eck

It's possible. There are parallels between main memory and disks. A disk seek gets you hundreds or thousands of bytes and requires (blocking for) millions of cpu cycles; a cache line fill gets you 64 (sometimes) bytes and requires (spinning for) hundreds of cpu cycles. You want to minimize both of those, but with memory, the difference is much less dramatic.

Everyone here seems sure that "Black & Berg" is an actual security company that issued a challenge and actually intended to pay someone money. Does anyone have any independent sources on that?

Just from the look of the site, it seems so much like a farcical joke on HBGary-type companies, I wonder if it's not a viral marketing campaign.

I've programmed both, though, not a PIC for many years now... PICs are just weird.

The AVR is a pretty typical Harvard RISC: 32 registers, memory load/store, direct/indirect jumps, stack pointer pointing at ram, etc. It's still 8/16 bit, so some pairs of 8 bit registers work like 16 bit registers.

The PIC has one accumulator register; ram is a set of (manually selected) register banks. Instructions operate on the accumulator and an operand in the selected bank. The stack is totally separate from other RAM, so it's fixed size (8 maybe?) and can only be used with call/return.

I don't want to be judgmental; there's certainly some strange virtues to PIC, like the lack of any real state to save/restore on ISR entry/exit; but they really just don't fit my mental concept of a CPU.

Exercise for the reader: Given the stack and register bank selection, which one of these has better C compilers?

(Still, if you're learning something new, gamble is right: ARM is the future.)

I'm a bit confused by the usage of "percolation". Generally speaking that refers to, in a tiered retrieval system, a document being initially indexed in a small temporary index then gradually "percolated" upwards to larger more permanent indices. Such systems allow updates without merging into the main index.

See: http://nlp.stanford.edu/IR-book/html/htmledition/dynamic-ind...

If it were acceptable to infringe a trademark for a few months and make $130k, and the only consequence was being forced to stop, a lot of people would think (correctly) that doing so is a profitable activity. Apple is demonstrating that it is not.

He sold a product with the trademarked Apple logo on it without authorization from Apple, and they're protecting their trademark. How is this a story?

Is there a trademark exception for being a kid? No.

Is there a trademark exception for "same shape as different colored component of product"? No.

Now, had he sold a white case without the logo, that would be a totally different story...

Section 1.1.1: "Data transmitted over a network, whether wired or wireless, is susceptible to passive monitoring."

Translation: We don't like those creeps on the other side of the building either.

I'm confused; in the example, couldn't he avoid the filesort entirely by just adding an key on t2.t1id? Is it really optimal to be sorting the whole table -- no matter how optimally -- on any common serving path?

To me, the interesting question would be, were a private citizen to build a tracking device and attach it to an FBI vehicle, would they have a problem with that? My guess is they would arrest you and figure out something to charge you with -- even though, by their own logic, since you could follow an FBI vehicle around to see where it goes, attaching a GPS is equivalent.

A logical standard would be "that which isn't legal for a private citizen to do, the police need a warrant for" -- but hey, I'm an engineer, not a lawyer...

Domains seem sort of like property but on some level are really speech. Were the NY Times to write an article that said, "The IP address of www.whatever.com is 1.2.3.4" would DHS try to shut down the presses? Ultimately, conveying that piece of information is what this plugin (and indeed, the DNS system before DHS got to it) does.