NAT works by creating a translation table: The private network has lots of IP address, each of which are talking on a relative few ports. The external side has 1 IP address talking on lots of ports. Each PrivateIP+Port combination on the inside needs one PublicID+Port combination on the outside. Since there are only 16 bits of port and a 10.* network has 24 bits of private IP space, this system would collapse under moderate load.
When analysts sour on commercial WiFi, it probably means that there is no way for a major corporation to make money off of it. That doesn't mean that WiFi can't be successful.
Broadband is becoming fairly cheap. A decent commercial quality DSL line probably runs about $200-$300 a month. Add a $1,000 one time fee for some high end access points and you can WiFi enable your coffee shop or book store. At those rates it can become feasible to offer WiFi for free, just to differentiate your business from competitors.
For that matter, public parks in cities can get WiFi services to encourage their use. Bryant Park in Manhattan (6th Ave and 42nd St) started all sorts of programs to get people using the park, displacing the drug dealers. WiFi is just another service that can be offered in this vein.
WiFi will be successful because it is "too cheap to meter". Only equipment manufacturers will make money on it, though.
Part of the problem is that Dynastar454 chose to target "North American" with his joke. "North American" isn't funny to anyone. Americans (US-ians) and Canadians have a long tradition of poking fun at each other. Because we are so similar culturally, it is amusing to describe each other in ways that would indicate a real rivalry.
If Dynastar had used "America" as his target, all the Canadians would have laughed. I swapped in Canadian and all the Americans laughed. We could have swapped in France and the whole world would hae laughed.
All three [copyright, patent] were designed to force the properties of materials on information so it could be owned and exchanged in a capitalistic society just like other goods.
Many slashdotters rant that ownership of information is immoral, but I doubt that many really consider the alternatives.
Once the printing press was invented/commercialized by Guttenburg, It suddenly became orders of magnitude cheaper to produce books. In the disorder that followed - the disorder present whenever technology leaps ahead of law - publishers found new material to publish by simply walking into the shops of other publishers and buying copies of their books. Publisher Foo would pay an author to create content, then publisher Bar would buy one copy of Foo's book and copy the contents. Bar would unercut the price, since he didn't have to make back the money paid to the author.
In the long run, there is no motive to pay authors. Authors need to eat in order to write. Therefore the government stepped in and created copyrights.
With patents, the issue is slightly different. If the invention was the product, there is little to protect the inventor from copying, hence little opportunity for the inventor to benefit.
If the invention, however, was not the product, but a process or procedure to make the product, it could be protected by keeping it a trade secret. Trade secrets are just that - secrets. Therefore, there is no way for another inventor to use one invention as the starting point for another. This is bad for innovation.
Patent law was created to give inventors limited monopolies on their ideas in exchange for publishing them for other inventors to use.
Patent and copyright law is useful and beneficial for society. The problem today is two-fold: protection is granted to overly broad "innovation" and protection is granted for increasingly long periods of time. I think we all intuitively agree that "1-click" purchasing shouldn't be patentable. But a truly new and innovative scheduling algorithm might be worthy. And it seems that exclusive periods are much too long for many inventions.
A few ideas I have for reform...
Increase the standards for getting a patent - currently almost anything can be patented, the burden of proof that something doesn't deserve a patent falls to accused infringers.
Reduce the exclusive period in accordance with the type of invention. Pharmaceuticals require huge investment, so drug companies need 15-20 years to make back their investment. Software patents should run out after no more than five years.
Reduce the exclusive period for copyrighted material.
Require that copyrighted material also be published. Therefore source code would need to be published in order to be protected.
Trademarks and Servicemarks are not so much about intellectual property, but about allowing consumers to reliably identify producers. Chrysler should not be able to name a car "Honda" and trick consumers into buying their product.
If for some reason Canada was wiped of the face of the earth tomorrow people would still be able live elsewhere.
Sure there would probably be major setbacks. It might take another five years to get the NHL to the point were we are now but things would eventually get back to normal.
So RMS's statement that Canada is no longer essential is true.
I vacationed to Canada and it seemed OK. But if I _had_ to I could learn to vacation in Europe, or Asia even. I know both Europe and Asia are working on getting there[sic] people to speak English.
If Canada went away tomorrow it would be a real shame (understatement of the year) but it would not be the end of the world.
