It's a reference to an old TV show, "The Six Million Dollar Man". Lee Majors played Steve Austin, who got terribly injured. The Powers That Be rebuilt him as a bionic wonder able to do things no mere flesh-and-blood human can do. The doll^H^H^H^Htelve-inch action figure had a vamera view-finder type of thing that you looked through his head to use (his bionic vision), and a cool karate chop action arm.
1. Pragmatically? If your kids know HTTP and HTTPS are blocked and that thousands of otherwise off-limits boobie pics are available through NNTP using Outlook Express, what makes you think they won't look there? How about when they use Tor to get porn, or SSH tunnel their HTTP through the school geek's BSD box?
2. You are suggesting a technical solution for a social problem. If you can't install Cybersitter or NetNanny or such, how do you stop porn on your kid's cell phone via Edge/GPRS? Via SMS? How about when your system is locked down and your kid runs next door to visit the neighbor -- and his system that's not locked down? How about when your kid brings home a Penthouse from school? Will the firewall stop that? What about when you think your kid and the neighbor are on the net, but they're actually _having_ sex instead of looking at it?
The metadata idea has already been mentioned several times in this thread. PICS ratings and such work when they're used. They're not a substitute for parenting either.
Using technology to restrict what a child does is not parenting. It's useful to have cabinet locks for small children, cell phones to call older children, etc. It's true you can't always be right beside your kids, but changing the way the Internet works isn't going to change that. There really is no technological substitute, though, for teaching kids right behavior. There is no substitute for discouraging wrong behavior. There is no substitute for spending time with your kids, for checking in on your kids and catching your kids in the act when they do misbehave. There is no substitute for discipline when the kids have done wrong.
Your ideas about what's right for kids and what's wrong for kids of certain ages may be very different from someone else's, but the fact that the kids need to be taught what you believe is right and wrong and that there are consequences for their behavior is pretty much indisputable. You may not like belting, spanking, room confinement, or whatever, but there needs to be some way to reach the child, and not just alter their environment within your four walls. The world is not within your four walls, and the Internet is a connection to the world. If you want to shelter your kids from the world, don't let them on the Net without supervision.
Okay, so HTTP porn on port 40001. Now, FTP porn on 40002? Gopher for porn on 40003? NNTP for porn on 40004? SMTP porn on 40005? SSL HTTP porn on 40006? 8-bit telnet with zmodem porn on 40007?
See the problem? Ports are for services. Porn is not a service, it's content of a service.
Maybe we should set hijack the Content-disposition header and set it to "Content-disposition: nasty". Sure, it'll break attachments, at least as far as there's overlap between attachments and porn, but who cares? Your children are safe from your lack of supervision while the rest of us work around your concerns.
When I was a child, I was told what to do and what not to do. If my parents weren't in the room with me the entire time, they checked in on me often enough that they'd catch me doing things I wasn't supposed to do, or at least make me reasonably afraid they would. When I got caught, I lost access to things like computers. When I was in real trouble, I got the belt.
Maybe that's what the Internet needs -- it needs parents who discipline their children for doing things children shouldn't do.
Come on everybody, we've got to stop the proliferation of unsupervised, spoiled, undisciplined children! Think of the Internet!
Also, Franklin, Commodore, TI, and Atari had systems with some form of OS in the ROM. Some Franklin systems had something called F-DOS in ROM which I think was mostly a ripoff of AppleDOS.
Notice that these examples are not modern hacks to try it out at home, but all commercially shipped systems from the late 1970s to early 1990s.
AMD and Intel still have documentation on DOS in ROM for embedded systems on their websites, and AMD even recommends Datalight's solution.
Oh, there's no doubt there's some serious issues with the numbers and how they're calculated. An industry standard for minimum number of units tested and minimum number of hours tested would be nice. At least disclosure of the testing conditions should be a minimum.
I'd like to see the industry do it without getting government involved. A simple law that clearly states that the manufacturers must describe the testing procedure in order to use the number for marketing would be great if the industry doesn't do it on its own. Of course, in the US there's never a simple law passed on one topic because it's good for the people. They always play politics with pork barrel projects, social issues, and such that they roll into one huge bill. Upon voting for or against said bill, one politician will get attacked in that it's bad for the environment while another will get attacked that it's bad for fiscal policy. One will defend his vote because it's good for small business and the other because it's good for "the poor" or "the arts", both without specifying who that is and how it actually helps. So yeah, it'd be nice if the industry could come up with a standard.
It would be mathematically equal, but I'm not sure it'd be equally _valid_. Given the initial defects and the possibility of misdesign causing heat-related losses or such, some stretch of time is really necessary. Testing 5 million for one hour proves little more than that the expected life is longer than one hour. Testing 200,000 for 25 hours would likely, despite the smaller but still sizable sample size, mean much more. Testing 20,000 at 250 hours would likely mean more still.
5,000 units at 1000 hours (41 days) or 10,000 units at 500 hours seems much more likely. After all, why make hundreds of thousands of something you're not sure are going to work at all?
The license with the best name for explaining the attitude based on name of license alone I think is the Creative Commons "Attribution-Share Alike" license.
