Why is IPv6 so complicated?
-
IPv4, so simple, just four numbers from 0 to 255 giving the highest IP address as: 255.255.255.255 Some reserved, such as 192.168.0.0-255 or 192.168.1.0-255, and a few others I can't remember. So then, we run out of IPv4 addresses and we need another. If I had to do this, I would have first had "IPv5" with five sets: 255.255.255.255.255 - the IPv4 would be contained with 255.255.255.255.0, which is a heck of a lot more addresses! After that just 255.255.255.255.255.0, etc. Compatibility with older IPv4 addresses is simple, all IPv4 addresses are just "IPv5", "IPv6" or "IPv7", etc. with the last one, two or however many sets being 0. Simples! An IPv4 address or 123.234.123.234 would be represented in "IPv8" as 123.234.123.234.0.0.0.0 without any further complication! Why did they have to come up with a new, complex system that is not backwards compatible and is therefore much harder to implement and therefore not being taken up as quickly as it could have been? Your thoughts and deeper knowledge appreciated. [update] The zeroes could be leading instead of trailing, eg. 0.255.255.255.255
- I would love to change the world, but they won’t give me the source code.
Is this similar to "32 bit CPU is enough, why 64 bit is so complicated?" The point is to enable every sand grain to be connected: geology - Can every grain of sand be addressed in IPv6? - Skeptics Stack Exchange[^] Like it or not, China is progressing on IPv6 very fast, simply for your 5G phones and IOTs.
-
System Administration in a private enclave; you have to know your DC/Realms and gateways, and you're a fool if you don't.
"Never attribute to malice that which can be explained by stupidity." - Hanlon's Razor
-
Is this similar to "32 bit CPU is enough, why 64 bit is so complicated?" The point is to enable every sand grain to be connected: geology - Can every grain of sand be addressed in IPv6? - Skeptics Stack Exchange[^] Like it or not, China is progressing on IPv6 very fast, simply for your 5G phones and IOTs.
Who claims that 32 bit is simple? I know it can be, but when I was teaching elementary computer architecture (as a software man, seeing architecture from a software man to software students) trying to explain the IA-32 mess has kept me forever away from attempting to learn IA-64. I had lost the architecture battle: I was fighting for MC68030 based *nix workstations (so it is long ago!), but the department head overruled the decision preferred by the majority of the teaching and technical staff, demanding that we go for Intel. In my lectures, I did sneak in some examples from non-Intel architectures, but couldn't give the students exercizes on anything but IA-32. If Intel did a thorough cleanup with IA-64, it might be far easier to understand. I don't know. IA-32 gave me so many frustrations that I never wanted to touch another IA-whatever architecture. The only thing that is impressive about IA-32 is how they can make that mess spin around as fast as it does. You would think that interpreting hodgepodge of instruction and addressing formats, the MMS and everything, would take so many clock cycles than any other achitecture could easily beat it. The fact that IA-32 essentially knocked down all competiton proves one thing: The hardware implementation guys at Intel must be super clever. I wouldn't say the same about those guys drawing up the x86 architecture. But that was long ago, of course.
-
Who claims that 32 bit is simple? I know it can be, but when I was teaching elementary computer architecture (as a software man, seeing architecture from a software man to software students) trying to explain the IA-32 mess has kept me forever away from attempting to learn IA-64. I had lost the architecture battle: I was fighting for MC68030 based *nix workstations (so it is long ago!), but the department head overruled the decision preferred by the majority of the teaching and technical staff, demanding that we go for Intel. In my lectures, I did sneak in some examples from non-Intel architectures, but couldn't give the students exercizes on anything but IA-32. If Intel did a thorough cleanup with IA-64, it might be far easier to understand. I don't know. IA-32 gave me so many frustrations that I never wanted to touch another IA-whatever architecture. The only thing that is impressive about IA-32 is how they can make that mess spin around as fast as it does. You would think that interpreting hodgepodge of instruction and addressing formats, the MMS and everything, would take so many clock cycles than any other achitecture could easily beat it. The fact that IA-32 essentially knocked down all competiton proves one thing: The hardware implementation guys at Intel must be super clever. I wouldn't say the same about those guys drawing up the x86 architecture. But that was long ago, of course.
-
Translate your post to hexadecimal, you will see that IPV4 and your proposal are not straightforward as they look in decimal. Try to think as microcontroller.
-
In the case of DNS failure, it has saved me quite a few times when DNS crashed, and DNS on a large scale can die too look at the DYN Denial of service attack that affected most of the east coast, also for me numbers seem to be easier to remember than names or strings depending on the complexity so IPv4 was simple enough to remember.
-
IPv4, so simple, just four numbers from 0 to 255 giving the highest IP address as: 255.255.255.255 Some reserved, such as 192.168.0.0-255 or 192.168.1.0-255, and a few others I can't remember. So then, we run out of IPv4 addresses and we need another. If I had to do this, I would have first had "IPv5" with five sets: 255.255.255.255.255 - the IPv4 would be contained with 255.255.255.255.0, which is a heck of a lot more addresses! After that just 255.255.255.255.255.0, etc. Compatibility with older IPv4 addresses is simple, all IPv4 addresses are just "IPv5", "IPv6" or "IPv7", etc. with the last one, two or however many sets being 0. Simples! An IPv4 address or 123.234.123.234 would be represented in "IPv8" as 123.234.123.234.0.0.0.0 without any further complication! Why did they have to come up with a new, complex system that is not backwards compatible and is therefore much harder to implement and therefore not being taken up as quickly as it could have been? Your thoughts and deeper knowledge appreciated. [update] The zeroes could be leading instead of trailing, eg. 0.255.255.255.255
- I would love to change the world, but they won’t give me the source code.
politics man (read economics). they make everything anew so that common people won't know how things work anymore. it is not allowed for people to know all, how the car, the computer, the internet, the washing machine... works. you have time to learn only one thing in life and for the rest of things you can only be a consumer. that thing moves the business, by making new things that brake how old things work. and those new things are going to be explained by a mass of new technical abbreviations and acronyms that have almost exactly the same meaning as the old ones, but these are brand new. and the hipsters are going to love this and show off. at the same time science is moving in the opposite direction. scientists are trying to discover the unified natural law that explains everything.
-
politics man (read economics). they make everything anew so that common people won't know how things work anymore. it is not allowed for people to know all, how the car, the computer, the internet, the washing machine... works. you have time to learn only one thing in life and for the rest of things you can only be a consumer. that thing moves the business, by making new things that brake how old things work. and those new things are going to be explained by a mass of new technical abbreviations and acronyms that have almost exactly the same meaning as the old ones, but these are brand new. and the hipsters are going to love this and show off. at the same time science is moving in the opposite direction. scientists are trying to discover the unified natural law that explains everything.
-
Mark_Wallace wrote:
What annoys me most is that when I ping a domain name to find out its IP address, it sometimes comes up with an IPv6 address -- and there's no way in the fruggin' world I'm going to remember that!
All you need to remember is a single number: 4 ping -4 localhost Best Wishes, -David Delaune
Is that right? Cheers! I'll give it a try, next time a ping returns an ipv6 address.
I wanna be a eunuchs developer! Pass me a bread knife!
-
Hey, get a real man's job, and you'll find out.
I wanna be a eunuchs developer! Pass me a bread knife!