Pure computer science doesn't pay enough to justify a college education in it. People are getting out of it as fast as possible because they flat out, won't accept the standard of living it provides.
When you enter the industry you'll find all your managers are in their 20's and all the programmers are in their 50's. Recent graduates either get into management as fast as possible or quit.
Programmers in the business for 30 years still live in dumpy apartments and have virtually no goals in life because they're so damn poor. No government program is going to change the situation. People can't be made to work 30 years to live in a dumpy apartment when other jobs provide so much more.
The culture in US is based on selling. People in the front office, interacting with the customers, making the deals are always going to be valued more than the people in the back room.
You can elect as many democrats as you want and tax yourself as much as you want. Your country will still value front office workers more than programmers.
I think those only work if the content is unencrypted. HDMI content isn't always encrypted you know. HDCP content would come out garbled and you'd be $50 poorer.
Look back on 1999 when decss was written, anyone could play DVD's on anything, and there were a lot of high profile court cases with celebrity programmers testifying.
Since then, HD content has been distributed in encrypted WMV and many other encrypted formats and no-one has even tried to crack those.
Part of the problem is there aren't any hackers anymore. The only guys who even care about the method of decrypting intellectual property are in India. Everyone else just wants to pay the fee and watch it.
Part of the problem is the culture is no longer obsessed with hacking PC's to play video. No-one wants to be able to play video on their PC or have complete control of intellectual property. The modern culture is to buy a commercial products to do that, pay for the intellectual property, and use it as it is intended.
The experience of playing intellectual property on the newest gadgets has made up for content protection. Having a snazzy media player or display has proven much more valuable to the modern hacker than defeating the encryption of the content.
The content industry has also proven that even if you can break encryption, you'll never be able to traffic it over the internet. They'll shut you down and arrest you so why bother with the encryption.
1999: That's the last time any hacker would gain access to an insanely great media upgrade or even care to.
After a 1 month Summer in Calif* and several years of declining temperatures, we feel the climate is cooling down from particulates more than it's heating up from CO2. Everyone knows sulfur from China's factories is reducing the amount of energy reaching Calif*. The sunsets today are a lot redder than they used to be.
If being too dumb to synthesize useful sentences qualifies you as psychopathic, then there's probably an alien race vastly smarter than us which thinks every human is psychopathic. Most politicians rank above 13 on that quiz. George Bush is too dumb to get above 6.
Predicted when IBM hyped the cell processor a few months ago that you'd to better with an AMD cell processor once AMD hired IBM's managers. Sure enough, you'll do better with an AMD cell processor.
Ever wonder why Russians are always first in pushing the human body to the limit of its exposure to harsh environments? Russians also hold the record for longest time fixing nuclear reactors without radiation suits, longest time spent underwater in sunken submarines with no air, longest exposure to exploding rocket fuel, largest populations in the coldest climates.
Given that space has many times more radiation than Earth and financial limitations require Russians to train fewer astronauts for longer missions, the longest time in space seems no less like Russians refactoring their predicament into a bragging right than their submarine records.
Through the years of Mosaic dominating, then Netscape dominating, then IE dominating, then Mozilla starting over from scratch, then Firefox starting over from scratch, then Firefox getting some code, it's a lot like watching coal miners in the 60's.
No matter what happens in the world. No matter what problems the world has moved onto, there is always this club which eats, sleeps, and breaths web browsers. They insist that winning back the lost users in 1998 is the most important breakthrough, that it wasn't Mozilla rewrite #20 but this version. This is the version which is going to get back the users they lost to Microsoft in 1998.
Just like coal miners saying the future isn't in space, it's underground, these web browser programmers seem to be eternally in 1998, endlessly chasing after the web browser trophy while maybe the world isn't watching anymore.
If everyone who already owns a car got rid of it and replaced it with a hybrid car, the cars would cost $100,000, China would consume all the world's oil and fill the atmosphere with sulfur making the new cars, and Japan would invent something better than hybrid cars, leaving u.s. in prehistory again.
Most of the cost in a hybrid car is the petroleum and coal required to mine the iron, lithium, manganese, copper, aluminum, cast the iron, weld the steel, melt the plastics, synthesize the battery dielectrics, and paint it. Then in 5 years the Lithium Manganese battery has to be replaced.
So when you junk your current car and pay China $20,000 to build a hybrid car, you're consuming $20,000 of oil and coal and saying you're not because you're buying a finished product.
Unfortunately, despite the parallel universe over at ap.org, China counts just as much on the energy scoreboard as u.s.. whether you burn oil h.e.r.e. or pay China to burn it there.
