"A database management system is nothing but a fancy filesystem with structured files, yet they are often used in servers and perform just fine."

Yes that is true but "perform just fine" for a database system means something different from "perform just fine" when you are talking about a file system.

For example many people would be happy to see their DBMS do 500 table inserts per second but a file system that could to only 500 writes per second very slow.

All this talk about what should go inside the kernel is pointless now that for years we've had a modular kernel. You can configure in or out what ever you like. If you want a user space file system you can have that or not have it. It's all very configurable.

You are correct. What is needed is a new instruction set. But we don't need new machine level instructions. Source code level is good enough. A programming language or API that works at a higher level is what's needed. Currently you have to get down in the details of synchronization to make threads work. What's needed s a more declarative approach where you can "mark" sections of the program with informatin about how they interact.

Object Oriented Programming is not a bad way to write parallel code. If the objects model real-world objects that each can be on it's own thread. Problem is that most dumb programmers idea of "object" is a GIU widget.

In the end it is more about educating developers.

Any computer with at least two cores or two CPU chips. I have two systems one is a dual xeon powoered HP running Linux with VMware's "server" and the other a dual core Apple iMac running Mac OS X with VMware's "fusion. They both work well. But in any case max out the RAM.

In fact if you are buying a computer just for this purpose, Buy yhr one that has the largest maximum RAM. You want at least 4GB. Also get the best disk system you can afford.

Processor speed is not really what you need but you do want a modern Intel Core2 and loads of RAM

Study something very close to home. I think the ecology of cracks in the side walk would make a graet topic. Where do nutrients come from? How to weeds compete for them and under what conditions do some win and loose.

Have the weeds evolved to fit this new environment that people craeted only in the last century. Are the wild weeds geneticaly the same as those in urban areas. Theory says evolution can happen with new environments. are we seeing this?

For Linux using the release of 1.0 was something that happened late. the .9 series was long lived and mature. I'd been using Linux for a long time before 1.0 was released.

In people years, Linux 1.0 was more like a high school graduation than a birth. It meant that Linux was mostly grown up,

For Linux we do have a very good and well defined "birthday". The day Linux posted on usnet. Use that.

Use opendns.com as your DNS provider. I do this to filter ads and popups. It works with any Internet connected device from PCs to iPod Touch.

We are almost a binary planet system, and that is pretty rare.

Back when Pluto was a planet, 2 out of 9 planets were "binary". Is that "pretty rare" that's a judgement call.

Now that Pluto is no longer a planet 1 in 8 plants is a binary. Is that "pretty rare"?

But if we only consider rocky planets then one in 4 is "binary". That's 25%. Hardy rare.

We will have to wait many years untill you can see many rocky exo-planets before we can have a statistically significant number of them toknow if Earth is rare or not. Our current sample space is to small. Now all we can say is "25% but with a huge margin of error".

So if I work for IBM or Microsoft, every day I need to send an e-mail to the entire company saying what color socks I'm wearing and ask for others to tell me about their socks too. And then starting about 10:00am we can start planing who is going out to lunch and who has a car large enough for the group.

When ever you are up at level when your upstream pipe is running full speed then you are already in effect throttling. So the idea of limiting your customer's bandwidth is not even a question. You ARE doing it now.

The question then is how to do it fairly. To do nothing is to throttle very un-fairly

My opion is to prioritise the packets. First off
Interactive protocols should go ahead of "background" stuff, that is "stuff" where the user is not sitting there at the screen like email, downloads (FTP and p2p) Next those using only low bandwidth should go to the front of the queue. (Let those not using there full share go first in line for more.)

This way those doing normal web surfing see fast page loads and those downloading pirate copies of DVD get whatever bandwidth you have left over. This is the only fair way. It is NOT fair to let those getting the DVD rip-offs stomp all over the others nd if you do nothing that is what happens.

One More Thing: You can cache pages. This will take a load off your pipe. There are any number of free page caches that will run onder Linux. And big disks are cheap now.

Everyone misses the point. We don't care how long it takes to LOAD a web site. For those on a typical home DSL line the speed of the connection determines how long a site takes to load.

What matters is AFTER the site is loaded and Javascript code is running locally in the user's computer. This makes web based applications possible, web mail, web calenders, web based office suites and so on. This is why Chrome matters

"Chances are if you don't like christian music, it is because you have a personal vendetta against anything christian, and well, I wont argue with bigots so you would be on your own."

The word vendetta is rather negative. One can have other religious beleifs with out hating Christians. To the non-indoctrinated Christian music sounds silly but harmless. Like people signing about how the Tooth Fairy loves them. "Vendetta" is the wrong description. "chuckles to himself and then moves on." is I think more common

How come when companies break the law they get to "agree" on the punishment?

This is like asking "How come cows can fly?" They can't so the question is meaningless. In your case companies do not do anything, people within those companies do things. Because there are many people within a company many things can be done at the same time. Some times even at cross purposes.

In this case one person did the price fixing. Then later another person agreed to pay the fine. Almost certainly this was not the same person.

OK so the battery car battery could be charged in 10 seconds but who has "kilo-amp" sized outlets. What are you going to plug your charger into?

Not "Brainless research" but "brainless reporting of research".

Or maybe brainless readers. Yes the problem is with people who get there science education by reading web Blogs and CNN headlines. The people who write forthose outlets are really trying to sell ad space. Don't blame the researchers because to ideot blogger jumped on the story and bend it around to snag a few more click-thoughs on his web site ads.

"I don't know" might be a frequent response) which (by my definition, at least) 'understands' the question."

Actually if the machine is being truthful "I don't know" is a very sophisticated answer. It means the machine has searched its store of information and deduced that the answer can not be derived from that store.

The trouble is that as a software engineer, I know that the designers will simply design the system to say "I don't know" whenever the system fails to find the answer.

Failing to find is MUCH different than knowing for certain something is not there.