I have to agree that this is a book that should be on everyone's shelf.
The very fact that both vi and emacs support regular expressions must mean they are a best-in-breed tool, because if there was a way for those two communities to disagree, they would have done it.
I love the fact that I can use the same expressions with grep, sed, vim, Perl, and Java. that being said, however, the critics are who warn that regex can be over used are correct: regex's are difficult to debug and to maintain, so don't go overboard.
I'm not a Mac user, although if I had the extra $ around I wouldn't mind buying one.
I always chuckle when I see someone spend extra for a 100 Mb or 1 Gb ethernet card, then plug their PC directly into a cable modem that does 1.5 Mb. I want to tell them to go to their local PC builder and get an ethernet card from the spares pile.
I thought that a big part of the excitement over the new Mac products was the high performance bus? Raw processor speed is always only part of the story. If the Mac can move data amongst ram, disk, processor, and video card much faster than the Intel/AMD systems, the net result is a more powerful PC.
I run a 633 celeron based eMachines. I pulled the OEM hard drive and replaced it with a 7200 RPM model. My user experience performance doubled - although the bechmarks wouldn't have changed. This is reasonable, since on an average desktop machine, the user's waits at boot and when loading an application from disk. Both those processes ran much faster with the high speed hard drive.
Just a note, besides web browsing/word processing I develop in Java. Tools such as Ant do a very good job of incremental compiling so I never have to kick off a make script and wait.
My Company is a multinational where each of the national components ran somewhat independantly. I've been a developer/architect for a core financial system is the US territory. The system requires about 10 developers/dbas/data administrators to keep it running.
We are currently in the process of replacing the system with a new system for a cost rumored to be around two hundred million dollars to the US territory.
The logic goes like this...
The global management decides to buy an app. for the full organization.
The global mangement also decides that they are going to charge back the development and liscensing costs to all the territories regardless of whether they use the new system.
A bean counter in the US says "Why are we paying fifty million dollars for a system we aren't using?"
No one has the cajones to say, "Global has forced us to through away $50,000,000. Is throughing away another $150,000,000 a good idea?"
Of course, 1 year earlier we couldn't get the $5,000,000 to do a rewrite on a modern Unix platform.
it should be pointed out that both sub-Saharan africa and southern india developed their own advanced civilisations independently
Diamond does discuss sun-Saharan Africa. He acknowledges the development of civilization there based on sorghum and the ass. He claims that this culture was less advanced because it had fewer less protein rich staple foods and less capable draft animals. It also had less room to spread, so it dominated sub-Saharan Africa but couldn't break out of the continent.
The wheat/bean/horse/oxen culture of the fertile crescent had the geography to spread all the way into India and to Britain. Plus, having more protein, more animals, more area, this culture developed more. It still took thouasands of year before the fertile crescent people were able to enslave the sub-Sahran people.
As for Dravidian, it seems to have developed in the Indus Valley, where civilization also developed independently. I don't have specific knowledge, but I would suspect that fertile crescent culture mixed with Indus culture when the Indo-Iranian (Persian, Hindi, etc.) speakers moved into India around 1700 BC.
An important point to keep in this thread is that when we discuss culture in this context, we need to think about questions like "What did they grow? What animals did they use? What farming methods did they use?" I'm not sure if the Dravidian speakers of Southern India adopted the culture (by this limited measure) of the fertile crescent people while maintaining their own language. Perhaps the Indus Valley culture was had enough in common with the fertile crescent culture to make this an easy transition.
after independently inventing the wheel, they used it for children's toys exclusively.
It is a pretty consistent observation that lots of cultures invented the wheel, but only those that had access to high quality draft animals used it. Remember that the horse and other draft animals (oxen, donkey, etc.) were extinct in the new world until (re)introduced by the Europeans in 1492.
A great book on the subject is Guns, Germs, Steel: The Fate of Human Societies by Jared Diamond. Diamond argues that two dominant cultures have arisen - A Western culture that traces its roots to Fertile Crescent in modern day Iraq and the an Eastern culture that traces its roots to the Yellow River Valley. In both of these places nature and geography conspired to create a package of tools that allowed these cultures to spread.
Both these places had the following...
