Sure, it's easy to install Windows 11 anyway. I've done a few of these as a test. What's not clear yet it how these machines will handle the major annual updates.

Right now, the current version of Windows 11 is 22H2, but we're due any day now for a 23H2 release, and it's not clear if a machine that bypassed the CPU checks will be able to just get that update in the normal way. If it doesn't get that update, 22H2 should continue to get security patches for the next 12ish months (currently shows as expiring October 8, 2024), but then it's suddenly abandonware again unless you do a manual OS re-install.

No, don't just remove the requirement. TPM was a good thing. People don't understand it, but it DOES increase security in a meaningful way, and if the line wasn't drawn it'd be another decade before a lot of people ever saw it. You don't have to understand it to benefit.

It would be nice, though, to have some kind of patch for older devices to allow them to upgrade after clicking through several warnings, and maybe with a certain features withheld, for laptops that can't just add a $20 usb TPM module as a practical work-around. And oh, look: we *can* indeed trick Windows 11 into installing on these machines.

The worse requirement is the arbitrary cut-off for 8th Gen (intel) CPUs. There's a lot of 6th and 7th gen business equipment out there, including stuff still for sale on the secondary market, that does include the TPM and still performs better, with better instruction set support, than some of the low-end stuff released today new. That's a place I'd where I'd like to see a better story.

Microsoft doesn't make (most of) the machines... but sure, let's make this a requirement. While we're at it let's go after Apple, too, which historically is MUCH worse for this. Microsoft is killing off PCs made before August of 2017 in October 2025, so that's 8 years. Apple abandoned 2015 Macs in 2021.

Next up: compare this against schools in a few rural districts. Since the original study was done in the context of a large city (LA), where you'd expect air quality to be poorer, pick a few rural districts with better initial air quality, install similar-looking boxes with fans in all of them, where only about 1/2 boxes have actual filters, and compare this with LA results. While we're at it, also try a few small-to-mid size cities. Say, Lincoln NE, Madison WI, Little Rock AR, where we measure the average outdoor air quality at each location and correlate it with the change in results.

These devices remain popular because they have exactly the functions needed for the material. and not all the other smart device features you don't. It's that exclusion that makes them important. Even when you have a more-enlightened classroom situation, the SAT, ACT, and other standard testing environments need to be able to know you're not looking things up... for example during the science or history portions of the same larger test. So the devices fill a niche, and that's fine. But given the technological advances in that field, there's no good reason a device first released in 1990 still needs to cost the same $100 it cost back then. Given the simple screen and modest CPU requirements, I'd expect them to be able to make a reasonable profit at less than half that price.

This misses the point. From what we know so far, the reason for the dismissal is Stack Exchange insisting singular "they" is not good enough, and the new code says you *must* use someone's preferred gendered pronoun if they have identified it for you. Avoiding it with "they" and proper names is still rude and a CoC violation. Moreover, in this specific case the mod had neither to that point either used singular "they" or intentionally misgendered anyone, but was fired for asking about clarification whether "they" and proper names would still be okay. The irony is the not even Stack Exchange can follow their own policy, as the use of "they" in the press communication fails to properly gender the mod.

They could market it as "Green Edition" or whatever, where expectations are set appropriately, and people would pay a premium for the lower quality.

Gotta love it

... nothing to see here.

> To initiate the exploit, a user simply needs to open an attachment received by email, messenger, or other file transfer service.

When you can convince a user to open a malicious attachment, there are many many options open to you. This is nothing new.

This is after they shortened it.

Windows releases used to be good for **10 YEARS** (mainstream plus extended). IMO, this was one their few big advantages in the server market over linux options, where even LTS linux distros only tend to be good for 3 or 4 without forced updates.

  Now, even in the best case, you only get two years, and that's after they extended Fall (H2) releases up from just 18 months.

"Still in support" is the key line, but they don't have to change any other policies to make this hurt. This was already changed a while back so the Spring (H1) feature updates only have 18 months of support and Fall (H2) feature updates only have 24 months of support.

So the best you can do is update to the latest in Fall, where you won't be bugged again for two years... and at that point, if you skip the intervening updates and go straight to latest, you get two more years. If you don't want the current Fall update, you'll be bugged again much sooner.

They should have named it "Java 10", to signify a greater jump, and to avoid breaking changes with all of code out in the wild that looks for "Java 98" to indicate the old version.

Not exactly. Some people are better **software engineers** than others. That's why they were promoted to Level 4. But the idea is, within a given level, the difference should be close enough that a big company doesn't have to negotiate salary with 10,000 different employees individually. Here's what "Level 4 Software Engineer" makes. If you want more, do better work to get promoted to Level 5 (or whatever is next up the chain).

MS has their "Windows To Go" product now that will even let you run a special Windows install from a flash drive.

This is what VMs are for.

But even putting that aside, why two completely different systems? That's not substantially different than just carrying two devices. Why not just select a different boot drive? If you're really paranoid this could be a hardware switch that completely changes which disk is at the end of a single sata connection, so there is absolutely no possibility of the system installed on disk having online content with the system installed on the other.