Doesn't matter whether it is good. Sometimes stopping something bad is worse - an example is alcohol and the worse effects of prohibition.

Alcohol gives nothing indispensible, and if alcoholic drinks were not invented, it arguably wouldn't have been an impediment to society. But once alcohol is invented, prohibition typically has an even worse effect.

Suppose alcoholic drinks were being invented today, one might ask whether it is a good invention, and arguably get a truthful answer in "no". But that answer doesn't matter. Similarly, information might not want to be free or want anything at all being inanimate, but technology is at such a stage where preventing copying entails unacceptable surveilance.

I don't understand how people use Android devices without root. Not even a firewall !!
Check the headphone jack before rooting. And don't break after it within a year - after which you'll have no warranty anyway for most devices.

That is ok because there is not much of a need to look beyond Novell. There is also something called open (relatively) specification, and a bigger something called target platform. GNU/Linux is great for software where GNU/Linux is the target platform. Wine is great for software where wine is the target platform - which isn't for much of software.

So this is a very bad example. Free reimplementation doesn't matter in looking beyond, but other aspects of Microsoft's and others' behaviour around the ecosystem matters. Brings me back ot my original point, this is similar to how Android sucks, but Cyanogenmod / AOSP / Xposed rock.

Internet.

The great thing about Android is not Android itself - but the hacker community around it. The Xposed framework has utter disregard for the Android CDD - and supports 2 applications side by side. You might have to make sure it works for the specific device. Or rather, you might have to find a way to make it work for the specific device.

This is a bigger problem. Using software in wine is not really looking beyond Microsoft.

One wouldn't. Research, buy, take responsibility. The idiot mentality US has encouraged where sue-happy buyer has zero responsibility for even understanding the idea of the device he is buying is the reason for companies to not offer interesting things in the US - epecially in electronics.

Highly uninformative post as it doesn't mention this situation is only for NT. On UNIXes, port 1521 (or whatever port is selected in installer) is enough.

Oracle's emphasis is rarely on Microsoft's operating systems. Not only in RDBMS, but many other products support UNIXes primarily, and Microsoft's operating systems secondarily.

Right. So you need to look beyond Intel and Microsoft. Lenovo IdeaPad A10 - 10.1 inch, quad core cortex A9, Android, tiny charger, weighs a kilogram with charger. If you don't find it in the US, probably buy from China because the US is bought and paid for by Intel and Microsoft.

1. English is an accidental enabler of hardware keyboards. English is one of the languages which have the least number of "characters". And it has become very popular. But popularity and brevity of characterset are not related much.
Software keyboards are giving back the rest of the world their more expressive charactersets. And they are loving it - see higher popularity of Android in areas with non-English speakers.

2. One hardware keyboard design has to be created for every language. At least every characterset. The world is a lot more varied in languages and characterset than Americans realize. I am currently in a country where a currency note has the amount printed in 16 languages. Most languages here have 75 or more unique 'characters', some have hundreds, making hardware keyboards impractical.

3. While US is a country predominantly using English language, and many other languages with Roman character set, US doesn't like Android as much as the rest of the world. Especially Asia. US likes its iPhones. And guess what? iPhones have very popular slideout (or fixed) keyboard cases.
For Asia, the home of Android, with dozens of languages with millions of speakers each, hardware keyboards just don't make sense.

You say "completely wrong", but not provide any argument against any of my statements.

Failover is used, but cluster solved it already, 30 years ago. You make no argument against any statement of mine in this context.

Overhead of multiple operating systems is well known. You make no argument against any statement of mine in this context either.

Power saving is something I added extra, but I don't see any argument from you about it either.

Are you the author of RFC 3514 ?

Overhead the GP is talking about is probably the operating systems that those 30 virtual machines are running, which is a huge duplication of effort.

Big enough infrastructure probably means that (say) the virtual machines are running web servers - you could run just one big web server on an operating system instead of 30 operating systems AND 30 web servers.

And failover advantages of virtualization don't apply to this because even such big web servers should be run in a cluster - as the assumption of "big enough" infrastructure probably enables.

And power saving by consolidating services and switching off servers in low load periods is not really an argument because during low load periods some of the blades running some web servers could be shut down.

With alternating current, the electrons are not leaving the houses of the people. Except by diffusion of electrons, which would happen even when zero electric power is consumed without airgap.

Here, it is not primarily about what is beneficial to oneself. It is like doing something benefits oneself greatly, but harms everyone, including oneself, a little bit.

There the benchmark you put out does not hold.

This is the case of file paths changing across operating systems over different releases of operating systems. They can change due to some property in the operating system changing value. Or the place to keep that property changing. Or the property name itself changing. Or the process of determining the file path changing in some manner.

The reason for this change is that some one in operating system design team decided to change it, or the packagers of the software being used decided to change locations of files. The minds of these people are NOT deterministic.

You can analyze the whole operating system in your program, but you wouldn't know the location of the exact file you are looking for. This is because name of the file could have changed, or there could be many files of the same name. So even by analyzing the operating system finding file path in unknown environment is NOT deterministic.

Not an option in the mathematically challenged US of A where explaining "every other", even numbers, odd numbers would take all the allotted time.

There you were the problem for not buying Nexus. Or any other phone after reading reviews and some space to install alternative keyboard / calendar / messaging applications. Free.

And if you claim ignorance, you cannot claim benefit of free market anymore because free market needs informed participants for functioning properly.

No. Niche markets are always full of stupid shit. Slide-out keyboard phones are a niche market. No amount of Stupid Shit Index is going to alter that fundamental reality.

"Pretty" water bottles, with a lot of attention paid to their looks, leak. Braille books have typos.

For non-niche markets, Moto G and Moto E are available, much cheaper than your phone and with zero stupid shit.