This is one of the strangest sentences I have encountered in quite a while.

First, "belie" is very definitely the wrong word to use here. It means "to show to be false". And second, Ethernet is ubiquitous largely because of its simplicity... there is nothing surprising about that.

I don't have an opinion about Rossi's device, and skepticism is warranted. But outright rejection is not reasonable, for a number of reasons.

In the decades since the Pons-Fleischman debacle back in 1989, there has been a huge amount of evidence that the phenomenon they described is real. The problem is just that it has been grossly unpredictable. Lots of people have replicated their results, just not reliably.

In addition, researchers for the U.S. Navy and other laboratories have been investigating the exact same Nickel-to-Copper LENR reaction for many years, saying that it shows great promise.

Don't confuse outright rejection with skepticism. They aren't the same things. I am skeptical of Rossi's claims too, but until more evidence comes in I am keeping an open mind. Even if Rossi turns out not to be the one who has made this work, there is every reason to believe there is something to this technology.

Here's the problem: they're pulling the same blunder that got Intel in hot water back in the 90s.

People aren't stupid (although many of them act that way sometimes). If you can take a part that sells for $1,000, disable some of the functionality, and sell it for $650... then you can sell the whole unit for $650. It's a ripoff and people know it.

Now, if it's a matter of disabling cores that don't pass testing anyway, that might be an effective way to dump "defective" parts on the market and still profit from them. But it's pretty hard -- for very good reasons -- to convince people that it doesn't actually cost you MORE money to disable only those that don't pass tests... meaning you still could have sold those better parts for the lower amount.

Wow. I don't know how that sneaked in there. The final paragraph should simply have been:

"The 'not very practical' designation simply means you haven't been using your imagination."

First off, the kind of OTP situation discussed in TFA isn't even a very common concern: using forensic tech to recover the key from memory. Either you'd have to have some kind of computer with virtually no garbage collection of free memory, or the computer would have to be seized immediately after having sent the message. Just not something that is going to happen every day.

But to address your own comment: you must keep in mind that almost ALL modern encryption has the problem that a key must be generated and distributed. Public-key or asymmetrical cryptography is great for certain things, but even that relies on keeping a key secret (although the secret key does not have to be distributed).

The point is: as long as your OTP key remains secret (if the key is pretty random and of course if it is used only once), then it is inviolable. There are many extremely practical uses for such technology.

For example: you can dispense with the "one time only" requirement if you simply want little more than a "keeping the kid sister's prying eyes off my email" level of security. You can dispense with the key distribution requirement if you have agreed upon a common external changing key source. And so on.

The "not very practical" designation simply means you haven't been using your imagination. common kind of OTP, nor is it a very practical one. The subset they describe is made by adding characters to a message from a random source to obscure the message.

That's only "a problem with my argument" if you assume I am an idiot. Of course I meant reliability in comparison to existing modern firearms, which are pretty darned reliable.

Of course I considered it. You are trying to insert my comments into some kind of weird context that doesn't exist.

2%??? Um... no. Let's not. Let's stick with more realistic figures. For example, in 2011, according to the FBI's "Law Enforcement Officers Killed and Assaulted", out of the many thousands of times officers drew their guns, 3 were killed with those guns. The actual figure is far closer to 0.01%, probably even less.

Now does that "99.99%" figure I gave start to make a little more sense to you?

If you're going to make an argument based on statistics, you should at least take the trouble to find out what those statistics are.

It's not even a good one.

The problem is not that such identification technology is impossible. The problem is that it is impossible to make it that reliable with today's technology, while still being able to perform its security function (keeping the bad guys from using the gun). You can have one, or you can have the other, but so far -- despite several products that have been hyped -- nobody has come even close to doing both. You have to have both or it's worse than useless, because it will get good people killed.