Let's say you work hard to produce something useful to me. You wish that I could use it and make changes useful to myself so long as I "pay it forward". So when I distribute works based on yours, I need to pass on the right to make changes to others. That's the GPL and similar licenses.
Let's say then that you worked hard to produce something you want to share with me, but you don't mind if I distribute copies with modifications without allowing other people to make further modifications. That's the BSD and similar licenses.
Let's say you want to be the only one who ever benefits from your hard work. Then you just don't release it and noone can make use of it but you.
If you want other people to use your product but not make changes, you release it as a sealed box and forbid people from modifying it. That's most commercial software licenses. It's odd that I can change my car or change my computer, but that something supposedly binds me from changing my software. Sure, I might void my warranty on my car or my computer by making changes the vendor can't support, but I'm not going to be hauled in front of a judge by Ford or Dell for making changes. Software companies seem to think they are above this type of right for their customers.
Anyway, there are lots of options for distributing your work. If it's your work, you get to decide how to distribute it. Microsoft doesn't think that having choice in how to distribute software is a bad thing. They were charging for software in a closed-source setting when the norm in other parts of the industry was to provide sources and much software was freely traded among user groups and published by hobbyist magazines. They exercised their freedom to choose licensing terms, but they claim that choice in licenses is bad. It seems SCO has a fast and loose interpretation of what they are even allowed to determine licensing terms for.;-) These companies aren't really about limiting choice in software licensing. They just want to make sure they have every advantage in pulling money into their coffers -- fair advantage or not.
The sooner we can help people understand that sourcing less expensive infrastructure to keep business costs down is very capitalist and that Microsoft et al who say it's important to spend more money (?!) running a business to be a capitalist are just blowing smoke, the sooner we'll be able to quit advocating so much and get back to developing.
If it's been rewritten from the original code, then it's still copyrighted by the University. There's no reason to doubt that MS either would leave the copyright notice in place or do a full rewrite. I believe a lot of negative things about MS, but I don't think they're boldly ripping off a TCP/IP stack.
I'm just saying there's a good reason for it to be compatible -- because it was at least at first derived from one of the original implementations of TCP/IP. This is a great example of BSD-style code doing a lot of good. Linux uses it. Windows uses it. OS X probably uses it since that's derived largely from the BSDs. And the world can communicate better because the code was BSD licensed.
I'm a big fan of the GPL (version 2 anyway), of BSD, the Mozilla License, the Artistic License, the MIT license, etc. They are all pretty good at accomplishing what they are intended to accomplish. In this case, the BSD license was used to allow everyone to have a working TCP/IP stack whether open or closed source, and that's a good thing. OTOH, I'm glad certain projects are GPLed, because when everything becomes a one-way street with open source code, you get the kind of fragmentation we had with Unix in the 1980s.
The TCP/IP stack in windows is fairly standard and compatible because it's mostly copyright the Regents of the University of California at Berkeley. Given that, it should be perfectly compatible, and not just fairly. "Embrace and extend" is a business strategy at Microsoft, you see.
No, it's not the competition. Not entirely at least. It's also not much about the net neutrality. Those things help prices, but not so much the speeds available. It's more about the population density. The regulations on communications in the US don't help either. Distance is the enemy when it comes to high-speed signals. The more land you have between people, the higher your fiber, wire, or radio costs for the same signal. The cost is not necessarily linear, either.
Japan's overall population density is 337/km or 873/sq mi. The US is 31/km or 80/sq mi.
It's much more expensive to wire Iowa for broadband for three million people than it is to wire Orange County, California to reach about the same number of people. Think of the cost of fiber, and of the signal attenuation on copper wire for sections that aren't optical. Part of every phone bill in the US has traditionally been to subsidize dialup in areas that make no economic sense to service on their own. The reason 56k analog modems were limited to 53k was that the line voltages needed for 56k were considered unreliable and possibly dangerous on the US's aging and sometimes very long phone circuits. It's not uncommon for a telephone office in lesser populated areas of the US to offer service to a ten to fifteen mile radius. I bet there's more than one CO in Yokohama, Japan.
Yokohoma Japan, Orange County California, and the state of Iowa all have around 3-4 million people.
Population densities for comparison:
Tokyo: between 5600 and 5900 people per square kilometer by most reports (about 14700/sq mi I figure) Nakano-ku: ~20,000/sq km (over 50k/sq mi) Yokohama: around 8,200/sq km (around 21k/sq mi by my own conversion) Orange County: 1,392/km (3,606/mi) Iowa: 20.22/sq km (52.4/sq mi)
And now the five largest cities in the US (as of 1990): New York City: 23,705/sq mi Los Angeles: 7,427/sq mi Chicago: 12,252/sq mi Houston TX: 3,020/sq mi Philadelphia: 11,736/sq mi
There are about 15 cities in the US with over 10,000/sq mi population densities.
Now, per 3 million people, it's clearly most feasible and affordable to provide a high-speed connection across 169 square miles (437.35 km) of Yokohama compared to 9498 square miles (2,455 km) of Orange County or 56,272 sq mi (145,743 km) of Iowa.
One quarter of the US population live in what the Census Bureau considers rural areas. About 30% of Bronx County NY, 23% of Kings County NY, and 25% of Philadelphia County PA lived in poverty in 2004. These are not statistics which lead to major infrastructure projects nationwide being quick and cheap to undertake.
One could make all sorts of points about New York and New Jersey getting much faster connections than the rest of the US because they have much higher population densities than most states. However, Ma Bell pretty much has a mandate for that not to happen by her services. No one else seems to have stepped up and done it in those areas either. With the regulations stating who can do what with a wire, and the nature of negotiating licenses, rights of way, etc being so cumbersome it's no surprise that the people who don't already have cable runs don't want to lay their own. Cable vs. Phone only goes so far, and increased competition won't necessarily help if the natural features of the population and the regulations regarding telecom keep people charging about the same prices even once there are more competitors. Companies will use customer service and advertising to retain and gain customers long before price cuts.
I can tell you anecdotally it's cheaper for me in a town of 45,000 people to get 6Mbps DSL than for a friend of mine twenty miles from me to get 256kbps wireless. Any ideas why he hasn't ponied up the money? Maybe it's beca
Thank you for posting the article text. Now I can rip it apart bit by bit without waiting for his server to come back. Please keep in mind I'm speaking to the article's author when I say 'you' after this point, because I'm replying to the article text.
1. We don't need to talk about a Linux company to compare it to Microsoft. We're not comparing a company to a company. We're comparing the products of many companies and individuals and the advantages and disadvantages of that vs. Microsoft's products. Many users of Linux don't depend fully on one company for updates, fixes, and support. So to say basically that because Microsoft ties us to one source for these things that means that we must fall into the same trap for other operating systems is narrow-minded at best.
2. Any reasonable IT person will tell you that security is a process and not a product. Having a more secure base to start with is part of that process. Having code review is often part of that process. Running programs that aren't meant to make system-wide changes as users not authorized to make system-wide changes is part of the process. Most Linux distributions do a better job of _supporting_ these processes, and since it's open and editable, can be made moreso by many parties.
3. The FSF is not a hippy nudist farm commune, and the GPL is not a "do your fair share" agreement. The GPL allows people who have a purpose of their own to take a working system and do just the work they need done to support a change to do so instea dof writing a whole system from scratch and duplicating that parts that already work the way they need. Meeting your own needs and giving a little back for others having given you that opportunity is not communism. It's smart in a capitalist marketplace to take the lowest-cost route to your goal. Companies buy pre-existing parts to make their products all the time. Disney takes fairy tales with no copyrights, then copyrights the new work based on it, then lobbies to get those copyrights extended. Richard Stallman and Linus Torvalds may have their differences, but they both want you to be able to use their work to do your work. They're not askign for the keys to your car. They just want you to treat their work a certain way if you choose to use it. They don't even care if you don't give out changes you make. They just want to make sure you give out the sources of any binaries you make from _their_ hard work.
4. Employees are not valued on the price of what they work on. They are valued on the amount of money they make the company and the rarity of their skills. The only reason a Ferrari mechanic makes more than a Chevy mechanic is that fewer people know how to work on Ferraris and that the shop owners are able to charge more because the end customer can find fewer competitors since fewer shops work on Ferraris. The goal of the Linux community is not to drive up costs at an employer. It is to do the exact opposite. It's a freely available system which is meant to lower barriers, not raise them. The fact that it does the job of commercial Unix so well for so much less and has built so large a base of trained and experienced administrators and developers that the labor rates have dropped is a positive thing. It means Linux actually has a lower TCO, which is a good thing in a capitalist society. It's a point Microsoft tries to claim. Doing more business with fewer employees who need training that is easier to get is a goal of all good capitalist companies.
5. Linux is open entirely. The minds of some of its proponents are not. Please do not confuse the issue. Blind IT advocacy is bad, but some IT advocacy done with care is a wonderful thing. If Windows doesn't serve a department's needs, the IT department needs to make that clear to the people writing the check. If Linux doesn't meet the needs of a particular project, IT needs to advocate against Linux on that project. If something makes your job much easier and much less stressful while saving your company money, you should always support it vociferously.
You're being humorous, and it's pretty fuinny. There's a point here just waiting to be made, though.
Since it's LinuxBIOS we're talking about, it could be a thin client which also has the option to boot off any drive the system is capable of using for boot. So while you can't fit the Windows installation into the BIOS flash, you could have a well-featured small Linux in the BIOS which then boots into Windows, OS/2, FreeBSD, NetBSD, other BSDs, Darwin, Solaris, a full Linux installation, or anything else that runs on the PCs LinuxBIOS can be the BIOS for.
PCs used to come with DOS in ROM, and QNX has had kernel+GUI+other stuff in this kind of space for years although I've never seen QNX on the BIOS flash. It's cool to see someone doing it with LinuxBIOS. EPOC/Symbian, WinCE/PocketPC, Palm, etc all have GUIs, too. Maybe something like this could lead Linux to be truly competitive in that kind of market eventually. I had Debian Small running on my Psion 5mx, which was really cool. Still, even having a Sempron, Celeron, Geode, or C7 board with no disk and no Compactflash that gives me a small, power-efficient smart terminal to stick in the living room or kitchen would be great.
There's actually one reason. It's difficult to keep anything as thin as tinfoil in the right shape for very long due to terrestrial weather. A good tinfoil dish may work really well until the first 50km per hour wind, but good luck past that. Now if you had a wooden or plastic bowl of the right shape lined with tinfoil...
The market is more efficient if the barriers to entry are lowered, period.
If the guy down the street gets more business with magic markers and poster board than the guy up the street does with fancy, expensive signs, that's business. The guy with the more expensive signs either finds a way to deal with the disparity or eliminates it.
If company A has expensive skilled craftsmen and company B has just as good of quality and lower total costs with robots, that's tough for company A. Company B's just making a financially smart investment.
If company A pays Microsoft for their OS and company B does the same amount of work just as effectively on an OS they didn't pay for, that's great for company B.
If company A needs lots of changes from Microsoft and has to pay millions of dollars for them while company B went with something more open and can get the same changes for ten thousand dollars, then company B made the smart investment.
Capitalism is about bringing money in and lowering costs, not about looking for ways to give more of your money to vendors than is necessary. The whole idea that OSS is anti-capitalist is just more FUD.
If you think it takes no soul and nothing special in the heart to do one of the world's most dangerous, dirtiest, hardest, and most thankless jobs because of love for one's country, its ideals, and its people then you don't understand what being a real warrior is about. SEALs, Rangers, and the like are away from home for months and years. They dedicate themselves to mastery of very difficult skills. They must be very intelligent, and must be conditioned to think, move, and act quickly and decisively. Whether or not you are a Christian, any artist should understand what Jesus said about love. The common translation is usually phrased such as, "There is no greater love than this, that a man would give his life for his friends."
While shooting straight and the ability to "blow stuff up" are mere skills, mere skills are what the discussion was about. A programmer can be a designer, in general, as easily as a designer can be a programmer. Either one can be an artist. If you can't see the art in a well-thought and well-implemented program then you have little concept of programming are are being too focused on what you consider to be art.
As for heart and soul as applied to military personnel, give me Washington, Grant, Roosevelt, Churchill, Ivan Sidorenko, James Stockdale, John McCain, Lawrence Joel, Niles Harris, and Pat Tillman as men of heart and soul before most artists. Some people write poetry or paint pictures about love, hardship, heartbreak, loss, and courage. Some people live life so as to inspire that art. I believe it takes more of both to be the inspiration.
I am dubious of the claims made in TF Summary. I'm just trying to frame those claims in a more concrete way.
I'm pretty certain Cisco and Juniper have no problems keeping up. If the big bandwidth providers have a problem keeping up, it's certainly because they're using too little equipment. If they can't afford more equipment, then it's poor network design or undercharging for services.
It's the undercharging for services bandwagon they want everyone on, of course. Any way to make a perceived shortage without actually increasing their costs is a profit windfall. It's like oil being priced higher for perceived instability in the Middle East. Yes, tensions with Iran, for example, can actually hurt supply prices. But the cost to the consumer is much higher than the extra expenses to the oil companies. That's why oil companies have been recording record profits. Now I guess it's Ma Bell's turn to try.
When he's concerned about bandwidth demand outstripping computing power, that's not a fiber count problem. That's a router problem. He's saying the routers aren't gaining capacity to route packets as quickly as the number of packets to route is rising.
No amount of extra fiber will help if the routers can't keep up. Setting up more routers in the same interconnect centers will bring either bigger routing tables or higher latencies depending on how they're connected to one another. Setting up more interconnects which are more geographically dispersed and which route more directly between the endpoints will help, but that's a very expensive option. New buildings in new areas with new fiber running to them and new employees to man them simply cannot be made into a small investment.
Mesh networks, P2P content distribution, caching at the local level, multicasting, and some other technical measures can all theoretically help, too. So can spreading out the data centers of the big media providers and routing their traffic more directly that way, but again centralization of data centers saves a lot of money.
If demand is really growing too fast to handle (I have my doubts about the sky actually falling) one of the best ways to assure that bandwidth demands are met is to slow the increase in demand. The quickest and easiest way to slow increase in demand for something is to raise its price. That's an ugly thought for all of us on the broadband price war gravy train, but it's basic economics. Let's hope for a technological solution (or a group of them) instead, if it's really a problem and not just hype to hit our wallets in the first place.
There's also a related title that is more web-specific, so there's an option. It's called the Non-Designer's Web Book. There are also font-specific, scan-and-print-specific, and application-specific books by Williams. She even has her own computer dictionary.
Slashdot Mirror
It's a reference to an old TV show, "The Six Million Dollar Man". Lee Majors played Steve Austin, who got terribly injured. The Powers That Be rebuilt him as a bionic wonder able to do things no mere flesh-and-blood human can do. The doll^H^H^H^Htelve-inch action figure had a vamera view-finder type of thing that you looked through his head to use (his bionic vision), and a cool karate chop action arm.
1. Pragmatically? If your kids know HTTP and HTTPS are blocked and that thousands of otherwise off-limits boobie pics are available through NNTP using Outlook Express, what makes you think they won't look there? How about when they use Tor to get porn, or SSH tunnel their HTTP through the school geek's BSD box?
