and not allowing sign language in the classroom seems like it’ll set a lot of students back educationally. It essentially turns deafness into a learning disability, which it doesn’t need to be if you just allow sign language. (It also shuts the students out of mainstream Deaf culture, which I imagine a lot of them will resent later in life.) I am surprised that a school with this philosophy still exists, frankly.

They’ve been doing this for over a century, it’s probably the top deaf school in the UK, and has the support of nearly the entire deaf community.

Most of the students have either some degree of hearing or use cochlear implants. I think nearly all, if not all, students use either hearing aids or cochlear implants.

The classes are very small (eg 5-6 max usually), students are arranged in a U-shape around the teacher so they can read lips. And there’s a special wireless broadcast system so the teacher wears a microphone and sends the audio directly to hearing aids or cochlear implants.

Regarding deaf culture, most of the students use BSL on their own outside class, and my daughter learned BSL from her friends there that grew up with it. Coming from a mainstream primary, she found “her people” here, discovered deaf culture and a community that shares the same struggles she faces.

The idea is that by teaching in BSL the students are further restricted in their ability to function in a hearing society.

I’m curious if you are deaf yourself, or work with the deaf. All the teachers at the school are trained teachers of the deaf, some are even deaf themselves. And I haven’t heard any complaints about the aural nature of the learning (except from the reservations of a few parents before sending their kids there, and I don’t think any of these parents regrets this after their children started there.)



Documentation to guide devs on safe usage of C++ is enough?

So any language should be allowed as long as they instruct developers to be careful.


Compliance often works exactly like this.

I don't know if they do this, but those conventions could be enforced by a tool.

Theres C++ in military airplanes, they just cut out 90% of the features: https://www.stroustrup.com/JSF-AV-rules.pdf

And heres a nice video about it: https://youtu.be/Gv4sDL9Ljww?si=Z4riPMKAKcIKaU0s


Yes, in WebKit, SaferCPP guidelines are enforced by a static analysis tool.

My work bans raw new and delete, so we only use unique_ptr. It's not as memory safe as Rust's borrow checker but I've never seen a segfault.

Yes, they do this, and it's really not an unreasonable requirement.

Of course. It's just a coincidence that they're placing onerous restrictions on competi- I mean alternative browser engines. Restrictions which, of course, they're not obliged to follow themselves.

I am sure that Apple will make no other efforts to impede others from unwalling the garden. That would be completely ridiculous, and frankly, un-Apple-esque.


Both Chrome and Firefox are already compliant, so I don't see it as onerous, but the full context of the list is indeed an extremely loud and clear "FUCK YOU, WE OWN YOU" to regulators and other browser vendors.

Which of the restrictions do you feel they don’t abide by? It looks like they meet all their own restrictions

> Use memory-safe programming languages, or features that improve memory safety within other languages, within the alternative web browser engine at a minimum for all code that processes web content;

There is absolutely zero way to satisfy the latter part here. It's at best non-enforceable. If I'm using C++ and use std::span instead of a c-style array, is that good enough?


Why not? The wording is “features that improve memory safety”

It doesn’t say that it needs to provide absolute memory safety. Based on the linked WebKit guidelines, it seems like they meet the criteria.


That's the commenter, not from the Apple page as far as I can tell.

My point is the requirement is too broad. It cannot be meaningfully enforced.


It’s literally from their requirements page

https://developer.apple.com/support/alternative-browser-engi...

You have to request explicit permission to be able to be a browser on iOS. You can’t just ship an app. I assume part of that process is that you specifically demonstrate that you try your best to use best safety practices.

Again, it’s also not absolute safety. It’s just due diligence review.


Sorry if I wasn't clear. I meant the WebKit guidelines were from the commenter, not from the apple page.

> or features that improve memory safety within other languages, within the alternative web browser engine at a minimum for all code that processes web content;

This can't be analyzed in any real way, so its just another way that Apple will restrict web engines and claim it was due to "not enough use of memory safety language features"


Why does it matter if Apple themselves don’t link the WebKit docs? It’s literally their project and seems to meet their requirements.

There’s a lot of things in the requirements like funding that Apple cannot verify. I think you’re being too binary in this.

Some of it is very clearly intended to be a “show us you are at least considering these security measures and have practices in place to minimize known issues”. Again, for the third time, it’s clearly NOT a list for ongoing perfect security, given that there are other items on the list that deal with further mitigation strategies.


> It’s literally their project and seems to meet their requirements.

This is meaningless. Apple can carve out special exceptions for themselves all day long.


What is the exception? I’m saying they meet the same requirements they are asking for other browsers.

This is literally the question I started this thread with and you have gone in to a loop of saying “they can’t enforce this” without any response of substance.


Your "substance" is "trust Apple will enforce something correctly where there isn't a correct answer". I don't agree with that. Apple has a history of interpreting things favorably for themselves and locking 3rd parties from doing the same things for wave hands reasons.

If you are going to make guidelines, make them evaluable. These aren't. If you care about memory safety, either say use a memory safe language or point to an exact reference guide to use to allow XYZ language to satisfy it.


Then you’re basically strawmanning here because you’re applying your own interpretation to the rules as written.

If you would pause for a second and actually read the rules in their entirety it is obvious that the lines you’re fixating on aren’t meant to be absolute security measures and therefore don’t need to be continuously evaluated.

Your conjecture about Apple withholding the permissions for arbitrary reasons is not borne out of evidence. When asked REPEATEDLY to show where they’re giving themselves an exception to their own rules, you continuously fail to provide any example and are just hand waiving conjectures.

Maybe they are doing what you’re saying but you’re making an incredibly poor argument regarding it.


Apple's resistance to rust is truly mind-boggling

Based on the prose style, I'm assuming you copy-pasted a ChatGPT "deep research" answer?

The prose style and the fact that it was super repetitive. Every bullet re-described the copy-pasting. Definitely LLM slop.

We don’t need a “deterrent” against things being redacted in publicly released documents. We can have transparency without the whole world finding out the names of victims and witnesses, people’s phone numbers and SSNs, etc., every time a document is released.

This is wrong - Apple did implement multipoint pairing. I’ve had Bose headphones and cheapo Amazon headphones and both switched perfectly between various Apple devices.

No, they did not implement mulitpoint Bluetooth as per the spec in AirPods, they are doing proprietary handoffs between their own devices, multipoint pairing is not implemented for Airpods outside of macOS and iOS.

It doesn’t need “the cloud” (switching works offline) but it does need to verify that the device it’s switching to belongs to you, which it does using a keypair associated with your account.

Didn't I previously prove this by completing the pairing process anyway?



To be clear, there’s no actual evidence that he’s being evicted. Talking about “his eviction” is pretty premature. It also seems like he will receive the reward.


Their handwriting benchmark is not useful. The test cases aren’t even handwritten!

https://www.codesota.com/ocr/best-for-handwriting


That’s just the illustration. But this is misleading - I will fix it asap and show real examples. I’ve run the mistral ocr on other benchmark


Do you know of any good handwriting eval/benchmark? I haven’t been able to find one.


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

Search: