Well, first, what do you mean by "computer engineering"? Do you want to design graphics boards, or set up data centers? If you're really going to do serious electronic design, you need both the degree and the experience. If you're mostly going to install big systems, the theory may not be that helpful. An MBA might be more useful.
Forty years ago, being an EE who designed serious hardware paid about as well as being a lawyer.
(The IEEE has numbers on this.) That is no longer the case. Hardware design also isn't as much fun as it used to be. You're constrained on all sides - size, power consumption, cost, and schedule. There's less opportunity to be clever. Most electronics design work is now done outside the US now, anyway. When the manufacturing goes, the design work follows a few years later. Aerospace is much smaller than it used to be, too.
If you really are interested in how it all works inside, though, go for it. Computer engineering is far less superficial than IT. You're less tied to whatever fad is up this month.
So if I drive 10,000 miles on a road covered with potholes and the wheels fall off, it is the car manu's fault? Didn't think so.
Wrong. There have been many lawsuits and recalls involving wheels falling off cars.
2000 Ford Focus --
NHTSA CAMPAIGN ID Number: 00V303000
Component: WHEELS:LUGS:NUTS:BOLTS
Potential Number of Units Affected: 203,700
Manufactured: 03/1999 - 05/2000
To install a retention cap to keep the left rear wheel and brake drum assembly from flying off the car.
"Attorney Fred Pritzker has obtained another large recovery in a car-truck accident case. Our law firm represented the family of a man killed by a truck wheel that had come off of the truck, rolled off a bridge and killed the driver of a car that was traveling on the road below."
If a wheel falls off a car and somebody gets hurt, there's a very high probability the manufacturer will be sued. Especially if that's been seen multiple times on that model. That's just not supposed to happen.
This militates against the argument that the "imprimatur" of a publisher always adds to a journal's legitimacy.
It sure does. Especially since Elseiver has explicitly made that argument. Here's an official Elsevier position paper on open access: "By introducing an author-pays model, Open Access risks undermining public trust in the integrity and quality of scientific publications that has been established over hundreds of years. The subscription model, where the users pay... ensures high quality, independent peer review and prevents commercial interests from influencing decisions to publish. This critical control measure would be rmeoved in a system where the author - or indeed his/her sponsoring institution - pays."
The claim that "the vendor doesn't know the environment in which the software will be used" is bogus. Car companies have no idea where you will drive your car, or on what kind of roads. They have a far worse problem than any software maker. Yet
they have to accept serious liability obligations.
Provided that this is implemented as a constraint on commerce, it's not an issue for free software when distributed for free. However, companies that package up free software and resell it (Red Hat, Novell, etc.) will face liability. That's as it should be.
Sell out, agree to stay with the purchased company for a short transition period only (six months or so), and don't agree to any noncompete longer than two years. Make sure that the buyout payment is independent of any employment arrangement. Get legal advice on tis.
Then, now that you have some money, do something else.
Why? Because you haven't faced the marketing problem yet. If you do it on your own, there's a good chance you'll have trouble selling the product. Especially during a recession. And getting money for marketing may be tough.
It's an OK movie. If it weren't "Star Trek", it would probably rank with The Chronicles of Riddick.
Annoyances:
Way too much lens flare and depth of focus manipulation. Even indoor scenes have lens flare. Somebody spent too much time pushing the buttons on the editor.
Somebody likes plumbing too much. Most of the interior scenes have vast amounts of piping and tankage. It looks like some of the shipboard shots were filmed in a modern brewery.
How did the Grand Canyon move to Iowa?
OK, the bad guys are attacking populated planets that are members of a military alliance by hovering in one place over the planet and lowering a drill? And nobody does anything about this? Even when they try it in populated areas? You'd think somebody might have something around that could fly and shoot, and with their planet being threatened, might use it.
