No, post-release runtime factors in heavily.

On the contrary, development time goes up with the number of developers. The scope and complexity of the project can increase, which may or may not be a good thing. In the Unix-like world, it often isn't. Backwards compatibility and dependency avoidance are more important.

This is not insightful. This is typical denial. Hey mods, wtf?

You're not technically forced to upgrade, but I think many people would consider ceasing to provide effective security updates as the end of an operating system's general useful lifecycle if it's running on a networked computer.

Most Linux distributions don't provide long-term support, even for security fixes, for very long. As one of the better examples, long-term support for Debian Squeeze -- a stable release of a major Linux distro widely deployed on servers -- is scheduled to end in February 2016. Debian Squeeze was released in February 2011, making that a respectable 5 year window (certainly better than a lot of other platforms).

However, that is far short of the extended support period of more than a decade that Microsoft has committed to for Windows 7 (around 9 years if you consider that they do require SP1, which was released about 16 months after the original, but which didn't add the kind of monkey business we've been seeing with various "upgrades" to software in recent years).

You don't have to upgrade OS X either, but last year when Apple declined to issue a security patch for OS X 10.6 (Snow Leopard) less than five years after it launched and when it was still reportedly in use by around 20% of people on OS X, they also put an upper bound on their viable support lifetime under 5 years.

When you get to software on this scale just writing the specs is an insanely difficult problem. Most big systems fail due to poor requirements, and I think that is because good requirements are a LOT harder to get right than people appreciate, especially for waterfall-style projects that are always the favorites of big RFP projects like these.

Ok, so how do I set up my internal DNS server so that everything works fine when my ISP gives me a new prefix every 24 hours, or every time the router is rebooted, or every time they feel like changing my prefix?

The advantage of private addresses is that they're handed out by a DHCP server that you can control, not by your ISP.

Again, I think IPv6 users today are spoiled by tunnel brokers and the like who are giving out static prefixes. I don't think that the likes of Comcast are going to be going that route when they finally embrace IPv6 fully. If nothing else a static prefix is something they can sell, and you could also argue that not having static prefixes makes life easier for them if they want to change their internal network topology.

How exactly do you set up internal DNS when all your IPs are subject to change?

You don't have to change ISPs to have your IPv6 prefix change. Your ISP need only assign you a new one. It is in their interest to do so often so that you have an incentive to pay more for a non-residential connection, and also because it gives them more flexibility in renumbering.

As far as I can tell, any IT project costing a billion or more is 100% guaranteed to fail.

Also, it sounds like they decided to source IT from Lufier, Mephistopheles and Satan, which incidentally also guarantees it to fail.

The BDSM convention will have a higher percentage of protected sex, and nearly everyone getting screwed will be doing so consensually. Most of them will have a good sense of humor about what they are doing, heavy D/S tops aside.

I think that's a great idea, but how many people do you think they're going to invite to speak who hail from the other camp? The "What the fuck are you doing to everything we love" camp? I'm betting it's going to be a big, fat goose egg. And even if not, we've seen how Lennart responds to critics, it's a big part of why so many of us dislike him so.

Even CNN knows, google could have told you, WTF is wrong with you?

You are delusional if you think I said or implied that.

Actually, a new ice age is starting. We just managed to more than offset it, though...

No, I'm saying and I have said before and I will say again that cgroups are simple, did not require a new init system, and can be manipulated from shell scripts.

That's a problem, but not an insurmountable one, and again, one which could have been solved with a unit script processor (which itself could be a shell script) rather than a whole new init system — I've pointed that out repeatedly, as well.

How much does it cost to get your slashvertisements accepted verbatim?

China's 33-petaflops Tianhe-2, currently the fastest supercomputer in the world, while still using Intel Xeon processors, takes use of the home-grown interconnect, arguably the most important component of modern supercomputers.

But where does it take it? To the movies? I hear you can get in for just fifty cents.