More samples is good, but I question your statement that the sample size in question isn't large enough. The fact is that the DNA is present, it shows a link to a previously unknown Asian progenitor population, a population that also bequeathed their DNA to some south Asian populations and to indigenous Australians. Better sampling may identify other areas of the Americas where these genes ended up, which would help understand migration patterns not only from Asia to the Americas, but also within the Americas themselves.

That you're imagination is so limited should not reflect on science's need for basic research. Unless you have a crystal ball that accurately reports what basic research taking place now is going to blossom into value further down the road, all you have is a pretty goddamned mindless anti-intellectual rant.

My understanding is not that an Austro-Melanesian population contributed to American Indian DNA, but rather that some progenitor population in Asia bequeathed their DNA to both groups. This puts a somewhat different hide on the cat, underlining that the Paleo-Indians were not a homogeneous Siberian population, but rather themselves were descended from a patchwork of groups in Asia who entered and likely interbred for centuries in Beringia.

Simplistic solutions are attractive to bean counters, politicians, and even voters. All three groups are frequently unoccupied by unsophisticated people who may have certain degrees of tactical intelligence, but tend not be able to think through a given proposal, but rather look for things that are somehow emotionally or ideologically pleasing.

Yep, there are a lot of flashing lights and beeps sending code in the computer world. I still hear phones using "... -- ..." to indicate an incoming SMS.

OpenSSL has first-to-market advantage, and anyone who hasn't evaluated the quality differences will choose the simpler license. Plus there are other alternatives, like Amazon's new SSL-in-5000-lines which is also gift-licensed.

The time for OpenSSL to dual-license was when it was the only available alternative to entirely proprietary implementations. That might indeed have funded a quality improvement.

I don't know a thing about the quality of GnuTLS or the Amazon thing. I've seen enough of the insides of OpenSSL to know it's not pretty, but am not a crypto guy and this don't work on it.

Maintaining FIPS compliance did not make anything easier. It's essentially a prohibition on bug repair, as you have to recertify afterward. But the people who wanted FIPS were the only ones who were actually paying for someone to work on OpenSSL.

I don't think any of the other Free Software projects ever tried to be FIPS certified.

If you are one of the infringed parties, I'd be happy to talk with you about what your options are. bruce at perens dot com or +1 510-4PERENS (I'm not there today, but it will take a message). I am not a lawyer but I work with the good ones and can bring them into the conversation if necessary.

As a community we've managed to almost completely ignore that because of their use of dual-licensing, MySQL made 1.1 Billion dollars after 9 years in business, and that for a database that was written by one person, and the code base remained available under the GPL.

IMO, 1.1 Billion dollars is pretty damn impressive. Especially if you get paid that to make Free Software. Heck, sign me up!

Oracle was a bad actor, and Monty is now leading further development of that same code base under the GPL. But it did not have to be that way.

How do you prove damages or have the right to settle violations if you don't have copyright?

If you have been doing enough work to justify getting paid for the software, you have an ample amount of your own copyrighted work to base your claim upon. If you haven't done that much work, what are you suing for?

You can also get a grant of the right to sue from your contributors. You can include in the agreement how you will apportion damages: for example you could take the ratio of your lines of modified code checked in vs. that of contributed code checked in, and give that portion of damages to FSF.

You do have to get either copyright assignment or the right to relicense from your contributors, if you have contributors, or the whole scheme doesn't work. But maybe you get enough money to pay them some.

Yes, writing Morse Code Software is one of the creative and educational things you can do with Morse code.

It took me 60 days to get to 20 WPM, working for a long time every day.

In contrast, it took a lot less time to write an interrupt-driven, terminate-and-stay-resident Morse Code sounder program in 6502 assembler. And I learned the instruction set, too.

I'm not saying you don't want to do either. It just doesn't belong on the test.