They did close airspace during the conflict first up to 26000 feet and then up to 32000 feet. Unfortunately that wasn't enough.

I guess they either didn't know what weapons the rebels had or didn't want to admit it for fear of advancing the rebels PR.

Lets not forget that Nokia was floundering before Elop went in.

AIUI corporate officers are given wide lattitude to do what they belive is in the best interests of the company. Otherwise you'd get a flood of lawsuits whenever a descision turned out badly. So you would basically have to prove that Elop did not belive that going the MS route was in the best interests of the community.

OCR can still be useful for search even if the accuracy is too low for complete conversion.

Windows mobile and windows phone are sufficiently different (AIUI there is some comonaility at the lowest levels but the user and developer interfaces are totally different) that they may as well be regarded as different platforms.

"the world runs windows?" have you ever entered in a real server room? do you know google, facebook, yahoo, rackspace and every other "big player" on internet?

Even if they just ment "on the desktop" Do they mean Windows XP, Windows Vista, Windows 7, Windows 8 ? (Not even considering the various sub versions of these.) There might even be some Win95, Win98, WinME, NT4, Windows 2000 still around. Commercial companies can be very reluctant to spend money "fixing" something which isn't "broken".

Because shit won't break trying to upgrade from Windows Server 2003 to Windows Server 2008?

How about from Windows 2008 to Windows 2008 R2 :)

For maintaining a farm of identical servers, I agree with you completely. For maintaining Grandma's desktop remotely, I agree with you completely. But for maintaining an enterprise desktop environment, Microsoft simply has the best tools for the job. Linux AD-via-Samba quite simply doesn't even come close for the convenience of centralized GP maintenance, and has aothing anywhere near the convenience of drag-and-drop group-based software installation (though Linux does have non-stock application deployment packages available, like Puppet, that partially fill that last point). Linux has nothing even remotely like (W)SUS. And those two alone count as complete showstoppers when it comes to minimizing the number of people required to maintain a large network.

On the other hand Windows dosn't have anything like apt :) or the ability to replace major sections of the system without rebooting.
The whole idea of "deploying" applications is very much a "Windows way of doing things" too. In many cases even Windows applications can run from networked drives.
There's also the Windows profile mechanism, with it's half baked writeback caching, which makes no sense at all in many situations.

My experience is the opposite of yours with installing Windows/Linux. I've found that ghosting Windows installs requires that the hardware be virtually identical. Having a different disk controller, or switching between ATA and AHCI modes usually causes blue screens and failure to boot.

My experience is that it's Windows which is a lot more fussy about hardware and imaging that Linux. This having been the case for at least 15 years. Even to the point of Windows wanting reinstall drivers when a USB device changes which USB port it is plugged into.

Any modern Linux distro, however can quite happily run even by putting the installed hard disk into a completely different machine.

Ironically it's fairly recent additions to Linux, such as UDEV "persistant-rules" and using disk UUIDs which can make this more complex than it was in the past.

I support a university campus and I'm tired (not really, but it gets boring) of being asked for copies of university software by students for whom there is no licenced copy available. The reason? The teacher will be accostumed to using that software and doesn't even consider changing to another.

Both the students and the teachers don't understand how software licencing works. Ironically OSS can easily be used the way they want.

Mind you, I'm not even talking about changing to Linux or some open source program. I'm talking about students (teachers too) persistently asking for Windows XP-compatible software to be installed in their Windows 8 computers when we aren't allowed to do it and asking for us to help them when the magically appearing copy of our licenced software doesn't work with their computers' Windows 7 or 8

Even if a piece OSS written for Windows XP refuses to run/install under Windows 7/8 there are several possible ways to fix this. Whereas with proprietary software you are utterly at the "mercy" of the vendor. They may have gone out of businessor only want to sell a much later version.

Some time ago someone built their business practices around the tools that were available on the platform they chose at the time. If they choose a new platform there should be an expectation of flexibility in the business practices to match the new platform. It doesn't matter which direction you go or what you're changing. It will never be 100% the same as the old solution.

That's true even if you "stick with" Microsoft. With many organisations just recently having undergone a very complex and painful migration from Windows XP (plus Office 2003) to Windows 7/8 (plus Office 2010/2013).
With the irony that a change to Ubuntu (plus LibreOffice) may have been less of a shock to the end users.

From my experience you need less Linux sysadmins to begin with. Its easier to do remote admin.

Also whilst there might be plenty of MSCEs highly skilled Windows sysadmins are hard to find. (They might even be the same people as Linux sysadmins...) There's also a certain irony in "Power Shell" becoming an important Windows admin tool.

So the TCO numbers Microsoft claims are usually bullshit.

TCO numbers are generally political bovine excrement. It dosn't matter if they are applied to computer software, electricity generation or whatever. Usually trick is to ignore some of costs associated with the "right" choice.

I wonder how much influence the absence of Valencian Catalan support in Windows but availability of support for at least large parts of Linux systems influenced the adoption of this system (Windows can apparently be made to display some of its UI in Catalan, but the translation is incomplete, and the local Valencian dialect of the language is entirely unsupported).

You also see something similar with the "English" version of Microsoft Windows. With only US specific spellings being used whatever the location is set to...

The problem is worse on Android than on many other platforms because there are very few native shared libraries exposed to developer and there is no sensible mechanism for updating them all. If there's a vulnerability in a library that a load of developers use, then you need 100% of those developers to update the library and ship new versions of their apps to be secure. For most other systems, core libraries are part of a system update and so can be fixed centrally.

It used to be very common with MS Windows for libraries to be bundled with applications. A situation called "DLL hell". Which can be even worst when an application installer tries to update a "system" copy of the library.

Although you certainly have a point, the core problem is often that the documentation is poor.

A not uncommon problem being "solutions" which omit steps or assume that everyone knows how to find, what is in practice, an obscure option. Sometimes also having "boilerplate" which overexplains another part of the process.

Amazingly, security libraries are often in this category. Is there a really good writeup ANYWHERE about SSL, certificates and signing practices?

That would also have to include TLS, STARTTLS, how it can really be STARTSSL, etc. There's also the issue of what is actually part of the protocol and what is implimentation specific.

Case in point, there's a scary big number of posts from people telling developers how to turn off SSL chain validation so that they can use self-signed certs, and a scary small number of posts reminding developers that they'd better not even think about shipping it without removing that code, and bordering on zero posts explaining how to replace the SSL chain validation with a proper check so that their app will actually be moderately secure with that self-signed cert even if it does ship. The result is that those ten thousand developers end up (statistically) finding the wrong way far more often than the right way.

There are also cases where using a self signed certificate is rather more secure than using a CA to. The whole CA idea having all sorts of problems.
Though the example is one of those where only someone who didn't understand how things worked would need to ask such a question in the first place.