Naturally occuring staple foods - usually grains - that were easy to domesticate
Large wild animals that were easy to domesticate and useful as draft animals
Room to spread out while using the same tools
In contrast, the natives of the Americas had only a single staple grain - corn - and that one took thousands of years longer than wheat, barley, oats, and rice to domesticate and they had no draft animals.
As an added gotcha, when the American natives did manage to domesticate corn, there were barriers to spreading out. For instance, the people of Mexico - Aztec, Mayan, Toltec - would need to pack up and cross the American Southwestern deserts, then the great plains (which can't be farmed easily without steel plows), then the Appalachian mountains, before reaching readily farmable land in the Eastern USA. The Chinese and Middle Eastern peoples could spread all the way to Korea, India, North Africa, and Europe without hitting that much of a barrier.
Re:Buyouts (why MS or anyone hasn't done it yet)
on
My Visit to SCO
·
· Score: 1
MS probably believes they can use SCO as a pawn in a gambit. They pump some cash into SCO to keep the case going. If SCO wins, great for MS. If SCO drags the case out, creating FUD, great for MS. If SCO loses, oh well.
If MS buys SCO outright, however, they gain very little but put themselves at great risk. Assume that SCO's loses, and perhaps is even deemed to have made fraudulent claims. Then the multi-BILLION dollar countersuit is coming out of Bill's pocket. As it is, if SCO loses and is countersued - there is not much value to be siezed.
The worst outcomes for MS are really 1) SCO loses quickly 2) SCO is bought out by IBM, RedHat, or someone else and the suit is dropped.
The interesting twist is that, if they are really smart, SCO may be playing double agent. First they take a big pile of money from MS for licensing-wink wink-and make alot of noise. (It has to look good, so as to get more money from Bill). Then, at the last minute, they make a deal with IBM to trade their IP portfolio for a the pieces of IBM's consulting business that pertain to the "small fish" that form SCO's traditional customer base but that IBM is not quite comfortable with.
The net result would then be that IBM and SCO both shed portions of their business that they don't want, and use a big chunk of Bill's own money to compete with Bill's own server products. Not to mention that IBM has technology licensed to MS, which they'll watch like a hawk, ready to attack MS with lawyer Shock and Awe at the slightest provocation.
This is a plot worthy of Vince MacMahon!
Re:Why I'm Not Really Worried...
on
My Visit to SCO
·
· Score: 2, Funny
You've never worked with IBM apparently. They aren't immune to screw-ups.
Hold one end of the sinew in the palm of your hand and the other between your thumb and forefinger. Adjust the length so that the basket is in the exact middle.
Whirl it around your head. Release the sinew between your thumb and forefinger at exactly the right moment to cause the stone to fly off and hit your target.
As you might imagine, doing this accurately is quite difficult. The only people that typically use the sling are shepherds, since they have the three or four hours a day to spend practicing. As a comparison, a bowman only needs about 30 minutes of daily practice to be highly skilled.
Note that SCO has not made copyright or patent claims, but trade secret claims.
There are mutliple types of IP, some of which have legal protection and some which don't. Copyrights and patents have legal protection. Trade secrets, however, are protected only to the extent that the owner manages to keep a secret. This was a deliberate choice of the founding fathers. Protecting copyrights and patents (for a limited time) encourages innovators to publish their work, which is good for everyone. Trade secrets are, by definition, unpublished - which is bad for progress.
Trade-secrets are protected by various forms of non-disclosure agreements. However, if some party in contract to you disclose the trade-secret, the recourse is to sue for damages.
The vast majority of users will not have any need for this kind of bandwidth for quite a while. People doing heavy graphics/video processing will like it but 99% of the public will yawn.
There are two major benefits, however.
Ease of Manufacture - it is easier and cheaper to manufacture systems that use serial protocols. Cables are cheaper. There are fewer traces on the circuit boards. The chips themselves may be more complex but once the chip fab process is tooled, who cares. This means that Dell and HP and all can shave a few more bucks off a mother board. That is a "Good Thing".
Close the I/O Gap - Server systems, including mainframes, are often not distinguished from PCs and Workstations by the speed of their processors, but by the capacity of their I/O buses. I/O is the limiting factor in many commercial applications. Faster I/O means that cheaper commodity systems can host bigger databases or more heavily used web servers. In some ways this is an Intel attack on exotic equipment such as Sun's fiber based disk arrays.