If you thought close-range ship to ship marksmanship in Star Wars was bad, here it's worse. Nobody can hit consistently at point-blank range. It's 1880s gunnery technology. But they can latch onto an individual falling to the planet and beam them up.
Kirk's attitude wouldn't survive the first year at any known military academy. No matter who his father was.
Having escaped from a big ship under attack using a bunch of little shuttles, the crew would be POWs or dead. The shuttles can't fight and can't run.
Time travel. Bad time travel. The deus ex machina of bad SF.
Maybe someday there will be a David Weber SF movie, one that makes military sense. This isn't it. It's a mediocre space opera.
The State of California is doing some massive budget cuts. Santa Rosa is cutting so deep that they're turning off street lights in the middle of blocks to save money. Looks like we found some people who don't have much to do and can be laid off.
I would also like to hear how you explain MySpace's massive success
What massive success? Myspace made about $75 million per quarter at peak.
Their traffic peaked in Q1 2008, and is down 30% since then. Facebook
passed them in April 2008 and now has 3x their traffic. Myspace never made
enough of a blip in News Corp. earnings to show up as a line item.
Social networking sites have a life cycle like nightclubs, and it's short.
They start, if they're lucky they become cool, they grow, the losers move in, the
cool people move out, and they decline. Has-been social networking sites include
AOL, Geocities, EZboard, Nerve, Friendster, Orkut, and Tribe. Social networking
sites have to be valued like movies - they have to make money over
their run. They're not ongoing businesses. There's a long tail of trickling revenue after the peak, as with ongoing sales of DVDs of old movies. But the big money comes early if at all.
That's problem #1 with social networking sites. Problem #2 is that the demographic is terrible from an advertiser perspective. Remember, half of all clicks come from 20% of users, and that 20% buys almost nothing. That 20% of users is Myspace's demographic.
Myspace revenue comes mostly from their Google ads. Think about that for a moment. Myspace is a big site run by a bigger publisher with sizable ad-selling operations. Yet they're running Google ads, from which Google makes most of the money. If Murdoch could make online pay, they'd be selling their own ad space. The advertisers on Myspace are mostly either bottom-feeders (links to pages with more ads and similar junk) or small advertisers who haven't figured out how to opt out of having their ad appear there.
Re-engineering biological systems takes generations to debug. And a huge number of dud individuals during the development process. This is fine for tomato R&D, but generating a big supply of failed post-humans is going to be unpopular. Just extending the human lifespan is likely to take generations to debug. It takes a century to find out if something worked.
AIs and robots don't have that problem.
What I suspect is going to happen is that we're going to get good AIs and robots, but they won't be cheaper than people. Suppose that an AI smarter than humans can be built, but it's the size of a server farm. In that case, the form the "singularity" may take is not augmented humans, but augmented corporations. The basic problem with companies is that no one person has the whole picture. But a machine could. If this happens, the machines will be in charge, simply because the machines can communicate and organize better.
Incidentally, spam is way up in the last three weeks. Looking at a filter that sees mail from
about ten E-mail addresses, spam volume had been from 40 to 70 messages/day for most of
April. Around April 22, the spam level started to climb steadily, and now it's in the 180-200
messages/range. Is there a new botnet active?
If we want to approach this by regulation, it would help to make operating system companies financially liable for the damage caused to third parties by botnets able to attack their systems. Windows would become much more secure in a hurry. Remember, Windows is insecure mostly because Microsoft by default has a general policy of running anything that looks executable from any external source, from web browsers to USB ports to CDs and DVDs. Windows also lets installed software do just about anything. With liability for the manufacturer, systems would ship with most of that turned off.
This isn't that radical; automobile companies are routinely held liable for damage caused by design or manufacturing flaws. The radical thing is that software companies have managed to escape product liability.
Seriously. What assets do they have left that are worth selling? Patents? Software?
The way this works is that the Bankruptcy Trustee brings in an auction house. Assets like intellectual property and lawsuits are usually handled directly by the Trustee. The auction house handles the physical assets.
There's a whole food chain in Silicon Valley for disposing of defunct companies. Action Computer buys up many of the old PCs, the ones that work. Weird Stuff Warehouse buys up old networking gear and miscellaneous electronics. Consolidated Office Distributors buys much of the furniture (Their warehouse in San Jose looks like the one from Raiders of the Lost Ark, only bigger. That's where many of the Aeron chairs from the dot-com boom ended up.) There's a place in San Jose that buys steel shelving. Quickly, the office buildings and factories are cleaned out, cleaned up, and put on the market.
Most search systems do well if you can find the "magic word", the word or phrase which nearly uniquely describes what you're looking for. When you're searching for something which is described with common words, not terms of art, search engines usually don't do well. That's where to test Alpha, which supposedly has some "understanding" of its data.
Here's an example of something I was doing today. I'm sketching out a design for a special-purpose DC-DC converter, something I haven't done before. I'm looking at a data sheet for a transformer, and at the rules for describing a transformer to LTSpice, a circuit simulator. LTspice wants a value K, the "coupling coefficient". The data sheet for the transformer has various numbers about the transformer, but not the coupling coefficient. How do I calculate the coupling coefficient?
It turns out that the magic words for answering this question are "leakage inductance". Once you know that, you can find the Wikipedia entry that gives the necessary conversion formulas, and calculate the coupling coefficient. Until you find the magic words, though, it's tough. If you just go looking for "coupling coefficient" in Google, you're directed to theory papers. "Leakage inductance" is the number that appears in data sheets, because it's directly measurable.
If Alpha can answer questions like "How do I compute the coupling coefficient for a transformer given the data sheet parameters?",
it will be a nice capability.
Yet, if you put in the entire phrase "How do I compute the coupling coefficient for a transformer given the data sheet parameters?" as a query to Google, you get as a first result
a paper on how to model a transformer in LTSpice given data sheet information, which is exactly the right result to return. The answer is in that paper, and it's a good paper. Google does better at this than one might expect.
In AMPS, the cell phone technology being described, there's a 3600 baud control channel shared between all the phones in a cell. Text messages had to be crammed into that. Voice was analog FM, with the control channel telling the handsets which voice channel to use.
That's why SMS is so data-limited. The data channel was tiny.
Exercise: buy a newspaper and throw out all the sections that are 100% marketing. Entire sections, like Autos, Real Estate, and Wine (Wine?) go into the dumpster. The classified sections can go; that's all on-line, and on line it's searchable.
Of what's left, over half will still be pages that are all advertising. Throw out those pages.
About 15-20% of the original pages will be left.
Then throw out the pages than only have stories you already saw on Google News. Throw out the
stories that came from PR Newswire. Maybe 2 to 3% of the pages will be left. That's the "content". The whole paper could probably be condensed down to about six pages. In many cities, less.
Today's newspapers make spam look like an efficient data transmission medium.
I haven't watched YouTube since they broke the player for Firefox.
There's some bug which, for some versions of Firefox, causes YouTube's player to randomly switch between full size and 1/4 frame size, every few seconds. It's been reported, it's Firefox-only, it occurs with Firefox 2 and 3, it's related to newer versions of Flash, and it's not fixed.
Then YouTube put in some kind of common login system with Google logins, and because I had both a Google login and a YouTube login, it forces me to a "Link your YouTube & Google Accounts" page, which I don't want to do.
The US used to have a 45MW shipboard nuclear power plant on the USS Sturgis, a converted Liberty ship. It was used to power the Panama Canal locks during a period of low water at Gatun Dam, the usual power source. The U.S. Army had a whole range of small reactors running in remote locations from 1952 to the early 1970s. The main problem was that they never built enough of them to justify the support and training infrastructure required.
Implementing OpenGL efficiently isn't just a "big job" it's essentially the entire field of computer graphics hardware.
It's understood, though. And you can do it in sections. Start with an OpenGL implementation that does tessellation, geometry transforms, and fill in software. Build something that just does the fill part. (That's 1995 technology in PC graphics.) Then add the 4x4 multiplier and do the geometry on the board (that's 1998 technology on the PC, 1985 technology for SGI.). Once all that's working correctly and the read-back of the frame buffer matches the OpenGL spec for the tests, you can start to work on parallelism. (That's 2000 technology). Then comes programmable shaders and arbitrary computation on the graphics board, which gets hard.
It's a lot like Linux; a decade behind, but still useful.
There's not that much mystery about the things. Making a VGA emulator in an FPGA is no big deal.
If all you implemented was text mode and mode 13H, it would probably boot Linux.
Getting to a card that runs OpenGL is a big job, but not out of reach. The pipeline is well understood, and there are software implementations to look at.
As you get to later versions of Direct-X, it gets tougher, because Microsoft controls the documentation.
But the real problem is that you'll never get anything like the performance of current generation 3D boards with an FPGA. There aren't anywhere near enough gates. You need custom silicon.
This 2-axis rate gyro part came out in 2006. Analog Devices and Motorola have had comparable parts for years, but at a higher price.
It's only 2-axis. If they could do all 3 axes on a flat chip, that would be something. Usually, you need a second chip mounted vertically to the first one to get all three axes.
An elegant design is to use four MEMS gyros oriented along the axes of a tetrahedron. With that redundancy, you can detect faults. The Segway does that, for safety reasons.
The
Brother Quattro is a good example. Brother makes printers, machine tools, and sewing machines. All three lines come together in their computer-controlled embroidery machine. This one even has a built-in LCD panel. There's embroidery software, too, for designing embroidery work. (I had to learn about this once when an artist friend sent me a "company logo" file she's been given for a web site, but couldn't read. It turned out to be an embroidery machine control file. I found a program that could render it (stitch by stitch) and got a picture out.)
Computer-controlled embroidery machines have been around for 25 years or so, and before that, ones controlled by cams and chains.
(Yes, the admins running the servers are the government in this case.)
No, they're not. They're just a provider of communications services. Your phone company and ISP have no role in dispute resolution in deals you make with other parties.
Take a look at why Slashdot's pages load so slowly. There are several layers of "document.write(some javascript that loads something else)" just to load ads. The browser
can't do the loads concurrently; they all take place sequentially. Each "document.write" has to finish before the code in it can be run. Also, some of the CSS is being read from "s.fsdn.com", which is a rather slow server at times.
It can get worse. Try Rushmore Drive, the slowest-loading search engine home page known. This is a spinoff of Ask. There's enough
ad-related crap on that page that it takes 10-15 seconds to load. And this is without
any personalization or content-related overhead. It's all inept ad serving.
Those are both sites maintained by supposedly competent professionals. Sites where some
third-tier web programmer just cut and pasted code from other sites can be much worse.
We can probably deal with increases in Internet traffic just by improving ad-blocking.
Sure. Why not? If you can do work and exchange value there, you're going to need contracts.
Avatars are just remote connections of humans. Just as you can commit to a contract via
telegram, phone, fax, and web page, you can commit to one through an avatar.
It's possible to create works of art and music in Second Life. All the usual copyright issues apply. This will be more of an issue as the technology improves. We already have "Myspace bands." Bands have already played in Second Life, including a few big names.
In some circumstances, NPCs can create contracts. A vending machine, for example, creates a contract with you when you insert coins. If you don't get the product, you can sue the company operating the vending machine. The vending machine is an agent of the seller, acting within its authority. That's settled law (and a real issue for vending machines that sell things like iPods). The same legal concept applies to NPCs. This all derives from ancient law of master and servant.
Now of course with the multi-core CPUs they have to get these things right -- but back in the "old" days the hardware engineers could be more careless.
Multi-processor arbiter theory wasn't properly understood until the mid-1970s. Some early shared memory processors had race conditions with access to shared memory. There's a famous Intel patent on this. See Arbiter on Wikipedia for details. Ir turns out that you cannot resolve a race condition of that type in constant time. Sometimes you have to allow extra time for the arbiter to settle in one state or the other. Support to detect the need for that additional delay has to be designed in.
Engineers still get this wrong occasionally in I/O hardware. The people who design CPUs know all about it, but the lower-level people who design bus interfaces sometimes don't.
Slashdot Mirror
Well, first, what do you mean by "computer engineering"? Do you want to design graphics boards, or set up data centers? If you're really going to do serious electronic design, you need both the degree and the experience. If you're mostly going to install big systems, the theory may not be that helpful. An MBA might be more useful.
Forty years ago, being an EE who designed serious hardware paid about as well as being a lawyer. (The IEEE has numbers on this.) That is no longer the case. Hardware design also isn't as much fun as it used to be. You're constrained on all sides - size, power consumption, cost, and schedule. There's less opportunity to be clever. Most electronics design work is now done outside the US now, anyway. When the manufacturing goes, the design work follows a few years later. Aerospace is much smaller than it used to be, too.
If you really are interested in how it all works inside, though, go for it. Computer engineering is far less superficial than IT. You're less tied to whatever fad is up this month.
So if I drive 10,000 miles on a road covered with potholes and the wheels fall off, it is the car manu's fault? Didn't think so.
Wrong. There have been many lawsuits and recalls involving wheels falling off cars.
Component: WHEELS:LUGS:NUTS:BOLTS
Potential Number of Units Affected: 203,700
Manufactured: 03/1999 - 05/2000
To install a retention cap to keep the left rear wheel and brake drum assembly from flying off the car.
If a wheel falls off a car and somebody gets hurt, there's a very high probability the manufacturer will be sued. Especially if that's been seen multiple times on that model. That's just not supposed to happen.
This militates against the argument that the "imprimatur" of a publisher always adds to a journal's legitimacy.
It sure does. Especially since Elseiver has explicitly made that argument. Here's an official Elsevier position paper on open access: "By introducing an author-pays model, Open Access risks undermining public trust in the integrity and quality of scientific publications that has been established over hundreds of years. The subscription model, where the users pay ... ensures high quality, independent peer review and prevents commercial interests from influencing decisions to publish. This critical control measure would be rmeoved in a system where the author - or indeed his/her sponsoring institution - pays."
That gives the open access movement a big boost..
It's time for software to grow up.
I proposed this in 2000 as a penalty for Microsoft in their antitrust trial. That would have been a big step forward.
The claim that "the vendor doesn't know the environment in which the software will be used" is bogus. Car companies have no idea where you will drive your car, or on what kind of roads. They have a far worse problem than any software maker. Yet they have to accept serious liability obligations.
Provided that this is implemented as a constraint on commerce, it's not an issue for free software when distributed for free. However, companies that package up free software and resell it (Red Hat, Novell, etc.) will face liability. That's as it should be.
Sell out, agree to stay with the purchased company for a short transition period only (six months or so), and don't agree to any noncompete longer than two years. Make sure that the buyout payment is independent of any employment arrangement. Get legal advice on tis.
Then, now that you have some money, do something else.
Why? Because you haven't faced the marketing problem yet. If you do it on your own, there's a good chance you'll have trouble selling the product. Especially during a recession. And getting money for marketing may be tough.
It's an OK movie. If it weren't "Star Trek", it would probably rank with The Chronicles of Riddick.
Annoyances:
Maybe someday there will be a David Weber SF movie, one that makes military sense. This isn't it. It's a mediocre space opera.
The State of California is doing some massive budget cuts. Santa Rosa is cutting so deep that they're turning off street lights in the middle of blocks to save money. Looks like we found some people who don't have much to do and can be laid off.
I would also like to hear how you explain MySpace's massive success
What massive success? Myspace made about $75 million per quarter at peak. Their traffic peaked in Q1 2008, and is down 30% since then. Facebook passed them in April 2008 and now has 3x their traffic. Myspace never made enough of a blip in News Corp. earnings to show up as a line item.
Social networking sites have a life cycle like nightclubs, and it's short. They start, if they're lucky they become cool, they grow, the losers move in, the cool people move out, and they decline. Has-been social networking sites include AOL, Geocities, EZboard, Nerve, Friendster, Orkut, and Tribe. Social networking sites have to be valued like movies - they have to make money over their run. They're not ongoing businesses. There's a long tail of trickling revenue after the peak, as with ongoing sales of DVDs of old movies. But the big money comes early if at all.
That's problem #1 with social networking sites. Problem #2 is that the demographic is terrible from an advertiser perspective. Remember, half of all clicks come from 20% of users, and that 20% buys almost nothing. That 20% of users is Myspace's demographic.
Myspace revenue comes mostly from their Google ads. Think about that for a moment. Myspace is a big site run by a bigger publisher with sizable ad-selling operations. Yet they're running Google ads, from which Google makes most of the money. If Murdoch could make online pay, they'd be selling their own ad space. The advertisers on Myspace are mostly either bottom-feeders (links to pages with more ads and similar junk) or small advertisers who haven't figured out how to opt out of having their ad appear there.
This is going to take a while.
Re-engineering biological systems takes generations to debug. And a huge number of dud individuals during the development process. This is fine for tomato R&D, but generating a big supply of failed post-humans is going to be unpopular. Just extending the human lifespan is likely to take generations to debug. It takes a century to find out if something worked.
AIs and robots don't have that problem.
What I suspect is going to happen is that we're going to get good AIs and robots, but they won't be cheaper than people. Suppose that an AI smarter than humans can be built, but it's the size of a server farm. In that case, the form the "singularity" may take is not augmented humans, but augmented corporations. The basic problem with companies is that no one person has the whole picture. But a machine could. If this happens, the machines will be in charge, simply because the machines can communicate and organize better.
How did this retarded article get FP'd?
Mod parent up. Dumb idea.
Incidentally, spam is way up in the last three weeks. Looking at a filter that sees mail from about ten E-mail addresses, spam volume had been from 40 to 70 messages/day for most of April. Around April 22, the spam level started to climb steadily, and now it's in the 180-200 messages/range. Is there a new botnet active?
If we want to approach this by regulation, it would help to make operating system companies financially liable for the damage caused to third parties by botnets able to attack their systems. Windows would become much more secure in a hurry. Remember, Windows is insecure mostly because Microsoft by default has a general policy of running anything that looks executable from any external source, from web browsers to USB ports to CDs and DVDs. Windows also lets installed software do just about anything. With liability for the manufacturer, systems would ship with most of that turned off.
This isn't that radical; automobile companies are routinely held liable for damage caused by design or manufacturing flaws. The radical thing is that software companies have managed to escape product liability.
Seriously. What assets do they have left that are worth selling? Patents? Software?
The way this works is that the Bankruptcy Trustee brings in an auction house. Assets like intellectual property and lawsuits are usually handled directly by the Trustee. The auction house handles the physical assets.
There's a whole food chain in Silicon Valley for disposing of defunct companies. Action Computer buys up many of the old PCs, the ones that work. Weird Stuff Warehouse buys up old networking gear and miscellaneous electronics. Consolidated Office Distributors buys much of the furniture (Their warehouse in San Jose looks like the one from Raiders of the Lost Ark, only bigger. That's where many of the Aeron chairs from the dot-com boom ended up.) There's a place in San Jose that buys steel shelving. Quickly, the office buildings and factories are cleaned out, cleaned up, and put on the market.
To advertise a "skill contest" and not pay off on the scheduled date is usually a violation of the Deceptive Mail Prevention and Enforcement Act. Not good, Google.
Most search systems do well if you can find the "magic word", the word or phrase which nearly uniquely describes what you're looking for. When you're searching for something which is described with common words, not terms of art, search engines usually don't do well. That's where to test Alpha, which supposedly has some "understanding" of its data.
Here's an example of something I was doing today. I'm sketching out a design for a special-purpose DC-DC converter, something I haven't done before. I'm looking at a data sheet for a transformer, and at the rules for describing a transformer to LTSpice, a circuit simulator. LTspice wants a value K, the "coupling coefficient". The data sheet for the transformer has various numbers about the transformer, but not the coupling coefficient. How do I calculate the coupling coefficient?
It turns out that the magic words for answering this question are "leakage inductance". Once you know that, you can find the Wikipedia entry that gives the necessary conversion formulas, and calculate the coupling coefficient. Until you find the magic words, though, it's tough. If you just go looking for "coupling coefficient" in Google, you're directed to theory papers. "Leakage inductance" is the number that appears in data sheets, because it's directly measurable.
If Alpha can answer questions like "How do I compute the coupling coefficient for a transformer given the data sheet parameters?", it will be a nice capability.
Yet, if you put in the entire phrase "How do I compute the coupling coefficient for a transformer given the data sheet parameters?" as a query to Google, you get as a first result a paper on how to model a transformer in LTSpice given data sheet information, which is exactly the right result to return. The answer is in that paper, and it's a good paper. Google does better at this than one might expect.
In AMPS, the cell phone technology being described, there's a 3600 baud control channel shared between all the phones in a cell. Text messages had to be crammed into that. Voice was analog FM, with the control channel telling the handsets which voice channel to use.
That's why SMS is so data-limited. The data channel was tiny.
Exercise: buy a newspaper and throw out all the sections that are 100% marketing. Entire sections, like Autos, Real Estate, and Wine (Wine?) go into the dumpster. The classified sections can go; that's all on-line, and on line it's searchable.
Of what's left, over half will still be pages that are all advertising. Throw out those pages. About 15-20% of the original pages will be left.
Then throw out the pages than only have stories you already saw on Google News. Throw out the stories that came from PR Newswire. Maybe 2 to 3% of the pages will be left. That's the "content". The whole paper could probably be condensed down to about six pages. In many cities, less.
Today's newspapers make spam look like an efficient data transmission medium.
I haven't watched YouTube since they broke the player for Firefox.
There's some bug which, for some versions of Firefox, causes YouTube's player to randomly switch between full size and 1/4 frame size, every few seconds. It's been reported, it's Firefox-only, it occurs with Firefox 2 and 3, it's related to newer versions of Flash, and it's not fixed.
Then YouTube put in some kind of common login system with Google logins, and because I had both a Google login and a YouTube login, it forces me to a "Link your YouTube & Google Accounts" page, which I don't want to do.
The US used to have a 45MW shipboard nuclear power plant on the USS Sturgis, a converted Liberty ship. It was used to power the Panama Canal locks during a period of low water at Gatun Dam, the usual power source. The U.S. Army had a whole range of small reactors running in remote locations from 1952 to the early 1970s. The main problem was that they never built enough of them to justify the support and training infrastructure required.
Implementing OpenGL efficiently isn't just a "big job" it's essentially the entire field of computer graphics hardware.
It's understood, though. And you can do it in sections. Start with an OpenGL implementation that does tessellation, geometry transforms, and fill in software. Build something that just does the fill part. (That's 1995 technology in PC graphics.) Then add the 4x4 multiplier and do the geometry on the board (that's 1998 technology on the PC, 1985 technology for SGI.). Once all that's working correctly and the read-back of the frame buffer matches the OpenGL spec for the tests, you can start to work on parallelism. (That's 2000 technology). Then comes programmable shaders and arbitrary computation on the graphics board, which gets hard.
It's a lot like Linux; a decade behind, but still useful.
There's not that much mystery about the things. Making a VGA emulator in an FPGA is no big deal. If all you implemented was text mode and mode 13H, it would probably boot Linux. Getting to a card that runs OpenGL is a big job, but not out of reach. The pipeline is well understood, and there are software implementations to look at. As you get to later versions of Direct-X, it gets tougher, because Microsoft controls the documentation.
But the real problem is that you'll never get anything like the performance of current generation 3D boards with an FPGA. There aren't anywhere near enough gates. You need custom silicon.
This 2-axis rate gyro part came out in 2006. Analog Devices and Motorola have had comparable parts for years, but at a higher price.
It's only 2-axis. If they could do all 3 axes on a flat chip, that would be something. Usually, you need a second chip mounted vertically to the first one to get all three axes.
An elegant design is to use four MEMS gyros oriented along the axes of a tetrahedron. With that redundancy, you can detect faults. The Segway does that, for safety reasons.
The Brother Quattro is a good example. Brother makes printers, machine tools, and sewing machines. All three lines come together in their computer-controlled embroidery machine. This one even has a built-in LCD panel. There's embroidery software, too, for designing embroidery work. (I had to learn about this once when an artist friend sent me a "company logo" file she's been given for a web site, but couldn't read. It turned out to be an embroidery machine control file. I found a program that could render it (stitch by stitch) and got a picture out.)
Computer-controlled embroidery machines have been around for 25 years or so, and before that, ones controlled by cams and chains.
(Yes, the admins running the servers are the government in this case.)
No, they're not. They're just a provider of communications services. Your phone company and ISP have no role in dispute resolution in deals you make with other parties.
Take a look at why Slashdot's pages load so slowly. There are several layers of "document.write(some javascript that loads something else)" just to load ads. The browser can't do the loads concurrently; they all take place sequentially. Each "document.write" has to finish before the code in it can be run. Also, some of the CSS is being read from "s.fsdn.com", which is a rather slow server at times.
It can get worse. Try Rushmore Drive, the slowest-loading search engine home page known. This is a spinoff of Ask. There's enough ad-related crap on that page that it takes 10-15 seconds to load. And this is without any personalization or content-related overhead. It's all inept ad serving.
Those are both sites maintained by supposedly competent professionals. Sites where some third-tier web programmer just cut and pasted code from other sites can be much worse.
We can probably deal with increases in Internet traffic just by improving ad-blocking.
Sure. Why not? If you can do work and exchange value there, you're going to need contracts. Avatars are just remote connections of humans. Just as you can commit to a contract via telegram, phone, fax, and web page, you can commit to one through an avatar.
It's possible to create works of art and music in Second Life. All the usual copyright issues apply. This will be more of an issue as the technology improves. We already have "Myspace bands." Bands have already played in Second Life, including a few big names.
In some circumstances, NPCs can create contracts. A vending machine, for example, creates a contract with you when you insert coins. If you don't get the product, you can sue the company operating the vending machine. The vending machine is an agent of the seller, acting within its authority. That's settled law (and a real issue for vending machines that sell things like iPods). The same legal concept applies to NPCs. This all derives from ancient law of master and servant.
Now of course with the multi-core CPUs they have to get these things right -- but back in the "old" days the hardware engineers could be more careless.
Multi-processor arbiter theory wasn't properly understood until the mid-1970s. Some early shared memory processors had race conditions with access to shared memory. There's a famous Intel patent on this. See Arbiter on Wikipedia for details. Ir turns out that you cannot resolve a race condition of that type in constant time. Sometimes you have to allow extra time for the arbiter to settle in one state or the other. Support to detect the need for that additional delay has to be designed in.
Engineers still get this wrong occasionally in I/O hardware. The people who design CPUs know all about it, but the lower-level people who design bus interfaces sometimes don't.