Thought the best option would be a reusable lifting body mounted on top of a Delta IV heavy. It would be a space shuttle with the main engines and the cargo bay removed.
Capsules like the ones they want are really cramped and are already available through Russia.
The Discovery channel documentary in May showed one of the most oppressive beaurocracies ever concieved. NASA's slogan is "if you have an idea, save it for your recruiter". They have low end managers converting the input of every 5 engineers into nothing. The result was a lot of managers stating they solved the problems heroically, when they never solved anything at all or only approved the most hopeless, rediculous tile repair technique.
After that documentary, we figured the next flight would be just as defect ridden as the last one. Sure enough, the foam popped off, the tiles chipped, nothing changed.
The decline of western economies and hyper competition to get away from outsourcable jobs and into management has created an environment which can never hope to solve the kinds of problems you need to solve if you want to access space.
U.s. needs to let the pros handle space flight for a few years while it figures out how to encourage creativity instead of hide it on the other end of an undersea cable.
U.S.: have ideas, get fired. Asia: have ideas, get promoted.
The u.s. culture is extremely defensive nowadays. Managers fight for fewer and fewer opportunities, seeing any growth in their subordinates as a threat. American employees are pigeonholed into specific tasks. Their companies are literally at war with themselves.
Asian companies are in a dramatic offense position. Their employees are allowed to do whatever they want without being viewed as threats. An asian EE one day can be a software engineer the next day and a rocket engine designer the next day.
Nowadays the supercomputer contest is just a matter of who can buy the most Opteron PC's and Cisco routers from Newegg and connect them. You might as well buy a few million DVD's from Best Buy and say you have the world's largest hard drive.
Eventually small countries will connect all the computers of their entire population with distributed clients and call that the world's largest supercomputer.
This business of entering a command, waiting a minute for zillions of nodes across a slow network to start, and waiting another minute for all the nodes to finish is hardly what supercomputing used to be.
It would be more interesting to see who does the most work with the least latency or who does the most work with the simplest programming model. Anyone can write a massively parallel program to utilize every Opteron in the world but a computer which can do the same work sequentially seems like a much bigger step forward.
Learned a long time ago to ignore any political opinions given by computer scientists because agree with them and they'll just say the opposite. So after the whining about companies banning replication of their video codecs and software, it's now bad for China to replicate MIPS compatability.
Nevertheless, compatability with the MIPS standard seems like the most trivial thing they could have copied. There are much harder problems to overcome in building a CPU than what spec to follow. The MIPS spec doesn't define how to mass produce very precise arrangements of semiconductor features for the least amount of money. It doesn't define how to dissipate heat and reduce power consumption.
Also, one day people are going to figure out that whatever China's government says, it's 10 years behind their current status. China's government says its economy is only growing at 5%. In reality it's growing at 10%. They say they won't finish the olympic stadium until 2008. It's finished now. They say 3 gorges won't become operational until 2010. It's operational now.
So what do you think the current state of Chinese technology is now that their government says they're at 1995 levels?
There's a reason they don't make semiconductors in US. Making semiconductors consumes enourmous amounts of resources and requires nasty chemicals like arsenic. So instead of releasing freon into US's atmosphere you release arsenic into China's water, make China burn vast amounts of fuel to bake silicon, and make China kill vast numbers of Panda bears to mine materials.
Most of the time it feels like HBS is in a different universe from the way things are done in u.s.. Theoretically they teach these principles to managers who then manage us they way they're taught but it's never shown.
Maybe HBS graduates go into fields other than technology and those fields are run like HBS teaches. Maybe technology leaders only come from Stanford and they have an opposite philosophy.
Globally, Harvard seems more in line with the way things are done in India.
It's astounding that with u.s.'s economy based mostly on selling, they're getting rid of mostly sales jobs. It's everywhere in the news: u.s. doesn't belong in implementation and design because the culture emphasizes selling. Getting rid of selling jobs is like making someone who doesn't like coleslaw eat coleslaw. Furthermore, sales jobs are the ones that lead to the executive jobs but they're keeping the executive jobs.
Now one thing you learn after college a lot more than you learn in college is exactly how to differentiate between jobs. The real world isn't defined as much by the type of programming you do as much as the scope of your responsibility.
Resume readers don't care if you're a Windows programmer, a UNIX programmer, a hardware designer, or a secretary. They want to see if you're a programmer, project lead, project manager, marketing manager, director, etc.