Adoption will be fairly fast because so many facilities are built right on the motherboard today. Since much of the market never installs a PCI board, there is nothing preventing them from buying a PC based solely on this new technology, particularly since the new hardware won't be expensive.
And the economies of scale in sharing more hardware throughout the line from consumer PCs to high end servers will be good for everyone. Now we'll be able to steal more equipment from work (just kidding).
I'm watching to see when the processors start talking serial directly. Getting rid of the exotic seven thousand pin packages for processors (and their associated sockets) will be another great savings.
Quick roundup - by a current user of VMS (on Vax!):
-- standard commands, using dir for examples -- dir foo.txt;* operate on all versions of foo.txt dir foo.txt;0 operate on the most recent version dir foo.txt;n operate on version n dir foo.txt;-n operate on the nth most recent version
-- The defaults -- dir foo.txt - show me all versions delete foo.txt - doesn't work, needs explicit;* edit/tpu foo.txt - open the most recent version
-- other interesting commands -- purge foo.txt/keep=n - keep the n most recent versions copy bar.txt foo.txt - create new version of foo.txt with version number incrmented copy bar.txt foo.txt;n - overwrite foo.txt version n
-- remember most recent version has largest num -- $ dir foo.txt foo.txt;1 - earliest foo.txt;2 foo.txt;3 - latest
Versioning is great. VMS has other cool features, as well. My favorite is the "environment" (LOGICALS in VMS speak) spaces that are shared at different system levels. To put it into Unix speak, imagine that an appropriately privileged user could export an environment variable to every environment on the system. The logicals can be used for lightweight IPC, better than writing flag files to the filesystem. If I recall there are four namespaces of logicals: SYSTEM - shared by everyone; USER - shared by all processes for a given USER; PROCESS - shared by a process and all its subprocesses; and LOCAL - isolated to a process.
The worst parts of VMS are the bizarre device:[dir.subdir]filename.ext;version format and the lack of real pipes.
J2EE is a bloated, poorly designed class library well-removed from the hardware or operating system...
This isn't true. J2EE is a specification for the API of an application server. It really is a virtual OS, providing process management and resource allocation to applications. It is, in truth, very comparable to an OS.
The class library is that you write of is not the J2EE specification, it is the specification of the J2EE programmer's API, the way in which J2EE applications will request services from the container (or virtual OS) and the way in which the container will activate, message, and deactivate applications.
There is nothing to prevent a developer from writing a real operating system that speaks directly to hardware and implements the J2EE API. I don't debate that it would be a silly excercise.
Please note that I've chosen not to debate your use of the adjectives "bloated, poorly designed".
I think that describing J2EE as a OS spec and J2EE application servers as OSs is not always inappropriate. (Of course there are limits to the model). Most J2EE servers are implemented to run in a Java virtual machine. J2EE becomes the virtual OS running on that virtual machine.
Re:I bought one
on
iBox Episode 2
·
· Score: 2, Informative
My impression is that applying a factor of about 2 when comparing clock rates is a decent rule of thumb for the average user.
Ergo, an 800MHz PPC-Mac-OSX runs similar to a 1.6 GHz Pentium-WindowsXP machine.
This, of course, is a very rough rule of thumb based on general user experience. Efficiencies in the OSs and other parts of the architectures and configurations make a big deal here. This is not a reflection of comparitative FLOPS or any CPU benchmark.
Routing is difficult (not impossible) in a totally decentralized infrastructure
Sparsely populated areas have no coverage, thus densely populated areas become islands, or the few links in a moderately populated area between two dense areas become swamped with "backbone" traffic
Quality of service is difficult to maintain - imagine being the Joe Shmoe living a few block between the rest of the world and the Victoria's Secret website - you're the bottleneck for half the world's traffic
Who runs name-servers and the like
While a completely decentralized network may not be feasable on the grand scale, WiFi can still be a practical and worthwhile part of achieving "Universal Service" in urban and suburban areas.
I would envision a device like this:
Start with a WiFi BroadBand Router plugged into your Cable/DSL/T3 - whatever.
Add a second firewall to the LAN side - effectively creating a neutral zone for Wifi inside the internet firewall and outside the LAN firewall.
Create a VPN capability to allow authenticated systems access to the LAN.
