Agreed---he's being dishonest in some regard. My anecdotal experience with my Honda Insight (an $18k car, BTW) is that I get average 44 mpg on my regular commute, which is stop and go at an average speed of 20 mph. My particular commute, which can include at 5-10 minutes of idling per direction, absolutely kills non-hybrid cars gas mileage. When I do a trip the the airport, with mostly freeway speeds, I can get 55 mpg for an average of the whole trip.

I couldn't agree with you more when you write that "the end is what GPL is about"---and that's the crux of my analogy with communism. GPL works when everything is GPLed (or with a more permissive license). This means that the "end" is an elitist club of software with homogenous licensing. Everyone is open and playing by the same rules, nobody "gets ahead". This is perfectly wonderful ideal and it is, without a shadow of a doubt, an ideal that can only be created by "telling people what they cannot do."

It sounds to me like we agree. Whether it's what "GPL is about" or a "means to an end" is largely irrelevant from a practical perspective.

To be absolutely clear, I'm fine with people wanting to GPL their code, but they shouldn't delude themselves into thinking that they're not being restrictive or imposing rules. For anything that I open source, I think GPL is too restrictive. ge7 said it better than me when he wrote,

Straw man! You responded to what you wish I had said, not what I did say.

Oh BS -- GPL is a pie-in-the-sky communist-like ideal in which you get kicked out of the commune if you actually try to assert some independence. That fine for sitting around and singing Kumbaya, but real open source is definitely closer to a BSD license as stated by the OP. Don't fool yourself into thinking that GPL 'is not about telling people what they cannot do' -- it sure as hell is.

That joke would be way funnier if there were actually a difference between a grizzly bear and a brown bear (they're the same thing). I think you mean black bear...

Uh, I think that was President George W. Bush that reported that in his 2003 State of the Union. But you're absolutely correct, if GWB were reporting this I wouldn't trust it either.

That doesn't answer the question. Obviously that's how you would have to design it. The question is what code would you have to write to accomplish what GCD with blocks allowed.

I mean, programmers execute something on a separate thread and then make notifications on the main the thread all the time. The beauty in GCD with blocks is what little code you have to write to accomplish that task. Having your code stay contextual (like in the wikipedia example) is far better than splitting it across multiple functions. Additionally, that concurrent task was automatically handed to a system-wide load balanced thread pool.

So, to answer the question just show the equivalent code in Java that maintains the same benefits that GCD does.

I think that the usage patterns with GCD are a little bit more broad than you're giving it credit for. Your audio processing example, for example, works just fine with other more primitive parallelization libraries. You could have easily done that with NSOperation before GCD was around. The question is can you do the equivalent to the -analyzeDocument example in this wikipedia article, http://en.wikipedia.org/wiki/Grand_Central_Dispatch with your Java API? It's an honest question, BTW.

Oh *snap*! Of course, I'll just steal the content since its not legally available on my Linux version... man, you really got me there!

Oh, that's fantastic! You mean the Linux versions will offer the same books without DRM?

Didn't think so.

I agree completely -- you leave a loaded gun and a three year alone in a room together, what do you think is going to happen? Of course the three year old is going to play with and examine the new object! At that age they are just naive and curious explorers with little fear. The Wii controller has little to do with this.

RTFA. From the THIRD and FOURTH sentences in the article,

They addressed exactly the issue you cite in the fourth comment of the article. From the comment,

That's a really annoying answer, but completely correct.

The math classes aren't just about learning the topics, but learning how to approach a problem. I think your goal should be to have strong enough fundamentals in mathematics that when you encounter a math problem in a topic you haven't seen before, you are comfortable pulling out a book and figuring out how to solve it. It sounds like you're not very comfortable with mathematics, so a good solution is to take enough classes that you do become comfortable.

That's completely unavoidable. For example, politicians must decide

• whether power lines or cell phones cause cancer and how to regulate it,
• whether vaccinations cause autism and if we should ban vaccinations
• whether the H1N1 vaccination should be immediately distributed,
• whether humans contribute to global warming and if we should regulate our contributions.

These policy makers can't be expected to be experts in all of these fields. I don't actually expect Senator Lisa Murkowski to understand enough immunology to look at the lab notebooks of the scientists and know whether or not the H1N1 vaccination they created will kill you outright. What I do expect her to know, is who's opinion she should trust. When the scientists at the CDC tell her its perfectly safe and will actually prevent infection, she's just gonna have to believe that and make policy choices based on that information.

No, it's not. How can I even know if I agree with their analysis? For example, if these scientists are telling me that adding GFP to tomatoes that I eat won't cause cancer, I'm not going to take the time (years!) to learn about details of molecular biology in order to evaluate the biochemical pathways they used to back their claims. I'm just going to trust that they know what they're saying. Even if Joe Blogger tells me they will cause cancer because genetic engineer is bad, I'm also going to just have to trust that when the scientists tell me his claim is bogus, that they actually listened to what he had to say.

Yes! If the scientists won't even listen to the claims, then they're not doing their job. We definitely agree here. That's the scientists responsibility to listen to them. But, I sure as hell don't want Senator Lisa Murkowski giving this other guy equal credence if the scientists have already refuted the claims. The politician and the layperson should not simply assume that there are two sides to this story and consider both claims. Neither is actually equipped to evaluate the claims; they simply aren't experts and haven't invested the time.

In my reading of the original article, the author decided to go the route of giving both the scientists and the blogger equal credibility. He took the time to investigate the claims a little deeper (as deep as his understanding would go) and then found that in the end scientists were likely right. His conclusion was simply that he wasted his time and should have stuck with believing the expert. He's not arguing the other guy is automatically wrong just because he's not an expert, but he is arguing he, a layperson, should have stuck with believing the experts; the system worked.

The broader conclusion, from my perspective, is that this is generally going to be the case. Even if I invest 10 years of my life educating myself such that I completely understand the intricacies of some scientific field of study, I'm pretty sure I'll likely agree with the general consensus in that field as it currently stands. This is why I don't actually need to educate myself about everything; I can evaluate claims simply by considering the source. In the case of a general consensus amongst scientists, I trust that the scientific process was followed when evaluating those claims.