Things like Google, open source, wiki have leveled the playing field to where it doesn't matter if you study hardware, windows, AS/400, or UNIX. These things can all be learned by anyone at any time. In modern companies the skills at any given level of responsibility are being learned on demand as they're needed. Hardware designers one day are being used as UNIX programmers the next day.
Todays differentiation is in how much responsibility you're capable of having. Most resumes are being divided into management, sales and programming and as far as we can tell from the 36 checkboxes, management is the place to be.
You're obviously taking user interface design at an American college because the amount of pure torture in checking off 20 checkboxes by hand and reading a tiny graph 1/10th the size of the page can only be explained by the sheer dominance in inferiority of that university system.
Now if you went to an Indian college you would have allowed us to compare all 20 options at once on a readable graph without having to kill ourselves checking boxes. If you can't make it work, outsource it.
SGI started vanishing in the 90's, around the same time their low end programmers left to become project managers at VA I.O.U. Then when they left VA I.O.U. to become directors at Google, VA I.O.U. disappeared. The real wealth is in the employees, not the companies.
If you fight wars in the middle east, you get killed. If you withdraw your army from the middle east, you get killed. If you protest wars in the middle east, you get killed.
At least in my silicon valley experience, in 2000 it was 5 programmers to every manager and we did real implementation. Today it's 1 programmer to every 5 managers. All the work is in developing specs for products and developing business relations with vendors, but not in implementing any product. The implementation of the products is done by Asian companies who pay fees to use our specs, so it's not officially outsourcing even though it is.
It's efficient allright and writing specs for products is much more politically correct than outsourcing the product development.
Whether you call it improved efficiency or outsourcing, improved efficiency is basically another way of saying outsourcing.
Good news: They want to reuse at least some of the immense investment made in the space shuttle and they want to build something bigger instead of smaller.
Bad news: The glamour of having a real spaceship is going to dissappear.
The stacked heavy lifter is going to be so expensive and so inflexible, only 1 launch is going to be possible per year, containing all the cargo for the entire year.
The delta IV would have been a better match for the CEV. We old timers just have a lot of bad memories of solid rockets.
The Atlas V, of course, depends on Russian engines. It won't meet the requirement of strategic access to space if Russia decides this female president stuff isn't for them.
Slashdot Mirror
Pure computer science doesn't pay enough to justify a college education in it. People are getting out of it as fast as possible because they flat out, won't accept the standard of living it provides.
When you enter the industry you'll find all your managers are in their 20's and all the programmers are in their 50's. Recent graduates either get into management as fast as possible or quit.
Programmers in the business for 30 years still live in dumpy apartments and have virtually no goals in life because they're so damn poor. No government program is going to change the situation. People can't be made to work 30 years to live in a dumpy apartment when other jobs provide so much more.
The culture in US is based on selling. People in the front office, interacting with the customers, making the deals are always going to be valued more than the people in the back room.
You can elect as many democrats as you want and tax yourself as much as you want. Your country will still value front office workers more than programmers.
I think those only work if the content is unencrypted. HDMI content isn't always encrypted you know. HDCP content would come out garbled and you'd be $50 poorer.
Look back on 1999 when decss was written, anyone could play DVD's on anything, and there were a lot of high profile court cases with celebrity programmers testifying.
Since then, HD content has been distributed in encrypted WMV and many other encrypted formats and no-one has even tried to crack those.
Part of the problem is there aren't any hackers anymore. The only guys who even care about the method of decrypting intellectual property are in India. Everyone else just wants to pay the fee and watch it.
Part of the problem is the culture is no longer obsessed with hacking PC's to play video. No-one wants to be able to play video on their PC or have complete control of intellectual property. The modern culture is to buy a commercial products to do that, pay for the intellectual property, and use it as it is intended.
The experience of playing intellectual property on the newest gadgets has made up for content protection. Having a snazzy media player or display has proven much more valuable to the modern hacker than defeating the encryption of the content.
The content industry has also proven that even if you can break encryption, you'll never be able to traffic it over the internet. They'll shut you down and arrest you so why bother with the encryption.
1999: That's the last time any hacker would gain access to an insanely great media upgrade or even care to.
After a 1 month Summer in Calif* and several years of declining temperatures, we feel the climate is cooling down from particulates more than it's heating up from CO2. Everyone knows sulfur from China's factories is reducing the amount of energy reaching Calif*. The sunsets today are a lot redder than they used to be.
If being too dumb to synthesize useful sentences qualifies you as psychopathic, then there's probably an alien race vastly smarter than us which thinks every human is psychopathic. Most politicians rank above 13 on that quiz. George Bush is too dumb to get above 6.
