No one expects Apple to use Google's extensions related to encryption. Apple didn't support the *vanilla* RCS standard, making it impossible to share full-resolution photos and videos using RCS.

How about you focus on the facts, instead of pretending you know my motivations.

Root should *not* have access to rewrite the entire OS. This is what Microsoft was forced to learn the hard way, through the CrowdStrike fiasco. As a result, they are moving drivers from Ring 0 into a more protected layer, so that the drivers, even with root permissions, cannot cause an OS-level crash.

You are conflating user-level permissions, with role-level capabilities. To be secure, the installer must be a part of the OS, and the developer must only be able to supply a package for that installer to process. This prevents the installer from causing OS problems because the installer simply won't do things that an installer shouldn't be allowed to do.

A competent system administrator would know that they don't know enough to fully understand the details of every security patch. Such knowledge requires much more research and analysis than the typical system administrator has time to do, regardless of their level of competence. It's not their job to second-guess which security updates they should accept, it's their job to keep the systems updated, not to pretend they know more than the software vendor about which things are important.

It appears that three AI companies are neck and neck: GitHub, Claude, and Cursor. https://www.cbinsights.com/res...

My experience is a bit different from yours. I personally use GitHub Copilot, which lets you use models from all the major companies, including Claude. Whenever I've tried Claude models, I get good results, but the execution is *S*L*O*W*. Like, 3-5 times slower than, say, GPT-5.4. So I keep reverting back (for now) to GPT.

At least Gemini is *trying* to tell the truth!

Isn't it???

And these are apparently huge ones, at that!

That sounds like sales gobbledygook to me!

This was literally my point about how security was built into Android and IOS. Neither one lets installers do anything to the OS, the package can only be unpacked in ways that are specifically designed by the installer mechanism. It's not a user permissions thing, it's an installer capability thing. In the phone world, you don't write an executable to do your install, you create a package, which the OS itself unpacks.

By contrast, a Windows or Mac OS or Linux installer, is just an ordinary executable under the hood, that can do anything the user running the code, has permission to do. Because you the developer provide the executable to do the installation, this is inherently a less secure installation mechanism.

And yet, microplastics (or any plastics) are not soluble in water and peroxide. So the body's ability to remove microplastics, as observed in this study (or at least as observed in the lack of buildup), doesn't seem to be related to solubility.

You're right, what you are doing is not what I was talking about.

Your choice is binary for each new version. You can install an update now, or not. You can't choose to install, say, an updated version of Office that includes last year's security fixes, but doesn't include last year's feature updates. If you want to update, you get it all. If you don't want the new features, you also don't get the new security patches.

You missed my point entirely. My point was not to criticize people who don't have $500 for an emergency, nor would I criticize fat people, some of whom are in my family.

Most who point to statis like "two thirds of American's can't afford a $500 emergency expense" use this to promote more government programs, or UBI, or other such taxpayer-funded "mitigations." Injecting more money doesn't help people with this struggle, any more than enlarging grocery store produce sections helps fat people lose weight.

Google's proprietary extensions focus on features like encryption, and have nothing to do with the difficulties Apple users had sending photos and videos to Android users.

I'm simply agreeing with the lead author's conclusion. You are disagreeing with it. It's also not a particularly hard-to-understand conclusion, stated by the lead author.

Further, given that the human body is very, very good at removing contaminants in general, the conclusion passes the sniff test. It stands to reason that the body can also remove microplastic contaminants.

When you say "I do" do you mean you create different patch versions of your software, with each combination of security fixes and feature updates that can be chosen? Or that you create feature flags that the end user can select?

Either way, I'm skeptical. When you reach 30 feature flags, or 30 selectable updates, you reach more than 1 billion permutations. User 1 wants A but not B and C, User 2 wants A and B and C, and so on. There is NO way you can produce that many patches, or adequately test that many feature flags.

Even Microsoft doesn't have the bandwidth to produce and test that many permutations, and they have many, many more than 30 choices that users would have to pick from.

Sure, that's true. But there are also areas of your adult children's lives where you really don't have a lot of influence, because they have made up their own minds. Sometimes, there are opportunities that come when problems arise. But my 32-year-old son isn't currently taking financial tips from me or his mom.

Good point.