If it were that simple, it would be done already. For one thing, RF is FAR too easy to foil. People have been experimenting with that kind of technology for many years now, and not one has come even close to putting one on the market. Again, the main reason is maintaining reliability and security at the same time. One or the other is relatively easy. Both are not.

It's not even a good one.

Sorry to have to say that, but in good conscience I have to say that. You ignored some rather gaping holes in the problem domain. But don't feel bad; even the makers of safes and other security equipment famously suffer from tunnel vision in that regard. For example: in recent years not one but many manufacturers of safes with electronic combination locks focused on the security of the electronic lock but paid little attention to the physcial locking mechanism. As a result, one security researcher produced a YouTube video of his 4-year-old, with no knowledge of the combination, opening some of them with ease.

Security vulnerabilities surround everything we do. It behooves one to step back and look at the larger picture. We are not yet even close to the technology required to put reliable and secure "lockout" mechanisms on guns. At the moment it's still science fiction. Biometric identification is not the problem. The problem is making it absolutely reliable, while ALSO rejecting the unauthorized.

I don't have case citations at hand but you can look them up. In particular, find copyright cases surrounding player piano rolls, in which the courts ruled that it made absolutely no difference whether copyrighted works were used to control a machine.

John Philip Sousa was famously involved in some of those suits.

Fortunately there have been a few judges lately who have an actual head on their shoulders, and who have ruled that simply telling somebody their fly is open is not the same as rape.

But these B.S. laws, like CFAA and DMCA, need to disappear. They were ill-conceived and we KNOW that they cause problems. Not little problems, big ones.

I would keep the safe-harbor provisions of DMCA, and scrap all the rest of it. Same with CFAA.

F*cking clueless judges. There is precedent going back well over 100 years that software is irrelevant to the nature of the machine. Examples: different cards in a Jacquard loom do not make it a "different" loom. Different player piano rolls do not make it a "different" piano. Etc. There are actual court decisions to this effect.

Not to mention that it is also common sense.

It is nothing of the sort. They went a long way toward throwing away the tremendous gains of the 2009 "new" Start Trek movie.

The first movie took great pains to give them a brand NEW Star Trek world with all the possibilities that implies. It breaks all necessary ties with the past, and gave them a new start.

So what did they do? They made the second movie a blatant derivative of "The Wrath of Khan".

With all that possibility, they came close to throwing it all away. As it is, it was WAY too similar to that other Khan movie.

Pardon me, but I go to movies to see new things. This wasn't it. My rating: FAIL.

No, it isn't. My comment has nothing to do with trying out new stuff. I try out lots of new stuff... that's how I learned about all the stuff I was commenting about.

My comment was about people who put stuff out there they think is new, when it isn't.

You missed my point.

Of course new stuff is built by people who genuinely wanted to improve on the old stuff. And that *IS*, as you say, how we improve.

But the fact is that most of them fail, and many of them -- many that I've seen in recent years, at any rate -- fail because they did not know their history. If they did, they would have known that it had been done before and failed then, too. Often for very good reasons.

If you're trying to improve something, make sure it's an actual improvement, yes? Rather than a reversion to something that we already knew doesn't work well.

This.

In recent years I have seen so much change for the sake of change, it sometimes disgusts me.

Let's get something straight, folks: Change is only good if it's an improvement. Otherwise, change is BAD, even if it's just as good as the old thing. There are a number of reasons for this.

First among those reasons is that change has definite costs involved. Whenever you change something, people have to learn new ways, use something differently, etc., etc. If anybody can find some kind of major change that doesn't have a cost associated with it, I'd be delighted to hear about it.

Second, things are usually the way they are for good reasons. There are generations of people who came before who tried different things and arrived at their ways via hard-won trial and error. Changing something "just because" probably means you don't know your history and, as they say, will likely be doomed to repeat it.

When I think a change is GOOD, on its own merit, I am happy to jump on the bandwagon. But I don't drool over things just because they are new or in fashion.

This was predicted back in the 1930s, too. How did that work out for them?