Thanks for the link to Stone's article. It is a good read. I haven't gotten through it all the way yet, but he seems to be making a very subtle point, one that is lost on 99.9% of the pro/anti-GM activists.

He side-steps the GM is good/bad argument and really focuses on the people and their deskilling. In fact, he says that he's not making a judgement at all on GM crops.

The really interesting is deskilling and its affect on all of us. Many of us in the industrialized world have been deskilled to the point (for example) that we can no longer fix our own car, or build our own house, or forage for food in the forest.

This deskilling is a trade-off. In return for letting someone else make some difficult decisions (choosing seeds and deciding where to plant them) or do some technical work for us (fixing cars, building houses), we get some benefits like more money, more free time, etc. But, then we start relying on others to make decisions for us.

Are the benefits worth the loss of skill? Sometimes, but not always.

Imagine trying to build a bridge where the melting point of steel halves every 6 months. The fact that computers have been getting so much faster so quickly means that even the most fundamental things we learned about programming 2 years ago are no longer relevant.

Also, because our computers are getting so much faster, they can handle much more complicated programs. 50 years ago, a large program was 100 lines of code, in the 60s, it was thousands, in the 80s, it was hundreds of thousands. Now, large systems have millions of lines of code. These things are arguably the most complex things that humanity has yet designed. New computers can handle this. But, how much has human intelligence expanded in the last 50 years? I still can't remember my phone number...how can I understand 3 million lines of code?

Because our computers can handle it, people expect that programmers can program to match their speed. We're still lagging behind.

Thankfully (for programmers), Moore's law no longer holds. Computers are not doubling in speed every 6 months. Finally, programming practice can catch up with the hardware.

This little excerpt hits on one of the biggest problems in programming today---programming languages are horrible. The languages in current use are way closer to how computers "think" than to how humans do (even though they are much better than they used to be).

It takes years to become profficient in the major languages, and so it is specialists who do the programming. Unfortunately, it is also specialists (of a very different kind) who need to use the completed programs. The specialists of one field (say Doctors, Lawyers, airplane mechanics) have very precise requirements and very precise terminology for what they do, and cannot demonstrate the nuances of their field to other specialists (the programmers). This is why software is always late, buggy, expensive, etc.

Until we can make programming languages that non-programmer specialists can use to create interesting things in their own field, we are going to have the same old tired complaints about bugs, quality, and geeky programmers.

COBOL was a great language for its time. It made programming accessible to the financial field in the 60s and 70s. It served its purpose well and many accountants stock brokers were able to get their work done must faster and better using it.

But, there are two issues with it.

First, the language was designed around the original machines it was supposed to run on (and not the other way around). Therefore, as computers changed, COBOL became more and more outdated. Languages need to be designed for the people that use it, so they can easily express their intentions to the computer.

Second, it was great at number crunching, but that was it. Payroll, accounting, and all that other back-end stuff that used to take people months to do by hand. Try applying it to anything that we really want to do now (for the web or for mobiles) and it's a horrible mismatch.

Unfortunately, we are still stuck with the language, with some 180 billion lines of it floating around powering the world's financial systems, running our banks.

When I worked on Wall Street a few years ago, there were 3 floors of my building devoted only to COBOL programmers. All gray-haired and all about to retire. Knowledge of how our banks actually work lies with them. When they're gone (and they will be soon), the banking system will still calculate interest and print statements and do payroll, but no one will know how it works.

For once, I agree with SR (well...maybe not the first, but it is rare).

Andrew, I hope your father pulls through. It must be a horrible experience for you.

That being said. It's still unclear to me whether or not there are truly more incidents of cancer in our society today. Are we just better at detecting it? Are we just healthier in most other respects that cancer itself is easier to spot because we are not dealing with polio, smallpox, measles, and any number of vitamin deficiencies?

I honestly don't know and would really like to see some statistics one way or the other.

There was a wonderful NYT article a few months ago that compared medical records from the American Civil War to today and it is just amazing the difference in health between now and then.

http://www.nytimes.com/2006/07/30/health/30age.html?ex=1170824400&en=5f7121e33571dd42&ei=5070

--or--

http://preview.tinyurl.com/2eopye

(you may need a paying subscription)

Whereas back then, people in their 30s would suffer from crippling arthritis and a host of other horrible ailments. Today, it's something that doesn't even cross our mind.

I'm not trying to downplay the importance of cancer research or the trauma that it puts individuals and their families through, but I am trying to emphasize that we've come a long way.