I think you are pretty close, but I think that the real problem for the NSA is the possibility of real competition to provide internet access. Imagine how tough the job will be if the NSA had to get cooperation from hundreds of ISPs like they have in Japan. The duopoly here is very convenient for the NSA but a nightmare for the rest of us.

Had we declared the owners of the pipes to be common carriers and imposed open access rules upon them, we'd have something like what Japan has: fast internet access with hundreds of ISPs vying for my money. Instead, we have cable and telcos who operate on one principle: make sure that the CEO can have a few vacation homes sprawled across the world, send his kids to private schools until they are married and allow him and his extended family to live in gated communities. The members of the board of directors get similar benefits, but to a lesser extent.

Oh, one more thing. The corporations participating in the duopoly need to siphon enough money from the economy to capture the agencies that regulate them, except for the NSA, which in theory, can't be bought. Snowden proved that, but not in the way the NSA had in mind.

In sum, the duopoly will slow the net down, but it will also provide a few powerful leverage points for the government while concentrating revenue into a few companies willing to cooperate. Yeah, that sums it up.

This is an interesting point. No matter how much data they collect, someone has to look at it and make a judgement call about it.

Nice sig. Nice post, too. We each have to choose our battles carefully. In the meantime, enjoy life.

A farm of computers, eh? How big is it? Here's an article from Schneier that discusses the physical limitations of computation relative to brute force attacks for private keys:

https://www.schneier.com/blog/archives/2009/09/the_doghouse_cr.html

To put it simply, brute force attacks are about trying every possible combination in a counter. Just to run a counter through 256 bits, You're going to need all the power of the sun for 32 years or more, or a supernova. Take your pick. That doesn't include power for any other useful computation. And then of course, there is time. How much time do you have?

The computer scientists of the world who believe in freedom will be happy to put the kibosh on on any code that permits side-attacks on encryption software. That is where the weakness is more likely to be, not the encryption algorithms.

Now I could be completely wrong about this, but based on the best available information I have, I don't think anyone is capable of brute force attacks against strong encryption except for poorly implemented crypto or really weak passwords.

Yeah, I guess you're right. Your examples seem worse than forgetting to update the SSL certificates for Azure, or permitting a complete failure of the London Stock Exchange for 8 hours the day after the Fed announces the bailout of AIG. For an interesting analysis of Microsoft network security that was presented at DEFCON 18, go here: http://www.youtube.com/watch?v=RXFoS8HTI6E

If after watching that, you still feel that Google is doing such a poor job with reliability and security, good luck with Microsoft.

Wow. What a showstopper.

Given that Microsoft isn't alert enough to keep their SSL certificates up to date with Azure (one, among many other high profile failures), I trust Google more than Microsoft. In the end, I can only hope that my trust is well founded.

Microsoft will never let lawsuits go to trial if they can help it. They surely love those NDAs that help seal the deal. Barnes and Noble made a pretty big stink about it, and with it, offered good reasons to believe that Microsoft was one of the biggest offenders.

Bessen and Meurer are in agreement. They found that patents tend to substitute for R&D. I think they tend to substitute for customer service, too.

Those damn open standards keep getting in the way.

So lets see if your claims of progress are born out by the facts in the paper:

"Simply eyeballing the big trends shows that patenting has exploded over the
last decades. In 1983 in the United States, 59,715 patents were issued; by 2003,
189,597 patents were issued; and in 2010, 244,341 new patents were approved. In
less than 30 years, the flow of patents more than quadrupled. By contrast, neither
innovation nor research and development expenditure nor factor productivity
have exhibited any particular upward trend. According to the Bureau of Labor
Statistics, annual growth in total factor productivity in the decade 1970 –1979 was
about 1.2 percent, while in the decades 1990 –1999 and 2000 –2009 it has been a bit
below 1 percent. Meanwhile, US research and development expenditure has been
oscillating for more than three decades in a narrow band around 2.5 percent of
GDP. The recent explosion of patents, in other words, has not brought about any
additional surge in useful innovations and aggregate productivity. In new industries
such as biotechnology and software —where innovation was already thriving in their
absence —patents have been introduced without any positive impact on the rate
of innovation. The software industry is an important case in point. In a dramatic
example of judge-made law, software patents became possible for the first time in
the early 1990s. Bessen and Meurer, in a large body of empirical work culminating
in Patent Failure (2008), have studied the consequences of this experiment and have
concluded that it damaged social welfare."

That is from page 4 in the report. So for *30 years*, R&D as a percentage of GDP has stagnated despite a quadrupling of patents issued. is that progress? The authors state that there is *NO* empirical evidence to suggest that patent actually increase innovation or productivity. None. I agree with the author and you're free to disagree. But the facts even from the introduction seem pretty plain to me.

If you didn't make your invention by now, someone else would have done it just for the first mover advantage.

Do the same people who stridently assert that labels for GMOs are not required think the same after this?

This is probably the tip of the iceberg.

It only has to start in one state.

We're going to get labeling and that's the end of it. Political will is mounting. It's going to happen. Get over it.

Clearly, caveat emptor is your place. Who needs a disinterested third party to ensure that transactions are fair?