2. You are suggesting a technical solution for a social problem. If you can't install Cybersitter or NetNanny or such, how do you stop porn on your kid's cell phone via Edge/GPRS? Via SMS? How about when your system is locked down and your kid runs next door to visit the neighbor -- and his system that's not locked down? How about when your kid brings home a Penthouse from school? Will the firewall stop that? What about when you think your kid and the neighbor are on the net, but they're actually _having_ sex instead of looking at it?
The metadata idea has already been mentioned several times in this thread. PICS ratings and such work when they're used. They're not a substitute for parenting either.
Using technology to restrict what a child does is not parenting. It's useful to have cabinet locks for small children, cell phones to call older children, etc. It's true you can't always be right beside your kids, but changing the way the Internet works isn't going to change that. There really is no technological substitute, though, for teaching kids right behavior. There is no substitute for discouraging wrong behavior. There is no substitute for spending time with your kids, for checking in on your kids and catching your kids in the act when they do misbehave. There is no substitute for discipline when the kids have done wrong.
Your ideas about what's right for kids and what's wrong for kids of certain ages may be very different from someone else's, but the fact that the kids need to be taught what you believe is right and wrong and that there are consequences for their behavior is pretty much indisputable. You may not like belting, spanking, room confinement, or whatever, but there needs to be some way to reach the child, and not just alter their environment within your four walls. The world is not within your four walls, and the Internet is a connection to the world. If you want to shelter your kids from the world, don't let them on the Net without supervision.
Okay, so HTTP porn on port 40001. Now, FTP porn on 40002? Gopher for porn on 40003? NNTP for porn on 40004? SMTP porn on 40005? SSL HTTP porn on 40006? 8-bit telnet with zmodem porn on 40007?
See the problem? Ports are for services. Porn is not a service, it's content of a service.
Maybe we should set hijack the Content-disposition header and set it to "Content-disposition: nasty". Sure, it'll break attachments, at least as far as there's overlap between attachments and porn, but who cares? Your children are safe from your lack of supervision while the rest of us work around your concerns.
When I was a child, I was told what to do and what not to do. If my parents weren't in the room with me the entire time, they checked in on me often enough that they'd catch me doing things I wasn't supposed to do, or at least make me reasonably afraid they would. When I got caught, I lost access to things like computers. When I was in real trouble, I got the belt.
Maybe that's what the Internet needs -- it needs parents who discipline their children for doing things children shouldn't do.
Come on everybody, we've got to stop the proliferation of unsupervised, spoiled, undisciplined children! Think of the Internet!
This counts on having empty TV frequencies. Rural areas have fewer TV stations to contend with, anyway. Double the pleasure, double the fun.
Nutria already gave some examples, but I've got more for you.
Tandy 1000, HP's 95LX (and 200LX) palmtop PC with DOS (the 200 had MS-DOS 5.0), the HP 1000CX DOS palmtop, some of the early IBM Aptivas, the HP model 110 line of desktops, the rather famous GRiDLite (my GRiD laptops all loaded DOS from hardrive -- always wanted a GRiDLite too though), the IBM EduQuest Model 30 and Model 40 (I have a few model 40s, but only one still boots -- into OS/2 Warp because I'm not using the on-chip DOS), the Sharp PC-5000 portable, the IBM PCJr, certain IBM PS/1 machines, the Tandy 2500 XL, and some others.
Also, Franklin, Commodore, TI, and Atari had systems with some form of OS in the ROM. Some Franklin systems had something called F-DOS in ROM which I think was mostly a ripoff of AppleDOS.
Notice that these examples are not modern hacks to try it out at home, but all commercially shipped systems from the late 1970s to early 1990s.
AMD and Intel still have documentation on DOS in ROM for embedded systems on their websites, and AMD even recommends Datalight's solution.
Oh, there's no doubt there's some serious issues with the numbers and how they're calculated. An industry standard for minimum number of units tested and minimum number of hours tested would be nice. At least disclosure of the testing conditions should be a minimum.
I'd like to see the industry do it without getting government involved. A simple law that clearly states that the manufacturers must describe the testing procedure in order to use the number for marketing would be great if the industry doesn't do it on its own. Of course, in the US there's never a simple law passed on one topic because it's good for the people. They always play politics with pork barrel projects, social issues, and such that they roll into one huge bill. Upon voting for or against said bill, one politician will get attacked in that it's bad for the environment while another will get attacked that it's bad for fiscal policy. One will defend his vote because it's good for small business and the other because it's good for "the poor" or "the arts", both without specifying who that is and how it actually helps. So yeah, it'd be nice if the industry could come up with a standard.
IANA product tester.
It would be mathematically equal, but I'm not sure it'd be equally _valid_. Given the initial defects and the possibility of misdesign causing heat-related losses or such, some stretch of time is really necessary. Testing 5 million for one hour proves little more than that the expected life is longer than one hour. Testing 200,000 for 25 hours would likely, despite the smaller but still sizable sample size, mean much more. Testing 20,000 at 250 hours would likely mean more still.
5,000 units at 1000 hours (41 days) or 10,000 units at 500 hours seems much more likely. After all, why make hundreds of thousands of something you're not sure are going to work at all?
