Yep, most people used their PC's for content consumption, not creation. They're finding that consumption works just find on Android. What little creation they do works find from a browser on Android.

You're absolutely correct that Windows presents the Desktop before the system is actually ready to do anything. It gives the illusion of faster booting.

Personally, I think GUI mistakes are WAY harder to undo, sometimes impossible.
1. For one thing, there's no history of what was done so you can't see the exact mistake.
2. Windows that pop up, steal focus and take the next keystroke as response.
3. Accidentally hitting some sequence that's interpreted as a special command in a GUI. Sometime doing something like rearranging said GUI and having no idea what happened.
4. Many GUI interfaces have no way to export the settings in a human-readable form so you can see what the differences are from default. Makes migrating those setting very difficult. With text-based configuration files and diff, it's trivial.

At the end of the day, if you have no idea what you're doing you want the hand holding of a GUI to direct you into the most likely choices. But if there's no button, you can't do it. With a command line you can do about anything, but you might have to read up on it. The next time it's easy.

I agree with an earlier poster that the ideal GUI would generate a list of commands that it ran so you could see exactly what it was doing. It would also have ^z for all operations.

"A private company has no such luxury. It must be efficient to survive and profit. So its greed forces it to be efficient."

In a nearly ideal situation where consumers have perfect knowledge and competition functions properly.

Utilities are seldom this situation, mostly because it requires duplicate infrastructure. There's very little in the way of utility competition to bring about the good delivered through competition. You generally wouldn't want the capitalist mechanism to play out and correct the problem anyway because that would require allowing a private water systems to fail so that consumers can be educated about the importance of a well ran utility. This is why most utilities are heavily regulated state-sponsored monopolies.
It's easier to flirt with private education because the infrastructure is less of a problem and it's fairly easy to send kids to a different school if one fails. Private schools' largest failing is that they do not address the overall problem from a public-policy point-of-view. It's an opt-out versus fixing the system. It's a fix for the rich, screw the poor solution.
I should note that American public schools vary greatly. I grew up in the West where private elementary education is fairly rare, outside of a few Catholic schools. I hear that in cities where government corruption is notorious, the public schools are pretty bad, but that's hardly surprising.

The feature of Net Equalizer that lets you limit the number of active connections per client works well in limiting P2P traffic. But in other situations, just getting more bandwidth ends up making people happier and costs about the same as trying to limit it, if you include manpower. In an educational situation, Net Equalizer worked well for us. In a business setting, you should be able to mandate that users not do certain things, if management will back you up.

Another way to do this is to have more than one Internet connection and either route some protocols, users or servers over different connections. For example, it can work well to route ports 80 and 443 traffic over one connection and everything else over a second connection.

I used to use Squid for caching Windows Updates and it sped things up about 1000% percent.

I would recommend using something like Ntop to figure out where your bandwidth is actually being consumed and target that for caching.

Much like freeing up space on disks, you can waste time trying to figure out every little thing, or you can target the biggest files and get the most results.

The only down-side of Squid caching is that it can't work with https:

I think a lot of the stolen handsets end up overseas.

The problem is that much of the (perfectly good) hardware will not run Windows 7 or 8, or if the hardware is fast enough, there's no driver support. This might not be a big deal for a desktop box in a first world country, but there's a bunch of medical and industrial equipment that can't feasibly be upgraded.

The solution is to charge for yearly services packs for XP. If MS isn't willing to do it, see if a 3d party will.

When you acquire the sort of market dominance that XP enjoyed, you also acquire a social responsibility to fix the defects in the product or let a 3d party do it.

It might make sense if the "mandatory" part was limited to larger players in a given sector. e.g., over 20% market share or something. Certainly, vendors need more incentives to patch bugs, but I'm not sure this is the right way to go about it.

No, off by default is the right way for security. It reflects the correct way to think about security.

I'm not sure your statement about adoption forwards any logical point. Ease of use and security are generally considered to be a straight-line tradeoff. People don't use OpenBSD because they put other values (ease or use, more default packages, works with x, etc) above security. OpenBSD is a joy to use, until you find something that "just worked" in Linux and doesn't (easily) work in OpenBSD. Security isn't free.

I'm always amazed when people (especially other geeks) don't understand that many gifted computer people have weakness in other areas. Social skills being a prime example. Theo isn't some kind of demi-god, he's a person with one extreme strength and other weaknesses. Torvalds and Stallman aren't exactly the kings of diplomacy either. Being a bit anti-social is more of the norm for genius types.

That monkey needs to incorporate!

The "everything turned on by default" concept is part of why Windows is bloated and insecure.

This is changing, but part of what I like about Linux is that it makes less assumptions about what you are doing and assumes a (at least somewhat) skilled operator. Part of what I dislike about Ubuntu is that it makes too many assumptions about what I want to do. It's also why multiple Linux distros that target different audiences are a good thing.

The points here are that :
1. All optimizations or assumptions make something better for a certain task or use scenario.
2. Any optimization that makes something better for a certain task or scenario will be worse for a different task/use scenario.

Think of strong versus lightweight -- it's a tradeoff.

Seriously, I tried Menuet a few years ago. It's crazy fast on even ancient hardware.

It just needs a couple of killer apps.

Superheterodyne radar detectors would interfere with each other and lessen the sensitivity.

The Internet is a war zone. Not running some sort of script-blocker is like flying through an asteroid belt with your shield down.

Microsoft lulled users into poor security practices with "just works". Java is just too vulnerable to not have some kind of click-to-play or white-list.