Add some Quality of Service features to allow the owner of the hotspot the ability to throttle the bandwidth utilized by the public users of the hotspot
Put the whole kit and kaboodle in a little box that is easy to plug in and set up - like most of the home broadband routers.
Give broadband providers some money from the universal service fees and require them to allow their customers to give away reasonable amounts of bandwidth.
For instance, I have a cable modem, and I have been getting bandwidth of about 1.5 Mb/s down and 300 kb/s up. I should be able to plug in a simple box, get secure access to my lan and the internaet at full or almost full speed, and allow my neighbors to share my connection at 256 kb/s down and 50 kb/s up. I think this is a fair compromise. My neighbors get something decent for nothing. The broadband provider still gets to sell bandwidth. I get best speeds for my own use - it is my connection after all.
Perhaps the broadband provider gets some degree of control over the QoS for the public. As their excess bandwidth shrinks they could remotely throttle down the public speeds. (There would need to be an appropriate mechanism to prevent them from just throttling things down without cause, but they should be allowed to give priority to their paying customers. Maybe throttle-downs would reduce their fees from the Universal Service funds.)
Slashdot Mirror
NAT works by creating a translation table: The private network has lots of IP address, each of which are talking on a relative few ports. The external side has 1 IP address talking on lots of ports. Each PrivateIP+Port combination on the inside needs one PublicID+Port combination on the outside. Since there are only 16 bits of port and a 10.* network has 24 bits of private IP space, this system would collapse under moderate load.
Broadband is becoming fairly cheap. A decent commercial quality DSL line probably runs about $200-$300 a month. Add a $1,000 one time fee for some high end access points and you can WiFi enable your coffee shop or book store. At those rates it can become feasible to offer WiFi for free, just to differentiate your business from competitors.
For that matter, public parks in cities can get WiFi services to encourage their use. Bryant Park in Manhattan (6th Ave and 42nd St) started all sorts of programs to get people using the park, displacing the drug dealers. WiFi is just another service that can be offered in this vein.
WiFi will be successful because it is "too cheap to meter". Only equipment manufacturers will make money on it, though.
Here! Here! Well spoken, Bruce!
Part of the problem is that Dynastar454 chose to target "North American" with his joke. "North American" isn't funny to anyone. Americans (US-ians) and Canadians have a long tradition of poking fun at each other. Because we are so similar culturally, it is amusing to describe each other in ways that would indicate a real rivalry.
If Dynastar had used "America" as his target, all the Canadians would have laughed. I swapped in Canadian and all the Americans laughed. We could have swapped in France and the whole world would hae laughed.
Many slashdotters rant that ownership of information is immoral, but I doubt that many really consider the alternatives.
Once the printing press was invented/commercialized by Guttenburg, It suddenly became orders of magnitude cheaper to produce books. In the disorder that followed - the disorder present whenever technology leaps ahead of law - publishers found new material to publish by simply walking into the shops of other publishers and buying copies of their books. Publisher Foo would pay an author to create content, then publisher Bar would buy one copy of Foo's book and copy the contents. Bar would unercut the price, since he didn't have to make back the money paid to the author.
In the long run, there is no motive to pay authors. Authors need to eat in order to write. Therefore the government stepped in and created copyrights.
With patents, the issue is slightly different. If the invention was the product, there is little to protect the inventor from copying, hence little opportunity for the inventor to benefit.
If the invention, however, was not the product, but a process or procedure to make the product, it could be protected by keeping it a trade secret. Trade secrets are just that - secrets. Therefore, there is no way for another inventor to use one invention as the starting point for another. This is bad for innovation.
Patent law was created to give inventors limited monopolies on their ideas in exchange for publishing them for other inventors to use.
Patent and copyright law is useful and beneficial for society. The problem today is two-fold: protection is granted to overly broad "innovation" and protection is granted for increasingly long periods of time. I think we all intuitively agree that "1-click" purchasing shouldn't be patentable. But a truly new and innovative scheduling algorithm might be worthy. And it seems that exclusive periods are much too long for many inventions.
A few ideas I have for reform...
- Increase the standards for getting a patent - currently almost anything can be patented, the burden of proof that something doesn't deserve a patent falls to accused infringers.
- Reduce the exclusive period in accordance with the type of invention. Pharmaceuticals require huge investment, so drug companies need 15-20 years to make back their investment. Software patents should run out after no more than five years.