The license with the best name for explaining the attitude based on name of license alone I think is the Creative Commons "Attribution-Share Alike" license.
;-) These companies aren't really about limiting choice in software licensing. They just want to make sure they have every advantage in pulling money into their coffers -- fair advantage or not.
Let's say you work hard to produce something useful to me. You wish that I could use it and make changes useful to myself so long as I "pay it forward". So when I distribute works based on yours, I need to pass on the right to make changes to others. That's the GPL and similar licenses.
Let's say then that you worked hard to produce something you want to share with me, but you don't mind if I distribute copies with modifications without allowing other people to make further modifications. That's the BSD and similar licenses.
Let's say you want to be the only one who ever benefits from your hard work. Then you just don't release it and noone can make use of it but you.
If you want other people to use your product but not make changes, you release it as a sealed box and forbid people from modifying it. That's most commercial software licenses. It's odd that I can change my car or change my computer, but that something supposedly binds me from changing my software. Sure, I might void my warranty on my car or my computer by making changes the vendor can't support, but I'm not going to be hauled in front of a judge by Ford or Dell for making changes. Software companies seem to think they are above this type of right for their customers.
Anyway, there are lots of options for distributing your work. If it's your work, you get to decide how to distribute it. Microsoft doesn't think that having choice in how to distribute software is a bad thing. They were charging for software in a closed-source setting when the norm in other parts of the industry was to provide sources and much software was freely traded among user groups and published by hobbyist magazines. They exercised their freedom to choose licensing terms, but they claim that choice in licenses is bad. It seems SCO has a fast and loose interpretation of what they are even allowed to determine licensing terms for.
The sooner we can help people understand that sourcing less expensive infrastructure to keep business costs down is very capitalist and that Microsoft et al who say it's important to spend more money (?!) running a business to be a capitalist are just blowing smoke, the sooner we'll be able to quit advocating so much and get back to developing.
If it's been rewritten from the original code, then it's still copyrighted by the University. There's no reason to doubt that MS either would leave the copyright notice in place or do a full rewrite. I believe a lot of negative things about MS, but I don't think they're boldly ripping off a TCP/IP stack.
I'm just saying there's a good reason for it to be compatible -- because it was at least at first derived from one of the original implementations of TCP/IP. This is a great example of BSD-style code doing a lot of good. Linux uses it. Windows uses it. OS X probably uses it since that's derived largely from the BSDs. And the world can communicate better because the code was BSD licensed.
I'm a big fan of the GPL (version 2 anyway), of BSD, the Mozilla License, the Artistic License, the MIT license, etc. They are all pretty good at accomplishing what they are intended to accomplish. In this case, the BSD license was used to allow everyone to have a working TCP/IP stack whether open or closed source, and that's a good thing. OTOH, I'm glad certain projects are GPLed, because when everything becomes a one-way street with open source code, you get the kind of fragmentation we had with Unix in the 1980s.
The TCP/IP stack in windows is fairly standard and compatible because it's mostly copyright the Regents of the University of California at Berkeley. Given that, it should be perfectly compatible, and not just fairly. "Embrace and extend" is a business strategy at Microsoft, you see.
No, it's not the competition. Not entirely at least. It's also not much about the net neutrality. Those things help prices, but not so much the speeds available. It's more about the population density. The regulations on communications in the US don't help either. Distance is the enemy when it comes to high-speed signals. The more land you have between people, the higher your fiber, wire, or radio costs for the same signal. The cost is not necessarily linear, either.
Japan's overall population density is 337/km or 873/sq mi. The US is 31/km or 80/sq mi.
It's much more expensive to wire Iowa for broadband for three million people than it is to wire Orange County, California to reach about the same number of people. Think of the cost of fiber, and of the signal attenuation on copper wire for sections that aren't optical. Part of every phone bill in the US has traditionally been to subsidize dialup in areas that make no economic sense to service on their own. The reason 56k analog modems were limited to 53k was that the line voltages needed for 56k were considered unreliable and possibly dangerous on the US's aging and sometimes very long phone circuits. It's not uncommon for a telephone office in lesser populated areas of the US to offer service to a ten to fifteen mile radius. I bet there's more than one CO in Yokohama, Japan.
Yokohoma Japan, Orange County California, and the state of Iowa all have around 3-4 million people.
Population densities for comparison:
Tokyo: between 5600 and 5900 people per square kilometer by most reports (about 14700/sq mi I figure)
Nakano-ku: ~20,000/sq km (over 50k/sq mi)
Yokohama: around 8,200/sq km (around 21k/sq mi by my own conversion)
Orange County: 1,392/km (3,606/mi)
Iowa: 20.22/sq km (52.4/sq mi)
And now the five largest cities in the US (as of 1990):
New York City: 23,705/sq mi
Los Angeles: 7,427/sq mi
Chicago: 12,252/sq mi
Houston TX: 3,020/sq mi
Philadelphia: 11,736/sq mi
There are about 15 cities in the US with over 10,000/sq mi population densities.
Now, per 3 million people, it's clearly most feasible and affordable to provide a high-speed connection across 169 square miles (437.35 km) of Yokohama compared to 9498 square miles (2,455 km) of Orange County or 56,272 sq mi (145,743 km) of Iowa.
