I'd say it is warranted to credit MS with making serious effort to maintain backward compatibility. A win32 binary (not driver) which if implemented to spec on Windows XP w/ VC++6.0 probably still runs in Windows 10, at least until they kill off win32 in favor of UWP. Win32 and COM's fanatical (and much appreciated) adherence to ABI versioning means most software packages will run for a very long time via MS's compatibility shims. How many commercial or FOSS projects can you think of which provide those kinds of version compatibility shims? The only one I can think of is glibc and very few projects appear to target versioned symbols. Microsoft's backwards-compat-OCD exceeds what is possible with static linking.

15.04, and the battle was limited to the online reviews. The actor change went largely unnoticed by most viewers.

I would bet that 2.6.x still has a significant install base under the names RHEL5 & 6, centos 5 & 6, oracle linux 5 & 6... And aside from that middle brand, these are people who are probably paying for vendor support, which makes them more equal than other linux users.

Hire actual QA. Showstopping bugs prevent me from getting shit done. Looking at you, Windows Insider program.

That takes me back. When did we get automatic PCI IRQ steering, Windows 95 (20 something years ago)? I feel like this stopped being a problem when ISA went away.

It breaks fewer people's shit at once if there's a bug they didn't catch. It's like beta testing a new feature with a small group before deploying it to everyone. It's prudent.

We're just lucky it's not a ring -3 antivirus coprocessor embedded in the northbridge.

It'd be nice if this proper OS existed at the present time because I'm not going back to Amiga despite its lack of viruses. I'm reading between the lines here, but I infer you are setting up Linux or a BSD variant as this hypothetical proper OS, all of which have their fair share of vulnerabilities and are even harder for non-technical end users to configure correctly to avoid problems.

Sometimes you don't want to put a 450 kg warhead through somebody's window when an 8 kg warhead will do.
Sometimes you don't know how many targets there are until you're near the target.
Sometimes you need to use additional missiles if the first wasn't sufficient and can't afford the non-trivial flight time for a second launch.
Sometimes you want to go home without blowing things up and without wasting 1.6M USD.

There are advantages to having a reusable launch platform in the area, whether that be a UAV or a strike fighter.

Nevermind, it's not clear from the summary, but all of the articles mention this. Yes there is still licensing, no the rules are not as strenuous as a full pilot's license (no medical, etc).
FPV flight is still dead without a waiver. Interestingly, you can fly above 400' as long as you are within 400' of a structure (eg, for remote visual inspection of tall buildings).

http://www.faa.gov/uas/media/P...
Is that not the point of the remote pilot airman certificate?

Think of it more as Amazon trying to encourage the development of automated photo morphing technology. In a decade, we may have some awesome algorithms to obviate those photo editor people... what's the word... Photographers.

I'm glad you're able to repair defects that only occur on hardware you don't own, that you can't reproduce, and can't validate as fixed for hardware/firmware with no vendor support, having no published specification, and will be obsoleted in only a few years. There are hundreds of different platforms to support when nobody follows the spec. Most people consider that somewhere between "frustratingly difficult" and "damn near impossible."

Are we talking about the same Android? You seem to have enumerated where android is the weakest; poor vendor update frequency, many varied hardware platforms, a plethora of vendor or user customized software configurations.

Argument failure on my part. After reading subsequent posts you made clarifying your position, no, I don't see any advantage for the common consumer to go out and replace all their old things with new ones. Someone like me might like that I can use my phone to one-click reconfigure my tv and receiver to play video games or select a movie on netflix and have the tv switch inputs to whatever and just start playing it--heck I can do this now, but IoT should make it much easier. Granted there are a crapload of privacy concerns exactly like you and other commenters have cited that are of serious concern. That stated, given the low incremental cost of enabling IoT on a device, it's pretty damn likely to end up in all products whether you like it or not.