"Some people, when confronted with a problem, think "I know, I'll use regular expressions." Now they have two problems."

- Jamie Zawinski

If you want to avoid the overhead of spawning new processes you might want to look into IPython Parallel:

http://ipython.org/ipython-doc/dev/parallel/

If you use that you can keep your "engines" (= processes) running to avoid the overhead of spawning processes. But the inter-process communication will still be slow (I believe they also use pickling) unless you use MPI for communication (which limits the datatypes that you can transfer and adds some extra programming overhead).

But spawning processes is very slow. And more importantly communication between processes means pickling and unpickling objects which in my experience can be a showstopper due to the performance penalty. I guess this is a consequence of the fact that the multiprocessing module is very general and can run on several nodes. So my question is:

Will Python get a fast parallellization module for CPU bound problems on shared memory architectures?

But multiprocessing is heavyweight. It is very expensive to launch new processes. Which leads to my question to Guido:

When will Python support lightweight threads for CPU bound workloads?

It ain't over till the fat lady sings...

Get your quotes straight (there are multiple versions of the "first they came..." story but none with muslims:-):

First they came for the communists,
and I didn't speak out because I wasn't a communist.

Then they came for the socialists,
and I didn't speak out because I wasn't a socialist.

Then they came for the trade unionists,
and I didn't speak out because I wasn't a trade unionist.

Then they came for the Jews,
and I didn't speak out because I wasn't a Jew.

Then they came for the Catholics,
and I didn't speak out because I wasn't a Catholic.

Then they came for me,
and there was no one left to speak for me.

"For one, they've failed the address the perception that unions protect lazy workers at the expense of the productive ones."

That is not a perception but a fact. And at least in my old union they did it openly. I was employed at a public institution where part of the salary was fixed (based on seniority) and a minor part was individual. The individual part is however not negotiated between the employee and the employer but between a union representative and the employer (also for employees not in a union - effectively forcing people into the union). One of the negotiation tactics that our union representative used was to deny productive employees that the institution wanted to reward any increase in wage supplements since the union wanted the money spent on the low productivity employees.

Fortunately I left that workplace for another workplace where the unions are much less influential.

"I guess eventually a lot of professors would be too if they were to release texts or videos and compete with one another."

I believe that this is why the top universities are betting on this. My expectation is that the top universities will provide cheap education for the masses in the future. In other words: The MOOC revolution will be similar to how TV took over from movie theatres.

Second tier universities will simply disappear or the professors will be relegated to teaching assistants for the top professors. Or more likely: Second tier universities will play another role as meeting places for students taking courses from the best universities - and possibly with the professors as TAs. Or they will have to concentrate on areas where they can compete with top universities. I am happily looking forward to this development.

"You could have said the same thing about mass production, and predicted that the world would still be full of serfs and slaves 200 years in the future, but you would have been wrong."

But the problem is that until now technology has generally acted as a productivity multiplier for the general population. What will happen at some point is probably that humans are in fact not needed any more. And we will therefore only be able to earn money by capital investment (in non-human based production). And that will have a tremendous effect on the distribution of wealth in the World and will probably lead to revolutions and worse.

Robot (with RMS's voice): Dave, please don't use the term "open source" or I will have to restrain you and read the GPL for you until you repent. And if that does not help I will force you to use emacs until you learn to say "free software".

(although I am a vim user no flamewar is intended)

Obligatory: http://photos1.blogger.com/blogger/4772/1767/1600/PiMpoRioR.jpg

I agree. In a historical context small girls were probably given away (sold) as spouses in Arabia. And that is history. But all islamic schools insist that the behaviour of Mohammed is perfect and should not put into a historical context. And that is exactly my point. It is good that the parts of the life of Mohammed that are totally unacceptable in the world today are drawn into the clear sunlight. It must be made clear that if muslims want to live in harmony with the rest of the World they have to abandon the notion that the Quran is perfect and is not part of any historical context.

"Frankly, these provocations won't help anyone."

I agree that the Mohammed film in question is terrible. But it does address some important issues with Islam. And that will help us non-believers in the long run. For example the issue of child marriage. Aisha, one of Mohammeds wifes were 9 or 10 when the marriage was consummated. And we should of course (also as non-muslims) be free to debate if that is behaviour befitting for a prophet. Another example is the glorification of violence against non-muslims. At least as long as the main schools of Islam argue that the behaviour of Mohammed is perfect and should form an ideal guideline on how muslims should live in the 21st century.

Another point is that this film is only one pearl in a series of incidents where muslims are threatening fellow muslims or non-muslims over perceived blasphemy. The last example was a documentary on Channel 4 (examining the historical roots of Islam) where a planning screening was cancelled due to threats to the historian behind the film:

http://www.independent.co.uk/arts-entertainment/tv/news/channel-4-cancels-controversial-screening-of-islam-the-untold-story-documentary-after-presenter-tom-holland-is-threatened-8125641.html

I believe everyone with two eyes can see which religion produces the most violence. That is a no-brainer.

But no, I believe that you can find recent isolated acts of Christian violence due to perceived blasphemy. But I am also sure that you cannot find it at the scale that we see Islamic violence due to perceived blasphemy. And also on the everyday level: People are regularly executed for blasphemy in some islamic states. I haven't heard of any executions performed by Buddhists for blasphemy. Why is this the case if all religions share the same propensity for violence against those who don't share their faith?

Yes, blasphemy is an important filter for me in this discussion. As an atheist I believe that the best shield against oppresive tendencies in religions is the deconstruction of the religions. And Islam is the only major religion which is hindering this through organised violence at a larger scale.

And Anders Behring Breivik was a cultural Christian (according to himself) who committed a heinous crime. But for political and not religious reasons (as mentioned he was not very religious).