[governance] Uni.X to Uni.X .NETworking - The 10 Bit Length Field

[Jim Fleming]

Uni.X to Uni.X .NETworking - The 10 Bit Length Field

Note, with the 10 bit length field the value in the field will INCLUDE the
length of the 160 bit packet for backward compatibility.

"The next 16-bit IPv4 field defines the entire datagram size, including
header and data, in 8-bit bytes. The minimum-length datagram is 20 bytes and
the maximum is 65535."

That implies that code points (distinct values) 0 to 19 are not legal or
defined or programmed, etc.
[Legal is an awful word because the people connecting the devices are
writing the code that makes them work. Governance groups appear to want to
try to make writing code illegal.]

In stream-lined packet design and code bloat reduction it is not wise to
have bits sitting around undefined. One reason is that it is a waste of
bandwidth to be sending bloated fields that are not used. In a real network,
with large packet volumes bandwidth matters. It also matters in places
on Earth which are hard to reach. It costs a lot of money to send bits to
some places and the
people there do not often have the money to pay their half of the
connection. Some claim, they
should pay for their connection all the way to the big island. With
wireless, it becomes a moot
point because people meet in mid-air or both people pay for the entire
distance.

One easy way to handle the 0 to 19 code points is to use them to specify
lengths of multiple
fixed size blocks that follow the header. Unfortunately, that can move back
down the slippery
slope of fragmentation and back toward the realm of hacking via stack
overflows, etc. A
modified approach appears to be to use them as both a length specifier and
op-code for the
new features surrounding the DHT storage and the virtual LAN to LAN
bridging. If the code
is small and tight and well-tested, that should be an easy place to trigger
it.

For a complete commercial solution, all of the bits have to be defined. If
bits remain as reserved
or research bits then code forks and net forks will occur and the costs and
churn can make a
commercial offering not viable, or the consumers will end up paying for
research over and over
as they will likely see happening with the up-coming roll-out of the U.S.
Government's new
Internet Protocol coming soon from the DOD. You will like it and will pay a
fortune for it.
Kids will have WIMAX and better technology for free. Isn't governance
wonderful ?



_______________________________________________
governance mailing list
governance at lists.cpsr.org
https://ssl.cpsr.org/mailman/listinfo/governance