Predicted when IBM hyped the cell processor a few months ago that you'd to better with an AMD cell processor once AMD hired IBM's managers. Sure enough, you'll do better with an AMD cell processor.
Ever wonder why Russians are always first in pushing the human body to the limit of its exposure to harsh environments? Russians also hold the record for longest time fixing nuclear reactors without radiation suits, longest time spent underwater in sunken submarines with no air, longest exposure to exploding rocket fuel, largest populations in the coldest climates.
Given that space has many times more radiation than Earth and financial limitations require Russians to train fewer astronauts for longer missions, the longest time in space seems no less like Russians refactoring their predicament into a bragging right than their submarine records.
Through the years of Mosaic dominating, then Netscape dominating, then IE dominating, then Mozilla starting over from scratch, then Firefox starting over from scratch, then Firefox getting some code, it's a lot like watching coal miners in the 60's.
No matter what happens in the world. No matter what problems the world has moved onto, there is always this club which eats, sleeps, and breaths web browsers. They insist that winning back the lost users in 1998 is the most important breakthrough, that it wasn't Mozilla rewrite #20 but this version. This is the version which is going to get back the users they lost to Microsoft in 1998.
Just like coal miners saying the future isn't in space, it's underground, these web browser programmers seem to be eternally in 1998, endlessly chasing after the web browser trophy while maybe the world isn't watching anymore.
If everyone who already owns a car got rid of it and replaced it with a hybrid car, the cars would cost $100,000, China would consume all the world's oil and fill the atmosphere with sulfur making the new cars, and Japan would invent something better than hybrid cars, leaving u.s. in prehistory again.
Most of the cost in a hybrid car is the petroleum and coal required to mine the iron, lithium, manganese, copper, aluminum, cast the iron, weld the steel, melt the plastics, synthesize the battery dielectrics, and paint it. Then in 5 years the Lithium Manganese battery has to be replaced.
So when you junk your current car and pay China $20,000 to build a hybrid car, you're consuming $20,000 of oil and coal and saying you're not because you're buying a finished product.
Unfortunately, despite the parallel universe over at ap.org, China counts just as much on the energy scoreboard as u.s.. whether you burn oil h.e.r.e. or pay China to burn it there.
Thought the best option would be a reusable lifting body mounted on top of a Delta IV heavy. It would be a space shuttle with the main engines and the cargo bay removed.
Capsules like the ones they want are really cramped and are already available through Russia.
The Discovery channel documentary in May showed one of the most oppressive beaurocracies ever concieved. NASA's slogan is "if you have an idea, save it for your recruiter". They have low end managers converting the input of every 5 engineers into nothing. The result was a lot of managers stating they solved the problems heroically, when they never solved anything at all or only approved the most hopeless, rediculous tile repair technique.
After that documentary, we figured the next flight would be just as defect ridden as the last one. Sure enough, the foam popped off, the tiles chipped, nothing changed.
The decline of western economies and hyper competition to get away from outsourcable jobs and into management has created an environment which can never hope to solve the kinds of problems you need to solve if you want to access space.
U.s. needs to let the pros handle space flight for a few years while it figures out how to encourage creativity instead of hide it on the other end of an undersea cable.
U.S.: have ideas, get fired.
Asia: have ideas, get promoted.
The u.s. culture is extremely defensive nowadays. Managers fight for fewer and fewer opportunities, seeing any growth in their subordinates as a threat. American employees are pigeonholed into specific tasks. Their companies are literally at war with themselves.
Asian companies are in a dramatic offense position. Their employees are allowed to do whatever they want without being viewed as threats. An asian EE one day can be a software engineer the next day and a rocket engine designer the next day.
10,000,000,000% of U.S.'s software industry. 10% of India's software industry.
Nowadays the supercomputer contest is just a matter of who can buy the most Opteron PC's and Cisco routers from Newegg and connect them. You might as well buy a few million DVD's from Best Buy and say you have the world's largest hard drive.
Eventually small countries will connect all the computers of their entire population with distributed clients and call that the world's largest supercomputer.
This business of entering a command, waiting a minute for zillions of nodes across a slow network to start, and waiting another minute for all the nodes to finish is hardly what supercomputing used to be.
It would be more interesting to see who does the most work with the least latency or who does the most work with the simplest programming model. Anyone can write a massively parallel program to utilize every Opteron in the world but a computer which can do the same work sequentially seems like a much bigger step forward.
