torstenvl

Literally never heard that prior to hacking on the mg codebase.

I've used mg and find it generally useful. It's my default editor pretty much everywhere.

However, the code is a mess, with deliberately cryptic identifiers (e.g., `w_doto` for `cursorOffset`), redundant functions (e.g., `vtputc()` is a strict subset of the behavior of `vtpute()`) and a bunch of other weird choices (e.g., deeply-baked assumptions that one byte == one char == one column, even thought Big5 and ShiftJIS were already a thing when it was written).

I tried adding UTF-8 support to mg a couple years ago and basically gave up.

I'll probably be an atto convert.

Again, that is inarguably and obviously wrong. I cannot understand why you would say things you know to be false.

You seem to be conflating the act of writing to an arbitrary address with the act of assigning that arbitrary address to a pointer to begin with. They are distinct actions.

> it would be better if programmers read deliberate instances of UB as if they literally said "std::unreachable()"

Except that isn't true.

There are basically two kinds of undefined behavior.

The first is undefinable behavior, the kind that easily goes disastrously wrong, like writing to random memory locations.

The second is incidentally undefined behavior, something where there just isn't enough implementation consensus for the standard to settle on it, like realloc(p, 0).

There's no standard terminology I'm aware of to distinguish the two. But the distinction exists.

And where programmers read deliberate instances of UB, it's usually the latter, e.g., writing directly to an address memory mapped to device I/O in that particular environment.

That's the point.

In Coders at Work, Peter Seibel quotes Jamie Zawinski as saying that certain Netscape acquirees were people for whom "cross-platform means both Windows 3.1 and Windows 95."

I find that there are domains where LLMs are much faster and more skilled than I am, particularly in extremely well-documented but technical and complicated, but a lot of domains where they cannot do anything at all (mostly novel issues, weird architecture issue resolution, etc.).

Building a basic X11 window manager is almost a one shot prompt.

Modifying a UI toolkit to make it work with MSAA/IA2 is simply not possible.

There's a lot of room for deep work left... for now.

The data in that article support the opposite conclusion.

The reasoning the article takes to reach its incorrect conclusion is plainly erroneous. For example, it reasons that "Northern latitudes have later onset of winter sunrise and higher prevalence of SAD, with the latter increasing by 0.2% per 1° increase in latitude," ignoring that northern latitudes also have less sunlight overall.

My post did not ignore the position of the sun. Indeed, it's governed by it.

I am not interested in whatever you've got going on right now. Being dishonest about others' positions is bizarre. Refusing to clarify your own is bizarre. Let me know when you're willing to engage in good faith.

I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question.

What possible meaning could the phrase "noon happens at noon" have? That's tautological and meaningless.

And if your solution requires a wholly independent, newly built mass coordination mechanism rather than just a minor tweak to the existing one, then I think the burden you bear for arguing your position is extremely high.

All of the science supports permanent daylight time.

Even this article, while stating the opposite conclusion, directly scientifically supports permanent daylight time as beneficial for most people.

It points out sunrise being at 0751 versus 0851 for Toronto. That's the anticipated sunrise time on 1 January 2027. The anticipated sunset time is 1651.

That means that someone working in a cubicle for a standard 0800-1700 schedule will get 9 minutes of possible daylight under winter time, versus 51 minutes under daylight time (sunrise 0851, sunset 1751).

Lack of sunlight leads to terrible circadian rhythms. There is absolutely zero scientific support for the proposition that nine minutes of daylight is healthier for our circadian rhythms than fifty-one minutes of daylight. No person could possibly argue such a thing in good faith.

(Also, I have a terminology quibble with "Standard Time," since Daylight Time is the actual standard, and we observe it eight months of the year.)

MetaFilter is by far the most anti-social group of moderators on any site in history.

I am immensely skeptical that any sizable company would be willing to do that.

> shoe horn multiplatform UI into MFC

This is nonsensical. What are you even trying to say here? MFC isn't available on macOS or GTK.

> Just look at Unicode. It’s an unholy mess.

This seems utterly unrelated to the issue under discussion.

> Qt does what it does, even if the approach is not the same (arguably superior), there’s also Tk. There’s also libui and IUP.

None of those allow a single codebase to compile on Windows, Mac, and Linux, and get native UI elements on each.

Flatly incorrect on all possible counts.

Pro tip: if your substantive argument requires gaslighting people about the literal definitions of words, maybe you should take a step back.

> Lockout is a real risk, but there is nothing insecure about passkeys.

secure - adjective

se· cure si-ˈkyu̇r -ˈkyər

securer; securest

1

a: free from danger

b: affording safety (a secure hideaway)

c: TRUSTWORTHY, DEPENDABLE (a secure foundation)

d: free from risk of loss

Passkeys as a sole/required method of authentication increase the risk of permanent loss. In any other configuration they do not mitigate phishing risk.

It is not possible to say there is nothing insecure about them.

> To an individual, the greatest risks are instead permanent account lockout, automated account bans, and device loss. By using passkeys, you gain better security against man-in-the-middle attacks but face the higher probability scenario of losing access to your accounts.

I'm glad to see this view becoming more mainstream. Passkeys are grotesquely insecure.

The only possible way to consider them more secure is if phishing attacks were more common and more damaging than lockout, which is so implausible that I reject the idea that someone could take that position in good faith.

Yes. Agreed. Did you intend to respond to someone else?

Those comments may be fair. Defamation is not.

It appears you're calling something a scam because an early alpha version was early alpha quality. If so, that would be unprofessional and inflammatory.

I'm open to being wrong. So again, can you explain what you mean?

Can you explain what you mean by this? They haven't had a first release yet. They're still in beta.

It was never true.

> the vast majority of buyers are willingly choosing large phones

This is patently false. It is widely known to be patently false. I cannot understand why you would confidently assert something commonly known to be patently false.

https://www.cnet.com/tech/mobile/apples-iphone-16-still-lead...

Every single generation, without fail, larger phones are outsold by smaller phones of comparable performance and features.

Literally all available evidence directly contradicts you. There is not a single iPhone generation in which larger iPhones outsold smaller iPhones of comparable capabilities.

As was clearly, painstaking explained to you already, using easy-to-understand language, the 4.7" iPhones outsold the 5.5" iPhones. And today 6.1" iPhones outsell the 6.7" iPhones. Every single time, without fail, smaller iPhones have outsold and continue to outsell larger iPhones of similar capability. This is an objective and indisputable fact.

If you're not going to engage in good faith then don't engage. My post is right there for anyone to read.

Flatly incorrect.

In 2020, the iPhone SE sold over 24 million units.

In 2022, sales of iPhone mini 13, iPhone mini 12, and iPhone SE made up 12% of total iPhone sales.

Thats despite intentionally nerfing every one of those models.

If you look at comparable phones, the iPhone 7 and 8 (4.7") outsold the iPhone 7 Plus and 8 Plus (5.5") by significant margins.

Nobody wants large phones.

In what way is the fundamental idea of wxWidgets not sound? It works very well, and is pretty much the only toolkit to do what it does.