Such as high-resolution image editing and high-definition video editing. Compared to a web browser, these aren't quite as amenable to splitting an application into numerous "content processes," each with their own separate 2 GB RAM.

For a while, Linux supported an x86-64 ABI called "x32" that limits each process's address space to 2 GB so that more pointers will fit in the processor's data cache. It didn't become popular, in part because of a need to load three versions of the system libraries: 32-bit i686, x86-64, and "x32". In addition, porting x86-64 applications to use less pointer-heavy containers gave most of the cache advantage that "x32" would have provided. This includes switching from linked lists to gap buffers (or other dynamic arrays), from B-trees to T-trees, or from pointers to indices in a pool. Rust in particular has encouraged use of appropriately sized indices as a workaround for the borrow checker.

There's a widespread misconception that a 32-bit operating system is limited to 3 GB of physical RAM. I think this comes from Microsoft's practice of requiring drivers for 32-bit Windows Server to support PAE as a condition for certification, but not drivers for 32-bit Windows desktop. I seem to remember 32-bit desktop Linux being more PAE-friendly. PAE and content processes are how Firefox for 32-bit Linux managed to hang on this long.

Each function could be loaded the first time the user uses it. The device has to be online to query what is in stock at any given moment anyway. And I'd be interested in others' speculation about why the client side of the most widely used functions can't all fit in (say) 5 MB, which is twice the size of Doom.

I accept that, adding a clarification that I suggested the solution for the purpose of asking other people what these companies might know that I don't.

After scanning that many people, what are the odds of misidentification duplicates? It is a known problem with finger prints.

And/or demographic collapse.

Wikipedia officially requires articles to summarize their sources from a neutral point of view. However, Wikipedia did become overtly political with the PROTECT IP blackout of January 18, 2012, in protest of a proposed law seen to threaten Wikipedia's existence.

The first time you add a website to your home screen, it installs the website's service worker. You have to use the Internet for that, just as you have to use the Internet to download an application from Apple's App Store.

And I'm curious about what the blockers for even a partial PWA implementation have been during each of these 12 years.

All I've been asking is what features of Grab combined with missing features of PWA likely led to their continuing to choose native apps.

I don't see where I "assume[d] to know better than Grab".

I have several chairs in my garage I kept as trophys.

Science? Really? There's a lot of soft-brained, unscientific and technophilic pseudo-religion in the article.

Let's work with the argument's load-bearing phrase, "exploration is an intrinsic part of the human spirit."

There are so many things to criticise in that single statement of bias. Suffice it to say there's a good case to be made that "provincial domesticity and tribalism are prevalent inherited traits in humans", without emotional appeals to a "spirit" not in evidence.

No serious researcher wants a single company (much less NV) to "own the bridge that every quantum computer will need". Seriously, fuck NV and their proprietary bullshit. CUDA is bad enough, now they have to deal with some new-fangled version of NVLink? Hell Naw.

Well, bear in mind that nobody outside the US cares (or even heard about) about Youtube or Disney TV.

Cliques, not clicks.

Even apart from costs associated with proprietary software, the other reason to avoid vendor lock-in is to avoid self-propagating backdoors in the compiler. Ken Thompson described how to make such a backdoor with C in his 1983 "Reflections on Trusting Trust" speech. David A. Wheeler described "diverse double-compiling", a defense against compiler backdoors that relies on the existence of independent implementations of a language. Stable Rust doesn't have that because it's such a moving target, with widely used programs relying on language and library features less than half a year old.

See also "Reflections on Rusting Trust" by Manish Goregaokar

They knew this and decided to maintain Perl 5 alongside Raku (formerly known as Perl 6) for what is looking to be noticeably longer than the 11-year transition from Python 2 to Python 3.

Probably better to edit out genetic defects after birth than trying to eliminate them completely from the genome. Sometimes we have unfavorable traits that played a role in our survival in the past. Getting rid of them entirely may have undesirable future consequences.

So the solution to modern plutocratic corporate feudalists is . . . capitalism!