New ipv4 idea

[Linus No Beard]

All public IP addresses are bought in blocks that have a net mask of 255.255.255.0 as organizations that need more can just buy more blocks

7 ranges

    Range
A   1.0.0.0       -    9.255.255.255       Public addresses
B   10.0.0.0     -   10.255.255.255      Private addresses only range to have a default net mask of 255.0.0.0
C   11.0.0.0     -   126.255.255.255    Public addresses
D   127.0.0.0   -   127.0.0.255            Loopback

E   127.0.1.0   -   223.255.255.255    Public addresses

F   224.0.0.0   -   239.255.255.255    Multicast
G   240.0.0.0   -   255.255.255.255    Public addresses

This system frees 310,825,234 addresses while still being limited to 2^32 addresses

 

This system also fixes the waste of having 16 million loopbacks as that is just wasteful and it got rid of 192.168.0.0 - 192.168.255.255 as private addresses due to 10.0.0.0 - 10.255.255.255 being better suited for that goal.

[Lurick]

Why touch legacy stuff and break it instead of looking forward and focusing on a move to IPv6?

There are single companies that own entire /8 address spaces that still have them, you'd be better off getting those freed up than trying to reallocate address space like this.

[Eigenvektor]

Yeah, good luck updating millions of legacy devices. If we can do that change, then we can also change to IPv6 and solve the issue for good.

 

10 minutes ago, Linus No Beard said:

and it got rid of 192.168.0.0 - 192.168.255.255 as private addresses due to 10.0.0.0 - 10.255.255.255 being better suited for that goal.

There are actually three private ranges

10.0.0.0/8

172.16.0.0/12

192.168.0.0/16

[Linus No Beard]

2 minutes ago, Eigenvektor said:

Yeah, good luck updating millions of legacy devices. If we can do that change, then we can also change to IPv6 and solve the issue for good.

 

There are actually three private ranges

10.0.0.0/8

172.16.0.0/12

192.168.0.0/16

why 172.16.0.0/12 also yes my system would get rid of that because it is not needed when we have 10.0.0.0/8 that is 16 million private address and that might be a bit much

[Linus No Beard]

4 minutes ago, Lurick said:

Why touch legacy stuff and break it instead of looking forward and focusing on a move to IPv6?

There are single companies that own entire /8 address spaces that still have them, you'd be better off getting those freed up than trying to reallocate address space like this.

I just wanted to see if with the power of hindsight i could do better that IANA.

[Eigenvektor]

6 minutes ago, Linus No Beard said:

why 172.16.0.0/12

I don't know why, but that's what it is.

 

6 minutes ago, Linus No Beard said:

also yes my system would get rid of that because it is not needed when we have 10.0.0.0/8 that is 16 million private address and that might be a bit much

Or you could take away millions of IPs from AT&T, Apple, Ford, …, Comcast who each own 16 million public IPs. And millions more from the DoD. No need to change which addresses are public and private, simply make those addresses available to other companies.

[Lurick]

As for why each space for private addresses was chosen:

 

The Internet Assigned Numbers Authority (IANA) has reserved the following three blocks of the IP address space for private internets:

     10.0.0.0        -   10.255.255.255  (10/8 prefix)
     172.16.0.0      -   172.31.255.255  (172.16/12 prefix)
     192.168.0.0     -   192.168.255.255 (192.168/16 prefix)

We will refer to the first block as "24-bit block", the second as "20-bit block", and to the third as "16-bit" block. Note that (in pre-CIDR notation) the first block is nothing but a single class A network number, while the second block is a set of 16 contiguous class B network numbers, and third block is a set of 256 contiguous class C network numbers.

 

[wseaton]

RFC 1918 FTW

 

If you want to improve the universe, un-brainwash all those bondage and discipline OCD network admin types that insist on subnetting a witty bitty Class C network's private address space into it's absolute restrictive range.

 

Me: "Dude, use 255.255.0.0.....nobody cares....it's your address range.....it's private....I'm sure your switch can handle the extra octet ..... / sarcasm"

"him: That's not what I was taught in my engineering class. internal network ranges need to be strictly defined"

Me:"Dude, this is a small comany....not Boeing. I bet you were taught you need to VLAN each 254 address space to limit broadcast traffic"

"him: well....yeah....that to"

 

 

[Needfuldoer]

2 hours ago, wseaton said:

Me: "Dude, use 255.255.0.0.....nobody cares....it's your address range.....it's private....I'm sure your switch can handle the extra octet ..... / sarcasm"

"him: That's not what I was taught in my engineering class. internal network ranges need to be strictly defined"

Me:"Dude, this is a small comany....not Boeing. I bet you were taught you need to VLAN each 254 address space to limit broadcast traffic"

"him: well....yeah....that to"

And remember, you cannot use the .0 and .255 addresses in those big subnets, "because you can't use those addresses".

 

Someone told me this unironically. 

[Eigenvektor]

16 hours ago, Needfuldoer said:

And remember, you cannot use the .0 and .255 addresses in those big subnets, "because you can't use those addresses".

I guess someone once told them you can't use .0 and .255 in 192.168.1.0/24 and neglected to mention why or how the netmask figures into this.

[Guest]

As long as I can keep my localhost, I'd be okay with it. 

[Linus No Beard]

8 minutes ago, fpo said:

As long as I can keep my localhost, I'd be okay with it. 

You only need one tbh but i gave a whole 254 (because 0 and 255 cant be used just like normal ipv4)