One quarter of the US population live in what the Census Bureau considers rural areas. About 30% of Bronx County NY, 23% of Kings County NY, and 25% of Philadelphia County PA lived in poverty in 2004. These are not statistics which lead to major infrastructure projects nationwide being quick and cheap to undertake.
One could make all sorts of points about New York and New Jersey getting much faster connections than the rest of the US because they have much higher population densities than most states. However, Ma Bell pretty much has a mandate for that not to happen by her services. No one else seems to have stepped up and done it in those areas either. With the regulations stating who can do what with a wire, and the nature of negotiating licenses, rights of way, etc being so cumbersome it's no surprise that the people who don't already have cable runs don't want to lay their own. Cable vs. Phone only goes so far, and increased competition won't necessarily help if the natural features of the population and the regulations regarding telecom keep people charging about the same prices even once there are more competitors. Companies will use customer service and advertising to retain and gain customers long before price cuts.
I can tell you anecdotally it's cheaper for me in a town of 45,000 people to get 6Mbps DSL than for a friend of mine twenty miles from me to get 256kbps wireless. Any ideas why he hasn't ponied up the money? Maybe it's beca
Thank you for posting the article text. Now I can rip it apart bit by bit without waiting for his server to come back. Please keep in mind I'm speaking to the article's author when I say 'you' after this point, because I'm replying to the article text.
1. We don't need to talk about a Linux company to compare it to Microsoft. We're not comparing a company to a company. We're comparing the products of many companies and individuals and the advantages and disadvantages of that vs. Microsoft's products. Many users of Linux don't depend fully on one company for updates, fixes, and support. So to say basically that because Microsoft ties us to one source for these things that means that we must fall into the same trap for other operating systems is narrow-minded at best.
2. Any reasonable IT person will tell you that security is a process and not a product. Having a more secure base to start with is part of that process. Having code review is often part of that process. Running programs that aren't meant to make system-wide changes as users not authorized to make system-wide changes is part of the process. Most Linux distributions do a better job of _supporting_ these processes, and since it's open and editable, can be made moreso by many parties.
3. The FSF is not a hippy nudist farm commune, and the GPL is not a "do your fair share" agreement. The GPL allows people who have a purpose of their own to take a working system and do just the work they need done to support a change to do so instea dof writing a whole system from scratch and duplicating that parts that already work the way they need. Meeting your own needs and giving a little back for others having given you that opportunity is not communism. It's smart in a capitalist marketplace to take the lowest-cost route to your goal. Companies buy pre-existing parts to make their products all the time. Disney takes fairy tales with no copyrights, then copyrights the new work based on it, then lobbies to get those copyrights extended. Richard Stallman and Linus Torvalds may have their differences, but they both want you to be able to use their work to do your work. They're not askign for the keys to your car. They just want you to treat their work a certain way if you choose to use it. They don't even care if you don't give out changes you make. They just want to make sure you give out the sources of any binaries you make from _their_ hard work.
4. Employees are not valued on the price of what they work on. They are valued on the amount of money they make the company and the rarity of their skills. The only reason a Ferrari mechanic makes more than a Chevy mechanic is that fewer people know how to work on Ferraris and that the shop owners are able to charge more because the end customer can find fewer competitors since fewer shops work on Ferraris. The goal of the Linux community is not to drive up costs at an employer. It is to do the exact opposite. It's a freely available system which is meant to lower barriers, not raise them. The fact that it does the job of commercial Unix so well for so much less and has built so large a base of trained and experienced administrators and developers that the labor rates have dropped is a positive thing. It means Linux actually has a lower TCO, which is a good thing in a capitalist society. It's a point Microsoft tries to claim. Doing more business with fewer employees who need training that is easier to get is a goal of all good capitalist companies.
5. Linux is open entirely. The minds of some of its proponents are not. Please do not confuse the issue. Blind IT advocacy is bad, but some IT advocacy done with care is a wonderful thing. If Windows doesn't serve a department's needs, the IT department needs to make that clear to the people writing the check. If Linux doesn't meet the needs of a particular project, IT needs to advocate against Linux on that project. If something makes your job much easier and much less stressful while saving your company money, you should always support it vociferously.
You're being humorous, and it's pretty fuinny. There's a point here just waiting to be made, though.
Since it's LinuxBIOS we're talking about, it could be a thin client which also has the option to boot off any drive the system is capable of using for boot. So while you can't fit the Windows installation into the BIOS flash, you could have a well-featured small Linux in the BIOS which then boots into Windows, OS/2, FreeBSD, NetBSD, other BSDs, Darwin, Solaris, a full Linux installation, or anything else that runs on the PCs LinuxBIOS can be the BIOS for.
PCs used to come with DOS in ROM, and QNX has had kernel+GUI+other stuff in this kind of space for years although I've never seen QNX on the BIOS flash. It's cool to see someone doing it with LinuxBIOS. EPOC/Symbian, WinCE/PocketPC, Palm, etc all have GUIs, too. Maybe something like this could lead Linux to be truly competitive in that kind of market eventually. I had Debian Small running on my Psion 5mx, which was really cool. Still, even having a Sempron, Celeron, Geode, or C7 board with no disk and no Compactflash that gives me a small, power-efficient smart terminal to stick in the living room or kitchen would be great.
