The Guardian rightly pointed out that Hugh Grant's tapes said almost nothing new, that the Prime Minister had been implicated long ago. Yes, Murdoch has corrupted UK politics (well, he wasn't the only one to be fair, but he is a major player). The government knows that Murdoch's papers now decide who win in general elections (much as happens in the US when the moguls decide who gets airplay and who doesn't). Because of this power, the government is effectively powerless.
Mind you, that is its own fault. The British political system has essentially three sections - those elected on the grounds of democracy (the Commons), those selected - supposedly, but rarely, on the grounds of merit (the Lords), and those hired to do the legwork (the Civil Service). The downgrading of the powers of the House of Lords, the selling of honours and the use of appointments to keep favoured MPs in power after they are rejected by the people means the Lords is effectively useless. And, yes, that matters in a case like this. It limits who outsiders can hold hostage. By politicizing the otherwise apolitical Civil Service, you also have an organization that wants corruption - especially when it happens to the "wrong side".
Then-Prime Minister Gordon Brown, not to mention several other leading cabinet figures in the government, were certainly hacked. The midnight raid by police suggests that they're beginning to get more than a little concerned that there might indeed be national security implications. The efforts by the Civil Service to block criminal investigations at the time the story first broke also hints that hushing things up was a major concern, over and above any legal liability.
The part of the story that isn't getting much news is that much of the former Fleet Street operates in the same manner. This scandal won't just affect NotW but will touch ALL tabloids. There will be paid crackers, hackers and phone phreaks working for all of them.
(It's possible that the investigation will also implicate papers in other parts of the world - do you seriously imagine any paper that feeds off paparazzi hijinks won't have dabbled in such practices?)
Not quite. High treason as a capital crime was abolished back in the 90s I think. On the other hand, she can appoint whomever she likes (such as editors for rival papers) to the House of Lords in the Queen's Birthday honours (the only honours she can bestow without political interference). That's one thing that the Queen really should use that list for - promoting rivals to those who seek power and influence criminally. Democracy is great, but just like video games you really do want to be able to use cheat codes in special cases.
Not sure UUCP is in common use. The initial specs for Telnet appear in RFC 97 and are dated 15 February 1971. I think that if we're considering modern FTP to be a continuation of the original FTP, then modern Telnet is a continuation of RFC 97. I think that's the oldest protocol. ICMP is often considered IPv0, so that might be older still, but Telnet has FTP beat.
scp doesn't distract people with popups, doesn't eat resources, tends to be more stable, doesn't require a GUI interface, won't allow people to get distracted with porn and/or Cowboy Neil (or, indeed, porn of Cowboy Neil), is far friendlier to network resources than HTTP (HTTP is =HORRIBLE= for file transfers), and prevents people listening in to what you download.
From a security standpoint, very few FTP sites provide signatures, FTP and HTTP can suffer mitm attacks (all it takes is a trojaned machine anywhere in the chain between you and the site), FTP is Bad News because firewalls tend not to like it, and HTTP is Bad News because web clients have this obnoxious habit of trying to process anything they receive - which tends to be where most of the security holes come from.
From a server standpoint, HTTP connections keep connection state information for several minutes. This is a heavy resource drain. Web servers are also fairly heavy beasts - they've got to maintain a lot of internal state, the protocol is heavy, etc. It doesn't take much to exhaust a machine running, say, Apache 2.x of resources. On the other hand, any simple copy-a-file mechanism can afford to preserve almost no state at all, the processing of the requests is usually trivial and doesn't need hundreds of support libraries, and you don't need keepalives.
Which brings us back to security. Simple Is Good. Complex Is Bad. Every possible arc through the code is a potential haven for bugs. I dare anyone on Slashdot to even imagine how many arcs there are on a Cold Fusion site (you've the app, Cold Fusion itself, JRun4, Java and the web server, and any bug in any of that could potentially result in a security hole or a server crash). Now consider how many arcs there are in what is essentially a linear network copy process.
Compared to other cars of that vintage, it was. Ford sold mass-produced cars that were stocked high and sold cheap. The Fry's or Walmart of the car industry. They were popular because people could afford them, not because they worked well.
I agree (though if you are going to consider sftp, please also consider ftps), but it has been surprisingly durable. Rivals, historically, have included fsp, scp, rsync, uucp, WAIS, gopher and ftpmail. Some, like WAIS and gopher, also provided a far superior interface to the traditional FTP client.
Of these, scp and rsync are the only ones still in use today and I don't know of any anonymous FTP sites that provides scp, though I think kernel.org provides rsync.
About the only significant change to FTP since it began was that people used to use archie to find programs. (Archie, for those too young to remember, was a search engine specifically for anonymous FTP sites. You gave it a regexp, it gave you every site that had files that matched and the full directory path of those files. Because it was specialized, there was no risk of clutter. Equally, there was no chance it would survive into the era of web crawlers and generalized search engines.
I'm not anti-American, I'm anti-claiming-what-isn't-yours, no matter who does the claiming.
I'm absolutely against fraudulant claims. I don't care who makes them. I don't care where they make them, when they make them, how they make them. If the claim isn't backed up by reality then the claim should be utterly rejected and the claimant with it.
The Americans invented the Internet Protocol, via DARPAnet. THAT was the sum total of their contribution. And I'd hardly call IPSS a "failed network" - it was supporting graphical hypertext, MMORGs and home users many, many years before the likes of AOL even existed. The X protocols were replaced by IETF protocols, but that's about it.
Now consider the Internet in America today. Could they replace one set of protocols with another, with such amazing ease, lack of complaint and lack of financial hardship? IPv6 in the home has the answer to that. No they cannot. What the IPSS were able to achieve, the US part of the Internet cannot do.
The IPSS is technically still there - X.400 and X.500 are still used, X.25 is still in place, the same satellites and same wiring is used to link machines together today as was used in the 80s, the companies that ran IPSS still exist. That's quite impressive for a "failure". All that has changed is an adoption of IETF RFCs. And the IETF isn't an American group, it's a global group.
If someone were to ask me if America invented IPv0 and IPv4, the answer would be a categorical YES. If they also asked if America invented the bulk of services and standards used on the Internet today, I'd also say yes.
Likewise, if someone asked me if CERN invented hypertext, I'd say that honour really was split between British Telecom and Ted Nielson. But if they asked me if CERN invented the WWW, I'd say yes. The two are not synonymous. If that pisses you off, you're an idiot. If A != B, then there is nothing you can do to make A = B. Your opinion doesn't change history or the definition of words in English.
If someone asked whether the computer was invented in America, or indeed in England, I'd say no. They BOTH deserve credit, von Neumann for his contribution and Alan Turing for his. I WILL NOT accept personal bias in such debates, equal contributions get equal credit and to hell with those who put extremist views above what actually happened.
No, that doesn't work. You can't use a starter independently. You can use an X.25 network independently. In fact, X.25 had a lot of advantages over IPv4 - a better, more logical, addressing scheme and highly secure protocols. Combine the full X.400 and X.500 specs together and you've a system that the modern Internet is approaching and may actually achieve in the next 20 years. Internet 2 can be thought of as being the fast but stripped-down version, but despite Internet 2 software being mostly Open Source, I doubt any admin (including those using Slashdot) can name a regular Internet site that uses any of it.
If you want a car analogy, then X.25 is probably closer to picturing a 1880s battery-powered electric car (they did exist) being merged with an 1880s gasoline car to produce a fully-functional hybrid as good as any that exist today but over a century earlier. (It is infuriating in many ways that this did NOT happen with cars, and it is proof of the occasional value of government that when what was an upscaled DARPAnet merged with IPSS that they opted to take advantage of the strengths of both.)
Remember, X.25 routers included satellite links, the X.25 network used high-performance (for the time) land lines, the level of investment was infinitely greater than that put in by UUnet. (Or did you imagine they scrapped all those satellites and replace them with IP-aware ones?)
If Bell invented switchboards that could support routing IP traffic in real-time with only a basic firmware upgrade, and Bell had those switchboards on a global scale with lines that could carry Internet traffic for several continents, then damn right I'd call him a co-inventor. You notice that "co" bit? The bit you missed out with your feeble arguments, in perfect knowledge that I was talking about credit that was distributed and NOT given to one person or one group? Good. Remember it. Burn it into you brain if you have to.
Now, it just so happens Bell did not do that but the consortium managing IPSS did. They ARE therefore co-inventors. DARPAnet and UUnet would never have scaled globally - way too primitive, the engineers had no experience with massive, multinational delay-tolerant networks, the routing technologies needed just didn't exist in the States. This is not the mild improvement of replacing a hand crank with a shotgun starter or an electric starter. Using the example I gave, this would be like having a Prius no later than 1900. Which goes to demonstrate what happens when COMPETENT engineers don't treat others like dirt and grind their preferences into the ground. It is people like you who are to blame for cars being 100 years behind where QUALITY engineers would have been able to put them.
Your "Not Invented Here" attitude is not even archaic, historically scientists and engineers have always preferred to borrow from the best of the rest. It is an attitude that is incapable of success and incapable of admitting the success of others. It is an attitude that is corrosive, destroying original thought and creativity as it consumes observation and curiosity.
1) Packet headers in fields are easier to work with.
In IPv4, you have fields that are a single bit, a single nibble, a single byte, a word and a double word. Bit fields like priority/QoS are processed according to a value somewhere else. They are HORRIBLE to work with, the cost of actually digesting a header is HUGE!
In IPv6, most of this is cleaned out. Header entries that are never used aren't there. Those that are seldom used are defined as header extensions and aren't in the default structure. Everything else is a simple, easy-to-work-with size with a single meaning. So although the total header is larger, it doesn't matter. The cost of pulling the header into memory is insignificant compared to the cost of finding stuff in it. With IPv6, you pull in a word or a double word. This contains the data and nothing else, in the correct format, with no need to digest it. It is already in the right form.
2) Fragmentation in IPv4 is common, fragmentation in IPv6 is rare (which solves your objection, BTW)
In IPv4 or IPv6, if a packet is fragmented then a firewall or a proxy cannot do anything with the inbound packets until the system has all of them. The latter packets don't have the necessary information. This is a huge congestion issue.
With IPv4, you can't do a damn thing about it. With IPv6, each connection has an automatically adjusted MTU according to the largest packet that can pass through unfragmented. There are exceptions to this, but it can be considered the norm. This means packets won't be broken up, which means ALL packets can be processed and routed at the time of receiving them without the need to delay for reassembly.
3) Faster to perform routing
In IPv4, you have to search the router table for the best rule for a given connection. Router tables can have millions of entries, but if the router table is either tree-based or indexed-sequential then you only have to deal with around 20 steps for 2^20 entries.
In IPv6, you still have to search the router table but because of the way the addressing scheme works, you only ever have to compare 2 bytes of the address. This gives an effective router table size of 512, which means you can always find the right route in 9 steps. Less than half the number required for IPv4.
4) Less memory fragmentation
In IPv4, headers aren't properly aligned and that can lead to memory being either fragmented or never used. In the case of fragmentation, malloc will be more expensive than necessary. If you use a ring buffer for the headers, then you avoid fragmentation but it means you have to compare the bytes in the header with the bytes available in the buffer. Counting is also expensive. True, you can avoid this by treating everything as a byte stream and when you hit the field that says how many bytes there are read that many bytes minus the number already read. Byte streams are slooooow and expensive on a 32-bit or 64-bit architecture.
In IPv6, headers are properly aligned and leads to efficient packing and efficient queueing. If you pull an item off the queue, you can then push an item off the queue. Thus, you don't have to do byte counts, you simply need to record how many packets are on the queue. Because the header will always be an exact multiple of 32-bits in length (and almost always an exact multiple of 64-bits), you can use the entire width of the data bus, you can use native registers, etc. Just on usable bandwidth alone, you've quadrupled the best performance IPv4 headers can do.
5) Extensibility
In IPv4, the header contains everything it could ever need under any condition. Which means every device has to load the whole damn thing and digest the whole damn thing. This means that if the header does need extra information, tough. Nowhere to put it, unless you re-use an existing component and have all devices try to figure out what meaning that component has in the current context. (This has happened with the priority bits already.)
Not really, X.25 networks had gone global (International Packet Switch Stream) at a time the Internet was still purely an American toy. The Internet became global because the rest of the world had got there first - hardware-wise, at least. All the early transatlantic links were IPSS lines re-purposed, as was all the early European Internet capability. The Americans got the software side first.
Since the modern Internet is a marriage between software and hardware, and not one or the other alone, the only fair conclusion is that it was a global invention with no nation being able to claim credit for being truly first.
Not only further along the learning curve, but further along in mass availability. Mass availability = low cost. Low cost = more attractive to customers. I thought America had learned (the hard way) what happens when you ship inferior, expensive products after their car industry collapsed and Japan pwned them. I also thought America had learned (the hard way) what happens when you start behind your competition after they were thrashed by the USSR in the early space race. Catching up was damn expensive.
Address shortages are a very, very, very tiny, miniscule fraction of IPv6. If IPv6 was about address shortages, the IPng working group would have adopted TUBA.
You seem unwilling to even recognize any of the other features of IPv6:
Built-in security Built-in device mobility
Built-in network mobility
Built-in multimedia support
Extensible headers for dynamic protocol upgrades
Auto-configuration
Reduced latency
Improved router reliability (partly due to simpler routing protocols)
Native multicasting
Native anycasting
Superior QoS support
Don't even think of coming back with "but nobody uses these" - nobody was driving until the car was adopted either. Things have a habit of not being used when they're not available. When they are available, they are used. It's as simple as that.
^get there first^own ICANN and enough critical infrastructure to demand whatever the hell they like, no questions asked, regardless of any actual needs
Slashdot Mirror
The Guardian rightly pointed out that Hugh Grant's tapes said almost nothing new, that the Prime Minister had been implicated long ago. Yes, Murdoch has corrupted UK politics (well, he wasn't the only one to be fair, but he is a major player). The government knows that Murdoch's papers now decide who win in general elections (much as happens in the US when the moguls decide who gets airplay and who doesn't). Because of this power, the government is effectively powerless.
Mind you, that is its own fault. The British political system has essentially three sections - those elected on the grounds of democracy (the Commons), those selected - supposedly, but rarely, on the grounds of merit (the Lords), and those hired to do the legwork (the Civil Service). The downgrading of the powers of the House of Lords, the selling of honours and the use of appointments to keep favoured MPs in power after they are rejected by the people means the Lords is effectively useless. And, yes, that matters in a case like this. It limits who outsiders can hold hostage. By politicizing the otherwise apolitical Civil Service, you also have an organization that wants corruption - especially when it happens to the "wrong side".
Then-Prime Minister Gordon Brown, not to mention several other leading cabinet figures in the government, were certainly hacked. The midnight raid by police suggests that they're beginning to get more than a little concerned that there might indeed be national security implications. The efforts by the Civil Service to block criminal investigations at the time the story first broke also hints that hushing things up was a major concern, over and above any legal liability.
The part of the story that isn't getting much news is that much of the former Fleet Street operates in the same manner. This scandal won't just affect NotW but will touch ALL tabloids. There will be paid crackers, hackers and phone phreaks working for all of them.
(It's possible that the investigation will also implicate papers in other parts of the world - do you seriously imagine any paper that feeds off paparazzi hijinks won't have dabbled in such practices?)
Not quite. High treason as a capital crime was abolished back in the 90s I think. On the other hand, she can appoint whomever she likes (such as editors for rival papers) to the House of Lords in the Queen's Birthday honours (the only honours she can bestow without political interference). That's one thing that the Queen really should use that list for - promoting rivals to those who seek power and influence criminally. Democracy is great, but just like video games you really do want to be able to use cheat codes in special cases.
Not even close. Far, far older stories have been published as new on Slashdot over the years. Mind you, it's still impressive.
Hey, dying in ditches spreads disease! Please change to "falls into an active volcano", at least for environmental reasons.
How often do you download Firefox? In all fairness to other vendors, I think we should know so the figures can be adjusted.
Not sure UUCP is in common use. The initial specs for Telnet appear in RFC 97 and are dated 15 February 1971. I think that if we're considering modern FTP to be a continuation of the original FTP, then modern Telnet is a continuation of RFC 97. I think that's the oldest protocol. ICMP is often considered IPv0, so that might be older still, but Telnet has FTP beat.
scp doesn't distract people with popups, doesn't eat resources, tends to be more stable, doesn't require a GUI interface, won't allow people to get distracted with porn and/or Cowboy Neil (or, indeed, porn of Cowboy Neil), is far friendlier to network resources than HTTP (HTTP is =HORRIBLE= for file transfers), and prevents people listening in to what you download.
From a security standpoint, very few FTP sites provide signatures, FTP and HTTP can suffer mitm attacks (all it takes is a trojaned machine anywhere in the chain between you and the site), FTP is Bad News because firewalls tend not to like it, and HTTP is Bad News because web clients have this obnoxious habit of trying to process anything they receive - which tends to be where most of the security holes come from.
From a server standpoint, HTTP connections keep connection state information for several minutes. This is a heavy resource drain. Web servers are also fairly heavy beasts - they've got to maintain a lot of internal state, the protocol is heavy, etc. It doesn't take much to exhaust a machine running, say, Apache 2.x of resources. On the other hand, any simple copy-a-file mechanism can afford to preserve almost no state at all, the processing of the requests is usually trivial and doesn't need hundreds of support libraries, and you don't need keepalives.
Which brings us back to security. Simple Is Good. Complex Is Bad. Every possible arc through the code is a potential haven for bugs. I dare anyone on Slashdot to even imagine how many arcs there are on a Cold Fusion site (you've the app, Cold Fusion itself, JRun4, Java and the web server, and any bug in any of that could potentially result in a security hole or a server crash). Now consider how many arcs there are in what is essentially a linear network copy process.
Which is why we should scrap TCP and UDP, using SCTP and DCCP instead.
But almost nobody sends files via http. Way too primitive. FTP is still king there, followed by torrent.
Whilst I don't own a Model T or a car from 1971, I would personally prefer any of the cars from the Brighton Rally anyway. Much classier.
Compared to other cars of that vintage, it was. Ford sold mass-produced cars that were stocked high and sold cheap. The Fry's or Walmart of the car industry. They were popular because people could afford them, not because they worked well.
I agree (though if you are going to consider sftp, please also consider ftps), but it has been surprisingly durable. Rivals, historically, have included fsp, scp, rsync, uucp, WAIS, gopher and ftpmail. Some, like WAIS and gopher, also provided a far superior interface to the traditional FTP client.
Of these, scp and rsync are the only ones still in use today and I don't know of any anonymous FTP sites that provides scp, though I think kernel.org provides rsync.
About the only significant change to FTP since it began was that people used to use archie to find programs. (Archie, for those too young to remember, was a search engine specifically for anonymous FTP sites. You gave it a regexp, it gave you every site that had files that matched and the full directory path of those files. Because it was specialized, there was no risk of clutter. Equally, there was no chance it would survive into the era of web crawlers and generalized search engines.
I'm not anti-American, I'm anti-claiming-what-isn't-yours, no matter who does the claiming.
I'm absolutely against fraudulant claims. I don't care who makes them. I don't care where they make them, when they make them, how they make them. If the claim isn't backed up by reality then the claim should be utterly rejected and the claimant with it.
The Americans invented the Internet Protocol, via DARPAnet. THAT was the sum total of their contribution. And I'd hardly call IPSS a "failed network" - it was supporting graphical hypertext, MMORGs and home users many, many years before the likes of AOL even existed. The X protocols were replaced by IETF protocols, but that's about it.
Now consider the Internet in America today. Could they replace one set of protocols with another, with such amazing ease, lack of complaint and lack of financial hardship? IPv6 in the home has the answer to that. No they cannot. What the IPSS were able to achieve, the US part of the Internet cannot do.
The IPSS is technically still there - X.400 and X.500 are still used, X.25 is still in place, the same satellites and same wiring is used to link machines together today as was used in the 80s, the companies that ran IPSS still exist. That's quite impressive for a "failure". All that has changed is an adoption of IETF RFCs. And the IETF isn't an American group, it's a global group.
If someone were to ask me if America invented IPv0 and IPv4, the answer would be a categorical YES. If they also asked if America invented the bulk of services and standards used on the Internet today, I'd also say yes.
Likewise, if someone asked me if CERN invented hypertext, I'd say that honour really was split between British Telecom and Ted Nielson. But if they asked me if CERN invented the WWW, I'd say yes. The two are not synonymous. If that pisses you off, you're an idiot. If A != B, then there is nothing you can do to make A = B. Your opinion doesn't change history or the definition of words in English.
If someone asked whether the computer was invented in America, or indeed in England, I'd say no. They BOTH deserve credit, von Neumann for his contribution and Alan Turing for his. I WILL NOT accept personal bias in such debates, equal contributions get equal credit and to hell with those who put extremist views above what actually happened.
I suspect it's more a "what do we do with a brand name with a collapsing value and no product?"
Tod and Ryley: They pushed us, so we had them Replaced.
No, that doesn't work. You can't use a starter independently. You can use an X.25 network independently. In fact, X.25 had a lot of advantages over IPv4 - a better, more logical, addressing scheme and highly secure protocols. Combine the full X.400 and X.500 specs together and you've a system that the modern Internet is approaching and may actually achieve in the next 20 years. Internet 2 can be thought of as being the fast but stripped-down version, but despite Internet 2 software being mostly Open Source, I doubt any admin (including those using Slashdot) can name a regular Internet site that uses any of it.
If you want a car analogy, then X.25 is probably closer to picturing a 1880s battery-powered electric car (they did exist) being merged with an 1880s gasoline car to produce a fully-functional hybrid as good as any that exist today but over a century earlier. (It is infuriating in many ways that this did NOT happen with cars, and it is proof of the occasional value of government that when what was an upscaled DARPAnet merged with IPSS that they opted to take advantage of the strengths of both.)
Remember, X.25 routers included satellite links, the X.25 network used high-performance (for the time) land lines, the level of investment was infinitely greater than that put in by UUnet. (Or did you imagine they scrapped all those satellites and replace them with IP-aware ones?)
If Bell invented switchboards that could support routing IP traffic in real-time with only a basic firmware upgrade, and Bell had those switchboards on a global scale with lines that could carry Internet traffic for several continents, then damn right I'd call him a co-inventor. You notice that "co" bit? The bit you missed out with your feeble arguments, in perfect knowledge that I was talking about credit that was distributed and NOT given to one person or one group? Good. Remember it. Burn it into you brain if you have to.
Now, it just so happens Bell did not do that but the consortium managing IPSS did. They ARE therefore co-inventors. DARPAnet and UUnet would never have scaled globally - way too primitive, the engineers had no experience with massive, multinational delay-tolerant networks, the routing technologies needed just didn't exist in the States. This is not the mild improvement of replacing a hand crank with a shotgun starter or an electric starter. Using the example I gave, this would be like having a Prius no later than 1900. Which goes to demonstrate what happens when COMPETENT engineers don't treat others like dirt and grind their preferences into the ground. It is people like you who are to blame for cars being 100 years behind where QUALITY engineers would have been able to put them.
Your "Not Invented Here" attitude is not even archaic, historically scientists and engineers have always preferred to borrow from the best of the rest. It is an attitude that is incapable of success and incapable of admitting the success of others. It is an attitude that is corrosive, destroying original thought and creativity as it consumes observation and curiosity.
That and:
1) Packet headers in fields are easier to work with.
In IPv4, you have fields that are a single bit, a single nibble, a single byte, a word and a double word. Bit fields like priority/QoS are processed according to a value somewhere else. They are HORRIBLE to work with, the cost of actually digesting a header is HUGE!
In IPv6, most of this is cleaned out. Header entries that are never used aren't there. Those that are seldom used are defined as header extensions and aren't in the default structure. Everything else is a simple, easy-to-work-with size with a single meaning. So although the total header is larger, it doesn't matter. The cost of pulling the header into memory is insignificant compared to the cost of finding stuff in it. With IPv6, you pull in a word or a double word. This contains the data and nothing else, in the correct format, with no need to digest it. It is already in the right form.
2) Fragmentation in IPv4 is common, fragmentation in IPv6 is rare (which solves your objection, BTW)
In IPv4 or IPv6, if a packet is fragmented then a firewall or a proxy cannot do anything with the inbound packets until the system has all of them. The latter packets don't have the necessary information. This is a huge congestion issue.
With IPv4, you can't do a damn thing about it. With IPv6, each connection has an automatically adjusted MTU according to the largest packet that can pass through unfragmented. There are exceptions to this, but it can be considered the norm. This means packets won't be broken up, which means ALL packets can be processed and routed at the time of receiving them without the need to delay for reassembly.
3) Faster to perform routing
In IPv4, you have to search the router table for the best rule for a given connection. Router tables can have millions of entries, but if the router table is either tree-based or indexed-sequential then you only have to deal with around 20 steps for 2^20 entries.
In IPv6, you still have to search the router table but because of the way the addressing scheme works, you only ever have to compare 2 bytes of the address. This gives an effective router table size of 512, which means you can always find the right route in 9 steps. Less than half the number required for IPv4.
4) Less memory fragmentation
In IPv4, headers aren't properly aligned and that can lead to memory being either fragmented or never used. In the case of fragmentation, malloc will be more expensive than necessary. If you use a ring buffer for the headers, then you avoid fragmentation but it means you have to compare the bytes in the header with the bytes available in the buffer. Counting is also expensive. True, you can avoid this by treating everything as a byte stream and when you hit the field that says how many bytes there are read that many bytes minus the number already read. Byte streams are slooooow and expensive on a 32-bit or 64-bit architecture.
In IPv6, headers are properly aligned and leads to efficient packing and efficient queueing. If you pull an item off the queue, you can then push an item off the queue. Thus, you don't have to do byte counts, you simply need to record how many packets are on the queue. Because the header will always be an exact multiple of 32-bits in length (and almost always an exact multiple of 64-bits), you can use the entire width of the data bus, you can use native registers, etc. Just on usable bandwidth alone, you've quadrupled the best performance IPv4 headers can do.
5) Extensibility
In IPv4, the header contains everything it could ever need under any condition. Which means every device has to load the whole damn thing and digest the whole damn thing. This means that if the header does need extra information, tough. Nowhere to put it, unless you re-use an existing component and have all devices try to figure out what meaning that component has in the current context. (This has happened with the priority bits already.)
In IPv6, only core data is in the heade
Someone mod the parent up. Better yet, get Slashcode to go back to supporting that +6 freak moderation state and then mod the parent up.
Not really, X.25 networks had gone global (International Packet Switch Stream) at a time the Internet was still purely an American toy. The Internet became global because the rest of the world had got there first - hardware-wise, at least. All the early transatlantic links were IPSS lines re-purposed, as was all the early European Internet capability. The Americans got the software side first.
Since the modern Internet is a marriage between software and hardware, and not one or the other alone, the only fair conclusion is that it was a global invention with no nation being able to claim credit for being truly first.
I thought "remote jamming" meant jamming of remotes. Which would be awesome to do, next time the neighbors channel-surf at high volume.
Not only further along the learning curve, but further along in mass availability. Mass availability = low cost. Low cost = more attractive to customers. I thought America had learned (the hard way) what happens when you ship inferior, expensive products after their car industry collapsed and Japan pwned them. I also thought America had learned (the hard way) what happens when you start behind your competition after they were thrashed by the USSR in the early space race. Catching up was damn expensive.
But apparently not.
Address shortages are a very, very, very tiny, miniscule fraction of IPv6. If IPv6 was about address shortages, the IPng working group would have adopted TUBA.
You seem unwilling to even recognize any of the other features of IPv6:
Built-in device mobility
Don't even think of coming back with "but nobody uses these" - nobody was driving until the car was adopted either. Things have a habit of not being used when they're not available. When they are available, they are used. It's as simple as that.
IPv4 is inherently insecure. IPv4 is inherently immobile. IPv4 is inherently non-extensible.
IPv6 is inherently secure. IPv6 is inherently mobile. IPv6 is inherently extensible.
Now, tell me which makes the most sense for mobile devices?
^get there first^own ICANN and enough critical infrastructure to demand whatever the hell they like, no questions asked, regardless of any actual needs