- Reduce the exclusive period for copyrighted material.
- Require that copyrighted material also be published. Therefore source code would need to be published in order to be protected.
Trademarks and Servicemarks are not so much about intellectual property, but about allowing consumers to reliably identify producers. Chrysler should not be able to name a car "Honda" and trick consumers into buying their product.If for some reason Canada was wiped of the face of the earth tomorrow people would still be able live elsewhere.
Sure there would probably be major setbacks. It might take another five years to get the NHL to the point were we are now but things would eventually get back to normal.
So RMS's statement that Canada is no longer essential is true.
I vacationed to Canada and it seemed OK. But if I _had_ to I could learn to vacation in Europe, or Asia even. I know both Europe and Asia are working on getting there[sic] people to speak English.
If Canada went away tomorrow it would be a real shame (understatement of the year) but it would not be the end of the world.
The very fact that both vi and emacs support regular expressions must mean they are a best-in-breed tool, because if there was a way for those two communities to disagree, they would have done it.
I love the fact that I can use the same expressions with grep, sed, vim, Perl, and Java. that being said, however, the critics are who warn that regex can be over used are correct: regex's are difficult to debug and to maintain, so don't go overboard.
So HP-UX and Minix are out?
I always chuckle when I see someone spend extra for a 100 Mb or 1 Gb ethernet card, then plug their PC directly into a cable modem that does 1.5 Mb. I want to tell them to go to their local PC builder and get an ethernet card from the spares pile.
I thought that a big part of the excitement over the new Mac products was the high performance bus? Raw processor speed is always only part of the story. If the Mac can move data amongst ram, disk, processor, and video card much faster than the Intel/AMD systems, the net result is a more powerful PC.
I run a 633 celeron based eMachines. I pulled the OEM hard drive and replaced it with a 7200 RPM model. My user experience performance doubled - although the bechmarks wouldn't have changed. This is reasonable, since on an average desktop machine, the user's waits at boot and when loading an application from disk. Both those processes ran much faster with the high speed hard drive.
Just a note, besides web browsing/word processing I develop in Java. Tools such as Ant do a very good job of incremental compiling so I never have to kick off a make script and wait.
So, in two months we'll be able to buy a 64 bit workstation class PC with a high quality OS and a great UI? This can't be a great thing for Intel.
We are currently in the process of replacing the system with a new system for a cost rumored to be around two hundred million dollars to the US territory.
The logic goes like this...
- The global management decides to buy an app. for the full organization.
- The global mangement also decides that they are going to charge back the development and liscensing costs to all the territories regardless of whether they use the new system.
- A bean counter in the US says "Why are we paying fifty million dollars for a system we aren't using?"
- No one has the cajones to say, "Global has forced us to through away $50,000,000. Is throughing away another $150,000,000 a good idea?"
Of course, 1 year earlier we couldn't get the $5,000,000 to do a rewrite on a modern Unix platform.Diamond does discuss sun-Saharan Africa. He acknowledges the development of civilization there based on sorghum and the ass. He claims that this culture was less advanced because it had fewer less protein rich staple foods and less capable draft animals. It also had less room to spread, so it dominated sub-Saharan Africa but couldn't break out of the continent.
The wheat/bean/horse/oxen culture of the fertile crescent had the geography to spread all the way into India and to Britain. Plus, having more protein, more animals, more area, this culture developed more. It still took thouasands of year before the fertile crescent people were able to enslave the sub-Sahran people.
As for Dravidian, it seems to have developed in the Indus Valley, where civilization also developed independently. I don't have specific knowledge, but I would suspect that fertile crescent culture mixed with Indus culture when the Indo-Iranian (Persian, Hindi, etc.) speakers moved into India around 1700 BC.
An important point to keep in this thread is that when we discuss culture in this context, we need to think about questions like "What did they grow? What animals did they use? What farming methods did they use?" I'm not sure if the Dravidian speakers of Southern India adopted the culture (by this limited measure) of the fertile crescent people while maintaining their own language. Perhaps the Indus Valley culture was had enough in common with the fertile crescent culture to make this an easy transition.
It is a pretty consistent observation that lots of cultures invented the wheel, but only those that had access to high quality draft animals used it. Remember that the horse and other draft animals (oxen, donkey, etc.) were extinct in the new world until (re)introduced by the Europeans in 1492.
A great book on the subject is Guns, Germs, Steel: The Fate of Human Societies by Jared Diamond. Diamond argues that two dominant cultures have arisen - A Western culture that traces its roots to Fertile Crescent in modern day Iraq and the an Eastern culture that traces its roots to the Yellow River Valley. In both of these places nature and geography conspired to create a package of tools that allowed these cultures to spread.
Both these places had the following...
- Naturally occuring staple foods - usually grains - that were easy to domesticate
- Large wild animals that were easy to domesticate and useful as draft animals
- Room to spread out while using the same tools
In contrast, the natives of the Americas had only a single staple grain - corn - and that one took thousands of years longer than wheat, barley, oats, and rice to domesticate and they had no draft animals. As an added gotcha, when the American natives did manage to domesticate corn, there were barriers to spreading out. For instance, the people of Mexico - Aztec, Mayan, Toltec - would need to pack up and cross the American Southwestern deserts, then the great plains (which can't be farmed easily without steel plows), then the Appalachian mountains, before reaching readily farmable land in the Eastern USA. The Chinese and Middle Eastern peoples could spread all the way to Korea, India, North Africa, and Europe without hitting that much of a barrier.If MS buys SCO outright, however, they gain very little but put themselves at great risk. Assume that SCO's loses, and perhaps is even deemed to have made fraudulent claims. Then the multi-BILLION dollar countersuit is coming out of Bill's pocket. As it is, if SCO loses and is countersued - there is not much value to be siezed.
The worst outcomes for MS are really 1) SCO loses quickly 2) SCO is bought out by IBM, RedHat, or someone else and the suit is dropped.
The interesting twist is that, if they are really smart, SCO may be playing double agent. First they take a big pile of money from MS for licensing-wink wink-and make alot of noise. (It has to look good, so as to get more money from Bill). Then, at the last minute, they make a deal with IBM to trade their IP portfolio for a the pieces of IBM's consulting business that pertain to the "small fish" that form SCO's traditional customer base but that IBM is not quite comfortable with.
The net result would then be that IBM and SCO both shed portions of their business that they don't want, and use a big chunk of Bill's own money to compete with Bill's own server products. Not to mention that IBM has technology licensed to MS, which they'll watch like a hawk, ready to attack MS with lawyer Shock and Awe at the slightest provocation.
This is a plot worthy of Vince MacMahon!
I still say the PC-Junior was ahead of its time!
Poke arround the main site. there is some info there about starting your own WISP and selling service to your neighbors.
I've heard mixed reports about using Vonage and satelite ISP. YMMV.
Good Luck.
- Start with roughly 8 feet of sinew
- Add a small leather basket in the middle.
- Place a small stone in the basket.
- Hold one end of the sinew in the palm of your hand and the other between your thumb and forefinger. Adjust the length so that the basket is in the exact middle.
- Whirl it around your head. Release the sinew between your thumb and forefinger at exactly the right moment to cause the stone to fly off and hit your target.
As you might imagine, doing this accurately is quite difficult. The only people that typically use the sling are shepherds, since they have the three or four hours a day to spend practicing. As a comparison, a bowman only needs about 30 minutes of daily practice to be highly skilled.Damn, I watch too much History Channel.
Bull! /. readers don't have girlfriends!
Note that SCO has not made copyright or patent claims, but trade secret claims.
There are mutliple types of IP, some of which have legal protection and some which don't. Copyrights and patents have legal protection. Trade secrets, however, are protected only to the extent that the owner manages to keep a secret. This was a deliberate choice of the founding fathers. Protecting copyrights and patents (for a limited time) encourages innovators to publish their work, which is good for everyone. Trade secrets are, by definition, unpublished - which is bad for progress.
Trade-secrets are protected by various forms of non-disclosure agreements. However, if some party in contract to you disclose the trade-secret, the recourse is to sue for damages.
[I]f someone innocently receives trade secret rights, it will be very difficult to enjoin them from use or further disclosure of the trade secrets. Thus, IBM may need to pay SCO damages, but now that the secrets are widely known, SCO cannot attack Linux.
If SCO had a patent or copyright claim, Linux would be screwed. But SCO only has claims against those in "privity of contract" to not disclose.
Primer on trade secrets.
The vast majority of users will not have any need for this kind of bandwidth for quite a while. People doing heavy graphics/video processing will like it but 99% of the public will yawn.
There are two major benefits, however.
Adoption will be fairly fast because so many facilities are built right on the motherboard today. Since much of the market never installs a PCI board, there is nothing preventing them from buying a PC based solely on this new technology, particularly since the new hardware won't be expensive.
And the economies of scale in sharing more hardware throughout the line from consumer PCs to high end servers will be good for everyone. Now we'll be able to steal more equipment from work (just kidding).
I'm watching to see when the processors start talking serial directly. Getting rid of the exotic seven thousand pin packages for processors (and their associated sockets) will be another great savings.
Versioning is great. VMS has other cool features, as well. My favorite is the "environment" (LOGICALS in VMS speak) spaces that are shared at different system levels. To put it into Unix speak, imagine that an appropriately privileged user could export an environment variable to every environment on the system. The logicals can be used for lightweight IPC, better than writing flag files to the filesystem. If I recall there are four namespaces of logicals: SYSTEM - shared by everyone; USER - shared by all processes for a given USER; PROCESS - shared by a process and all its subprocesses; and LOCAL - isolated to a process.
The worst parts of VMS are the bizarre device:[dir.subdir]filename.ext;version format and the lack of real pipes.
If SCO accepted this plan, we would be able to find all of the infringing code. If we found the infringing code, we would know SCO's trade secrets.
SCO can't let that happen!
This isn't true. J2EE is a specification for the API of an application server. It really is a virtual OS, providing process management and resource allocation to applications. It is, in truth, very comparable to an OS.
The class library is that you write of is not the J2EE specification, it is the specification of the J2EE programmer's API, the way in which J2EE applications will request services from the container (or virtual OS) and the way in which the container will activate, message, and deactivate applications.
There is nothing to prevent a developer from writing a real operating system that speaks directly to hardware and implements the J2EE API. I don't debate that it would be a silly excercise.
Please note that I've chosen not to debate your use of the adjectives "bloated, poorly designed".
I think that describing J2EE as a OS spec and J2EE application servers as OSs is not always inappropriate. (Of course there are limits to the model). Most J2EE servers are implemented to run in a Java virtual machine. J2EE becomes the virtual OS running on that virtual machine.
Ergo, an 800MHz PPC-Mac-OSX runs similar to a 1.6 GHz Pentium-WindowsXP machine.
This, of course, is a very rough rule of thumb based on general user experience. Efficiencies in the OSs and other parts of the architectures and configurations make a big deal here. This is not a reflection of comparitative FLOPS or any CPU benchmark.
While a completely decentralized network may not be feasable on the grand scale, WiFi can still be a practical and worthwhile part of achieving "Universal Service" in urban and suburban areas.
I would envision a device like this:
- Start with a WiFi BroadBand Router plugged into your Cable/DSL/T3 - whatever.
- Add a second firewall to the LAN side - effectively creating a neutral zone for Wifi inside the internet firewall and outside the LAN firewall.
- Create a VPN capability to allow authenticated systems access to the LAN.
- Add some Quality of Service features to allow the owner of the hotspot the ability to throttle the bandwidth utilized by the public users of the hotspot
- Put the whole kit and kaboodle in a little box that is easy to plug in and set up - like most of the home broadband routers.
- Give broadband providers some money from the universal service fees and require them to allow their customers to give away reasonable amounts of bandwidth.
For instance, I have a cable modem, and I have been getting bandwidth of about 1.5 Mb/s down and 300 kb/s up. I should be able to plug in a simple box, get secure access to my lan and the internaet at full or almost full speed, and allow my neighbors to share my connection at 256 kb/s down and 50 kb/s up. I think this is a fair compromise. My neighbors get something decent for nothing. The broadband provider still gets to sell bandwidth. I get best speeds for my own use - it is my connection after all.Perhaps the broadband provider gets some degree of control over the QoS for the public. As their excess bandwidth shrinks they could remotely throttle down the public speeds. (There would need to be an appropriate mechanism to prevent them from just throttling things down without cause, but they should be allowed to give priority to their paying customers. Maybe throttle-downs would reduce their fees from the Universal Service funds.)