People have a right to support their own values in public no matter what they are based on. Organized religion does not manipulate government. It may manipulate people (voters), but so what? Voters are manipulated by all kinds of ideology. You have no more of a right that the government be influenced by your anti-religious values than others do for their religious ones. What sort of representation in government would you grant organized religion in return for taxing it? Separation of church and state works both ways. Many religious institutions do a better job of providing social services than the government and there are stipulations on how the funding is used, of course. This saves tax payers lots of money. More people have been murdered by atheist regimes (and in modern times) than by all religious ones put together. The influence of religion has given humanity far more in terms of hospitals, universities and other institutions that benefit society. We'd be far worse of without it than we are with it.

For decades Jane Elliott has made a business out of doing this sort of thing to/for people and she's still doing it today. http://www.janeelliott.com/

True. But the statement you quoted doesn't say that unique products are necessarily groundbreaking. If they are groundbreaking, making them also impossible to duplicate does mean quite a lot.

... eventually; often when its too late to do anything quick, easy or cheap to make it better.

"It's not that I can't figure it out. It's that I don't *care*." You should care. I once took a course in JBoss administration that included a section on tuning garbage collection. It convinced me that GC is a bad idea. It may make it easier for programmers but it shifts the burden for efficient memory management into a domain where it doesn't belong. Effective memory management is very sensitive to the design of the application. Shifting it to a generalized facility that runs garbage collection based on the short, medium or long duration of objects is an administrative nightmare. Someone writes an app that goes against the rules configured into the GC and the performance of the whole server suffers. So now we have training classes for Java programmers on how the GC works so they can write more efficient applications. How is that different from teaching C++ programmers a few simple code design disciplines (encapsulation, proper use of smart pointers, etc.) to minimize memory management problems in their code?

"Adult stem cells aren't pluripotent." Don't bet the family jewels on it. http://www.nature.com/nature/journal/v456/n7220/full/nature07404.html

Rob Pike worked at AT&T Bell Labs in 1982. I remember watching his demonstration of the Blit. I was using a production version in my work 10 years later. (I was not an early adopter. I got one when every one else seemed to already have them.) I guess this is another example of the way AT&T "stifled innovation". :-) http://mobile.slashdot.org/story/11/03/29/1437239/Ma-Bell-Stifled-Innovation-ATampT-May-Do-the-Same

Anyone who says Ma Bell stifled innovation didn't work there. (Of course ;-)

I used to work for AT&T Bell Labs. There was plenty of good research, development and innovation going on then. Yes, it took a lot of money because real innovation is not cheap and more competition only guarantees cheap products and services, not (necessarily) innovation. Yes, they were heavily regulated (even after 1984). Anyone who says Ma Bell stifled innovation did work there. Today's at&t is a result of the merger of some of the Regional Bell Operating Company's (RBOCs) that were spun off when the U.S. government broke up AT&T in 1984 (so are Verizon and Qwest). All their innovation came from what is now long gone and/or sold to Alcatel. There hasn't been much innovation for at&t to stifle, so the argument against the acquisition is really only about its effect on the cost to the consumer.

This is a very good point. The point about training users to use OpenOffice seems moot when you consider that it might be easier for people to switch to it from Office 03 to Office 2010. You have training overhead either way. Not only this but Go OpenOffice (http://go-oo.org/) and soon LibreOffice claim to offer better spreadsheet compatibility with MS Office. Plans to eliminate Java may help performance issues. If things continue to improve we might soon reach a tipping point where switching to LibreOffice from MS Office will be worth the cost savings and more and more companies will be doing it.

But is MS SQL Server any better? I heard a MySQL rep from Sun comment shortly after the Oracle deal to acquire Sun was announced that it might be Oracle's intent to use MySQL to compete with SQL Server. I don't see any less reason to believe his statements than yours above on the face of it. Innobase, the developer of the InnoDB transactional storage engine for MySQL, was acquired by Oracle some time before Oracle acquired Sun. Maybe Oracle has a higher opinion of MySQL than you do.

Seems like more of a reason for Oracle to keep MySQL going than Java.

I once took a course in JBoss administration that included a section on tuning garbage collection. It convinced me that GC is a bad idea. It may make it easier for programmers but it shifts the burden for efficient memory management into a domain where it doesn't belong. Effective memory management is very sensitive to the design of the application. Shifting it to a generalized facility that runs garbage collection based on the short, medium or long duration of objects is an administrative nightmare. Someone writes an app that goes against the rules configured into the GC and the performance of the whole server suffers. So now we have training classes for Java programmers on how the GC works so they can write more efficient programs. How is that different from teaching C++ programmers a few simple code design disciplines (encapsulation, proper use of auto_ptr, etc.) to minimize memory management problems in their code?