Yes, Dylan is right. Everyone knows that as a technology becomes more and more ubiquitous, important and mature, the more critical it is to know the details of this technology. This is why 50 years after mass electrification of our cities, almost 90 percent of people were employed as electricians. Electricity is very important, and anyone who didn't understand electric circuits was completely marginalized and forced into obscure, low-paid professions' such as business, medicine or law.
After all, it's very unlikely that we would see any real advances in productivity in the deployment of new technology. Thus if we deploy 10 times as many computers in the future, naturally, the number of jobs involved in programming, maintaining and using these computers will also be multiplied by 10. If not more.
As for those secretaries who foolishly fail to learn the low-level workings of their machines, shame on them. These people are idiots! They are fooled into thinking that they can produce adequate letters and documents with such eye-candy laden programs such as Microsoft Word. How dumb is that? Naturally these people should put all their efforts into learning how to use their machines properly - that is, without all those stupid icons and buttons and menus. I feel sorry for anyone who wasted their time learning irrelevant skills such as English composition or business communication as opposed to the all-important details of what's really going on on their computers.
Finally, I think we should all thank Dylan for having partially inspired' the Matrix Revolutions. In addition to his perceptive analysis of the skills that are critical to our future, he has made a huge contribution to the world of film.
Yours in breathless admiration,
Just some feedback on this article.
I beg to differ, most people shouldn't have to understand the workings of what's going on "under the hood" of a computer no more than a car, cd player or coffee machine. The whole aim of developing better user interfaces is to make it as intuitive as possible to use a computer.
"Getting inside" the computer is a good idea for anyoe who wants to program computers or is curious in general, but there should not be such a requirement. A drivers license does not require intimate knowledge of combustion engines, transmission systems nor the inner workings of an ABS braking system. Why should computers be any different?
Feel free to "print" this little blurb, name and web address may be published.
I want to warn everybody to install the Linux nforce Drivers Version 1.0-0261 with a KERNEL 2.4.19 (e.g. SuSE Linux 8.1). On our webserver (based on MSI K7N2GL - nForce2) we installed a kernel patch and therefore we had to reinstall the nforce drivers. After this the system was always having 20% system charge without doing anything. Our webserver was delivering instead of an average of 3 Mbit/s only a maximum of 600 KBit/s. Performance was crazy slow but everything worked. A change to unpatched Kernel with new drivers brought the same result. Patched Kernel with old drivers did not work at all. After reinstalling the old unpatched kernel (with the security holes in it) and reinstall of Drivers Version 1.0-0256 everything became stable and performing again.
Maybe someone has more experience, but I can only warn to use them. Since everything works with the new drivers and only performance is bad, maybe many users do not even remark it, because they do not use their system under heavy load (like our webserver).
Someone ought to lock this moron up just for the safety and well being of society. There is no excusing piracy, it is not free advertising as this imbecile suggests, and somebody ought to pee in your beer just for your running this article.
Regarding "AMD benchmarks Opteron against Itanium, Xeons, PIIIs", those Itanium2 numbers are very impressive considering the thing is only clocked to 1.5GHZ. It would be interesting know what, if any, ideas were taken from the DEC Alpha to make the Itanium2 as good as it is. The only fly in the ointment is that for the price of one Itanium, you could probably buy four Opterons.
Though I can see situations where going with the fewest and most powerful processors is advanageous. For example, many enterprise server packages are priced on a per cpu basis - as in thousands per cpu. And assuming floating point performance is the most important (as it often is on high end apps), having twice as many Opterons would probably be about the same (for floating point) or a little faster, from a performance perspective. But from a cost perpective, you could end up getting killed.
On the other hand, if you're running a Linux server with something like Apache, then the Opteron solution is probably better since in that setup, having extra CPUs doesn't cost you thousands more in licencing costs.
After having looked at those numbers, I think the best thing Intel could do to compete with the Opteron would be to offer more affordable versions of the Itanium.
After seeing those numbers, I certainly hope the Itanium doesn't die. It now looks like a decent processor.
What is also interesting is that it looks like if Compaq/HP were to spend a little money on the Alpha and get the clock speeds up to at least 2 GHZ, that processor would be competitive with the Opteron, from a performance perspective anyways - and probably cost the same or less than an Itanium. And who knows what performance we would have seen from the EV8 Alpha - probably a true Itanium killer.
Sign up for INQbot – a weekly roundup of the best from the INQ