That's true -- I forgot about that. The one caveat to that is that it is primarily targeted by the XenServer product: Xen, KVM, and QEMU tend to use the qcow or raw formats. I'm also worried about new features coming out from one or the other VMMs that aren't supported by the current format. Without an established way for anyone to adapt the format, I'm not sure it will provide a long term solution. I do have a specific feature in mind, but unfortunately you will have to wait for the colleague's PhD thesis to find out about it ;)
Go to Borders, get the Moleskine knock-offs made by Piccadilly. I found myself migrating to the larger size instead of the roughly 3x5 inch small ones.
Thank god, as that's easily the worst thing Stephenson's ever written -- like a rambling inept Slashdot comment, but stretched to 160 pages -- it's even structured around an extended simile between user interfaces and fictional automobiles, for fuck's sake!
On top of all that, it has the misfortune of being written at the worst possible point in time:
* too long after mainframes, minicomputers, or basic micros
* right before the death of BeOS (which he was infatuated with)
* right as Apple is about to die from its own bullshit
* long after NeXT has faded from relevancy but before the buyout
* In Linux's middle-aged (~2.2) years when a lot of things were bogged down
in wank -- well after the initial excitement, but well before BK and git
As an example, consider that no matter what interpretation of any tax law you get from an IRS employee, it will not hold up in court - if an IRS agent tells you to do A, then the IRS decides later to disallow A and that you should have done B, they will hit you up with taxes and penalties and you have no recourse.
UNLESS, you pay money for an IRS administrative ruling; then, you can use that response in tax court. The key is that the cost for such rulings are $50K. Each.
This is more like an ad than an article. Flow-based routers have been developed in the past, and the reason they are not prevalent is due to performance issues of one kind or another.
If they truly have solved the problem(s) that other flow-based routers had, great, but this can only be determined in the real world.