Open Watcom isn't GPL. It's running on Windows and OS/2. BSD and Linux are in the works.
Engadget does one with a DirecTV round dish.
This place has a bunch of links, including the above. One is a bicircle which looks kinda cool.
Or buy a stainless steel wok. Sure, it might not reflect the signal quite the same, but they rust much, much slower.
There's actually one reason. It's difficult to keep anything as thin as tinfoil in the right shape for very long due to terrestrial weather. A good tinfoil dish may work really well until the first 50km per hour wind, but good luck past that. Now if you had a wooden or plastic bowl of the right shape lined with tinfoil...
So, when will the casinos install these to help track down those Konami slot machines with the subliminal messages?
The market is more efficient if the barriers to entry are lowered, period.
If the guy down the street gets more business with magic markers and poster board than the guy up the street does with fancy, expensive signs, that's business. The guy with the more expensive signs either finds a way to deal with the disparity or eliminates it.
If company A has expensive skilled craftsmen and company B has just as good of quality and lower total costs with robots, that's tough for company A. Company B's just making a financially smart investment.
If company A pays Microsoft for their OS and company B does the same amount of work just as effectively on an OS they didn't pay for, that's great for company B.
If company A needs lots of changes from Microsoft and has to pay millions of dollars for them while company B went with something more open and can get the same changes for ten thousand dollars, then company B made the smart investment.
Capitalism is about bringing money in and lowering costs, not about looking for ways to give more of your money to vendors than is necessary. The whole idea that OSS is anti-capitalist is just more FUD.
Good catch. It seems it was a grandson and not a son.
If you think it takes no soul and nothing special in the heart to do one of the world's most dangerous, dirtiest, hardest, and most thankless jobs because of love for one's country, its ideals, and its people then you don't understand what being a real warrior is about. SEALs, Rangers, and the like are away from home for months and years. They dedicate themselves to mastery of very difficult skills. They must be very intelligent, and must be conditioned to think, move, and act quickly and decisively. Whether or not you are a Christian, any artist should understand what Jesus said about love. The common translation is usually phrased such as, "There is no greater love than this, that a man would give his life for his friends."
While shooting straight and the ability to "blow stuff up" are mere skills, mere skills are what the discussion was about. A programmer can be a designer, in general, as easily as a designer can be a programmer. Either one can be an artist. If you can't see the art in a well-thought and well-implemented program then you have little concept of programming are are being too focused on what you consider to be art.
As for heart and soul as applied to military personnel, give me Washington, Grant, Roosevelt, Churchill, Ivan Sidorenko, James Stockdale, John McCain, Lawrence Joel, Niles Harris, and Pat Tillman as men of heart and soul before most artists. Some people write poetry or paint pictures about love, hardship, heartbreak, loss, and courage. Some people live life so as to inspire that art. I believe it takes more of both to be the inspiration.
I am dubious of the claims made in TF Summary. I'm just trying to frame those claims in a more concrete way.
I'm pretty certain Cisco and Juniper have no problems keeping up. If the big bandwidth providers have a problem keeping up, it's certainly because they're using too little equipment. If they can't afford more equipment, then it's poor network design or undercharging for services.
It's the undercharging for services bandwagon they want everyone on, of course. Any way to make a perceived shortage without actually increasing their costs is a profit windfall. It's like oil being priced higher for perceived instability in the Middle East. Yes, tensions with Iran, for example, can actually hurt supply prices. But the cost to the consumer is much higher than the extra expenses to the oil companies. That's why oil companies have been recording record profits. Now I guess it's Ma Bell's turn to try.
When he's concerned about bandwidth demand outstripping computing power, that's not a fiber count problem. That's a router problem. He's saying the routers aren't gaining capacity to route packets as quickly as the number of packets to route is rising.
No amount of extra fiber will help if the routers can't keep up. Setting up more routers in the same interconnect centers will bring either bigger routing tables or higher latencies depending on how they're connected to one another. Setting up more interconnects which are more geographically dispersed and which route more directly between the endpoints will help, but that's a very expensive option. New buildings in new areas with new fiber running to them and new employees to man them simply cannot be made into a small investment.
Mesh networks, P2P content distribution, caching at the local level, multicasting, and some other technical measures can all theoretically help, too. So can spreading out the data centers of the big media providers and routing their traffic more directly that way, but again centralization of data centers saves a lot of money.
If demand is really growing too fast to handle (I have my doubts about the sky actually falling) one of the best ways to assure that bandwidth demands are met is to slow the increase in demand. The quickest and easiest way to slow increase in demand for something is to raise its price. That's an ugly thought for all of us on the broadband price war gravy train, but it's basic economics. Let's hope for a technological solution (or a group of them) instead, if it's really a problem and not just hype to hit our wallets in the first place.
There's also a related title that is more web-specific, so there's an option. It's called the Non-Designer's Web Book. There are also font-specific, scan-and-print-specific, and application-specific books by Williams. She even has her own computer dictionary.
Hmmm... it looks to me like the review already had a link to Amazon, too. So 363636-20 is not just a spammer but a liar too.