I disagree whole-heartedly. I've been using Boost for a number of years, and although it's been a bit hard to get into at times, and although I've barely scratched the surface of the available libraries, it has made my life a lot easier. Many of the available libraries complement the standard C++ library very nicely. No matter what parent post says, it's also fairly well accepted, to the point that it is not a long shot to expect a random competent C++ code to be able to work with it.
Personally, I'm pretty much treating it as an extended standard library these days.
(note : I am joking - I don't really want the faithful to die of radiation damage. I'm not Dawkins, ffs.)
I think your implications about what Dawkins wants to happen to believers are wrong and slanderous, and I think you owe the man an apology.
Now, to be completely clear - I feel overselling bandwidth is wrong. I feel the proper response to issues like this on the larger network is guaranteed access to the full amount of bandwidth sold at all times.
I disagree completely. Overselling is a very sensible tactic, and in fact about the only thing that allows you to get flatrate lines as cheap as they are are. Of course, it's a matter of degree, and many telcos are overdoing it, but on the other hand, a strict no-overselling policy would, in practice, lead to upstream capacities on part of the ISP that are used at maybe 20 percent at best - at peak times.
Besides, an ISP doesn't have just one big fat line to "The Internet"; they are part of the internet, and they have a number of connections to a number of other networks, with vastly different capacities. (And usually at least on Tier-1 or Tier-2 upstream provider that connects them to all those networks they cannot or don't want to connect to directly.) If you wanted to take this no-overselling rule literally, you'd have to prepare for some extremely unlikely scenarios, like for example, every single one of your 20 million customers wanting to download something from some obscure location in Madagascar at full speed, at the same time.
They are not prepared for this sort of thing for the same reason that traditional telcos aren't prepared for something like every single person in Chicago calling someone in NYC at the same time: That sort of thing doesn't happen.
The $500 PC discussed in the summary only has 512 megabytes of RAM.
What?! 500 USD for a PC with just 512 MB of RAM? Is this a joke?
I think you are missing the key difference:
Standing up in a physical meeting and showing your face and possibly your name while saying this has a much higher chance of actually swaying or supporting anybody's opinions than an anonymous post on slashdot could, regardless of how true or untrue it is.
The AC post on slashdot will probably be dismissed as just a troll, probably even by those who actually believe this.
The speed of an electron is pretty darn slow (on the order of inches per hour, IIRC)
How do Cathode Ray Tube monitors work then? I was under the impression that they're firing a constant stream of electrons from the back to the front. I don't think these vacuum tubes are filled so densely with electrons that this could possibly work in the tiny-game-of-billiards kind of way...
I suppose you're referring to the tables on the page "Comparison of DNSSEC and DNSCurve". Well, I was referring to the big navigation bar on the top of every of those pages.
... And he's using them where they are not supposed to be used.
<table> is only for tabular data.
The most famous example of an operating system that wrote its own TCP/IP stack from scratch would be Linux, but it is not the only one. You're not going to argue that Linux doesn't count because it is irrelevant these days, are you?
Serving coffee on aircraft causes turbulence.