Learned a long time ago to ignore any political opinions given by computer scientists because agree with them and they'll just say the opposite. So after the whining about companies banning replication of their video codecs and software, it's now bad for China to replicate MIPS compatability.
Nevertheless, compatability with the MIPS standard seems like the most trivial thing they could have copied. There are much harder problems to overcome in building a CPU than what spec to follow. The MIPS spec doesn't define how to mass produce very precise arrangements of semiconductor features for the least amount of money. It doesn't define how to dissipate heat and reduce power consumption.
Also, one day people are going to figure out that whatever China's government says, it's 10 years behind their current status. China's government says its economy is only growing at 5%. In reality it's growing at 10%. They say they won't finish the olympic stadium until 2008. It's finished now. They say 3 gorges won't become operational until 2010. It's operational now.
So what do you think the current state of Chinese technology is now that their government says they're at 1995 levels?
There's a reason they don't make semiconductors in US. Making semiconductors consumes enourmous amounts of resources and requires nasty chemicals like arsenic. So instead of releasing freon into US's atmosphere you release arsenic into China's water, make China burn vast amounts of fuel to bake silicon, and make China kill vast numbers of Panda bears to mine materials.
Most of the time it feels like HBS is in a different universe from the way things are done in u.s.. Theoretically they teach these principles to managers who then manage us they way they're taught but it's never shown.
Maybe HBS graduates go into fields other than technology and those fields are run like HBS teaches. Maybe technology leaders only come from Stanford and they have an opposite philosophy.
Globally, Harvard seems more in line with the way things are done in India.
It's astounding that with u.s.'s economy based mostly on selling, they're getting rid of mostly sales jobs. It's everywhere in the news: u.s. doesn't belong in implementation and design because the culture emphasizes selling. Getting rid of selling jobs is like making someone who doesn't like coleslaw eat coleslaw. Furthermore, sales jobs are the ones that lead to the executive jobs but they're keeping the executive jobs.
It's about time there were man pages for C++ the way there are for C.
Now one thing you learn after college a lot more than you learn in college is exactly how to differentiate between jobs. The real world isn't defined as much by the type of programming you do as much as the scope of your responsibility.
Resume readers don't care if you're a Windows programmer, a UNIX programmer, a hardware designer, or a secretary. They want to see if you're a programmer, project lead, project manager, marketing manager, director, etc.
Things like Google, open source, wiki have leveled the playing field to where it doesn't matter if you study hardware, windows, AS/400, or UNIX. These things can all be learned by anyone at any time. In modern companies the skills at any given level of responsibility are being learned on demand as they're needed. Hardware designers one day are being used as UNIX programmers the next day.
Todays differentiation is in how much responsibility you're capable of having. Most resumes are being divided into management, sales and programming and as far as we can tell from the 36 checkboxes, management is the place to be.
You're obviously taking user interface design at an American college because the amount of pure torture in checking off 20 checkboxes by hand and reading a tiny graph 1/10th the size of the page can only be explained by the sheer dominance in inferiority of that university system.
Now if you went to an Indian college you would have allowed us to compare all 20 options at once on a readable graph without having to kill ourselves checking boxes. If you can't make it work, outsource it.
SGI started vanishing in the 90's, around the same time their low end programmers left to become project managers at VA I.O.U. Then when they left VA I.O.U. to become directors at Google, VA I.O.U. disappeared. The real wealth is in the employees, not the companies.
If you fight wars in the middle east, you get killed. If you withdraw your army from the middle east, you get killed. If you protest wars in the middle east, you get killed.
At least in my silicon valley experience, in 2000 it was 5 programmers to every manager and we did real implementation. Today it's 1 programmer to every 5 managers. All the work is in developing specs for products and developing business relations with vendors, but not in implementing any product. The implementation of the products is done by Asian companies who pay fees to use our specs, so it's not officially outsourcing even though it is.
It's efficient allright and writing specs for products is much more politically correct than outsourcing the product development.
Whether you call it improved efficiency or outsourcing, improved efficiency is basically another way of saying outsourcing.
Good news: They want to reuse at least some of the immense investment made in the space shuttle and they want to build something bigger instead of smaller.
Bad news: The glamour of having a real spaceship is going to dissappear.
The stacked heavy lifter is going to be so expensive and so inflexible, only 1 launch is going to be possible per year, containing all the cargo for the entire year.
The delta IV would have been a better match for the CEV. We old timers just have a lot of bad memories of solid rockets.
The Atlas V, of course, depends on Russian engines. It won't meet the requirement of strategic access to space if Russia decides this
female president stuff isn't for them.