Hacker Newsnew | past | comments | ask | show | jobs | submit | 201984's commentslogin

What, like electronic payments using phones running American operating systems? With the bonus that you have two new gatekeepers that can lock your citizens out (Apple and Google).

Harmony OS looks great. I cant wait for the EU version. We will again have a hilarious discussion about US apps. Peeping tom and his lawless army of handsy friskers.

So you'd switch USA for China, and still have a foreign dependency embedded into the very fabric of your society. Do you not see a problem with this?

Oh, i mean a EU Phone OS. Eventually it has to happen. It will of course be entirely ceremonial and run on chinese made hardware with US apps.

Exactly, do they want people to buy their FPGAs or not? Charging per-seat licenses for developers and heavily restricting the free version mean people will buy from other vendors or just not use an FPGA at all.

I'll bite. Why is it the fault of the organization that gets broken into, rather than the fault of the attackers breaking into it? Even if the defender takes every reasonable defensive measure, they could still get pwned from some zero day that they had no defense against. Should they be fined into oblivion for something like that?

The question is whether the defender takes reasonable defensive measures or not.

The problem is that without having some kind of enforcement, businesses will decide that it is cheaper to not worry at all about security and thus their customers will have their data leaked/shared etc.

There's a world of difference between a company that puts effort into security and one that doesn't.


What is my incentive, as a shareholder in a medical company, to demand functional, bulletproof security, and to hold on to no more data than I need, and to encrypt everything? I'm never going to suffer as a result of breaches. Nor are any of my staff. so long as evidence doesn't show that they did it deliberately.

A cryptocurrency business or a diamond business, by contrast, has very strict security protocols, because if they don't, all the value gets wiped out very quickly. The rules basically absolve the healthcare company of fiscal responsibility.

This update OP is posting about may require jumping through certain hoops, but it does not require functionality of those measures.


Is this why every healthcare website has 2FA now? It's so annoying.

Medical records are the most valuable records to steal. They contain financial info like SSN for card fraud, but also demographic info that makes it crazy valuable.

Also, credit can be frozen and credit cards can be canceled. But the medical record, once stolen, can’t be frozen or canceled, so it’s always valuable

>an unaligned pointer in itself is UB, not only an access to it.

Can someone point to where the standard states this?


I think this is 6.3.2.3.7 in C99 about casting between pointer types:

> If the resulting pointer is not correctly aligned for the pointed-to type, the behavior is undefined.

However, unless I’m missing something, producing such a pointer from an integer is apparently not insta-UB? 6.3.2.2.5:

> An integer may be converted to any pointer type. Except as previously specified, the result is implementation-defined, might not be correctly aligned, might not point to an entity of the referenced type, and might be a trap representation

And later on 6.5.3.2.4:

> If an invalid value has been assigned to the pointer, the behavior of the unary * operator is undefined.

Which implies that the invalid pointer must have been obtained without being already undefined, right?



Why wouldn't 8 be enough? Surely 18,446,744,070,000,001,024 characters is enough for every writing system in the world.


Because that's not how Unicode works. It's not simply a table mapping numbers to all possible symbols. Unicode is full of special codepoints that have no meaning on their own, they serve as modifiers to other symbols and a single visible symbol can be formed by an arbitrary (in theory) long combimation of such codepoints. It doesn't matter how you encode it, it simply doesn't work as "codepoint -> symbol" and indexing in a unicode string is never O(1) and cannot be made O(1). Could we use a simple table approach? Maybe. But it wouldn't be Unicode


I actually wonder if the combinatoral explosion of attempting to enumerate every possible character combination would exceed 2^64 bits. My intuition is that it might, and also such a system would be unworkably unwieldy. The size of the spec document would also suffer from the combinatoral explosion. Imagine a system that tries to encode a unique entry for every possible Zalgo character.

Also, literally nobody wants to use 64 bit values to encode ASCII values. Even in our world of insanely large storage that would be breathtakingly wasteful.


Agreed, but it will take many generations for people to see characters in textual strings mainly as “code” instead of “data”.

>There are studies out there, that sometimes it is even benefitial to be slow and show a loading indicator because it could increase trust from users,

And I as a user absolutely hate programs that do this. Put an "updated" message with a timestamp if you want, but don't pointlessly waste my time.


Yes, there is probably no UX/UI approach that makes 100% of the users happy.


I'm 23 and IMO, the Windows desktop style peaked somewhere in Windows 95-2000. The first Windows I ever used was XP, so I'm mostly making that decision based off screenshots and emulators.

UIs back then were dense, didn't waste large amounts of space in a misguided attempt to be "minimalist", and had affordances for ease of use. There was no scrollbar hiding, no animations that made the user wait for no reason other than the designer's ego, very visible borders on windows and buttons that made finding/resizing them easier, large bars at the top of windows that let you move them around, and actual text for most buttons instead of icons that are anyone's guess what they mean. Thankfully some of this can be dialed back in the Windows 11 accessibility settings, at least for missing scrollbars and getting rid of time wasting animations, but a lot of programs don't respect those.

That's right there is a good indicator for which programs care about their users. I'm using your program because I want to actually do something, not waste time watching your designers show off.

I've disabled animations on my Android phone too, and it gives an extremely noticable speedup. Menus appear right when I click them, instead of a second later as they slide into existence. Too bad iPhones just replace the slide with a fade of equal duration; disrespect for the user's time like that is yet another reason I will never buy one.

Those older GUIs didn't try to hide the filesystem hierarchy either. It infuriates me to no end when I use a new OS and have to hunt down the way to show the disk root, or filename extensions, or hidden files. MacOS was especially bad; I had to look up a freaking keyboard shortcut that I never would have found on my own. The common reason is so "normal people" can use the interfaces, but I think that's infantilizing and is why tons of Gen Z don't know what files or folders are. Most people can learn .docx means a Word document, and C:\Users\TheirName is where their files are.

(Notable shoutout, the GNOME open/save dialogs are the absolute worst. I wish distros wouldn't default to it. People will just go right back to Windows 11 because it's somehow better.)

There's some improvements possible, for sure. I'd like to see some programs put hint letters over buttons when you press a modifier like Ctrl so you can easily see what the shortcuts are. I don't know of any that do, but it'd be very useful for more complex software like drawing programs or word processors.

edit: typo


> Notable shoutout, the GNOME open/save dialogs are the absolute worst. I wish distros wouldn't default to it.

I'll never understand who in their right mind would think that in a save dialog, if I start typing, I mean to search for something instead of trying to change the name of the file. It's really baffling.


The "constant migration to a new best thing" is a big problem. Once written, a program should be able to run forever, but this is only true on Windows for GUIs and on Linux only for some CLIs. Arch just recently dropped the original vi from its repos because "it no longer compiled" with stricter GCC settings, and if you want to run an older GUI, just forget about it. It's hars to blame people for only targeting the Web or Windows when those two will work forever, but on Linux you have to keep up with the endless treadmill of X11 to Wayland, GTK 2 to 3 to 4, Qt 3 to 4 to 5 to 6, pulseaudio to pipewire, etc., and if you miss just one you may as well give up.


The entire existence of docker is basically because Linux is impossibly bad at maintaining old software while Windows will still run some terribly old things (though the loss of Win16 is a big one).


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

Search: