At 1/14/2011 10:15 AM, Steve Barnes wrote: >Fred, I thought about posting back to you all day yesterday and >finally decided to.
Glad you did. I don't mind taking unpopular views. >I agree that there will always be some sort of IPV4 out and >available. They are constantly trying to get universities who were >given huge blocks that have used less than 5% to relinquish parts of >it. But as Tom DeReggi has stated that there is other benefits. I >am not excited about being a bleeding edge adopter but I am looking >forward to more training on this issue and being prepared for when >there is a benefit for my clients. WRT v4, since v6 lacks compatibility, we're stuck preserving v4 for everyone for a long time, so we should expect to use more CGNAT, and more efficient address assignment rules. This isn't necessarily a bad thing. I'm not so sure that these other benefits are real, or require v6. I saw how IPv6 was created, and what the rules were at the time. It was a very sorry process. They had previously adopted a much better IPv7, but misbehaving children on the IETF made total arses of themselves (Lyman was getting about 70 phone calls *per hour*) and convinced IAB to reopen the issue. (Specifically, Vint changed his vote.) Their objection was purely poltical; IPv7 (TUBA) was based on IS8473, CLNP. And thus it was tainted, even though CLNP was the *good* part of the OSI program. The good people were then frustrated and left. The B-team put IPv6 together, starting with a silly rule that it should only fix the address space problem, not any fundamental architectural issues in IP (some of which were addressed by TUBA). So 17 years later, in a very different world, we have a very costly proposal with very limited benefits. I am (not here, but in other fora) proposing that we migrate away from TCP/IP per se and towards a newer protocol suite. What I'm backing is simpler than migrating to v6, coexists better with v4, and offers much more real benefits to its adopters (user and ISP alike). >The reason I take this stance is I have been in the computer >industry for 26 years. I know almost ever DOS command there is and >can still write a pretty mean batch file menu system if needed in a >pinch. One day my largest client at that time with 100 workstations >and the new Novel 2.15 server asked me what I thought about this new >Windows 2.86 software. I told him that it was all a fad why would >you want to rum more than lotus 123 and WordPerfect. When Windows >3.0 came out I got a copy and started playing with it and I thought >I might be wrong. I setup a meeting with that large company and >told them I was wrong. They informed me that they already knew that >and due to my short sidedness they had just signed a service >agreement with another company. I lost a company that I had made >$150K off of the previous year. I vowed to never look at future >possibilities the same. Ironically, IPv6 was designed when Windows 3 was bleeding edge, Word Perfect dominated, and Novell was the king of networking. IP itself is older than MS-DOS. IPv6 is sort of like adding LIM expanded memory (remember that?) to DOS. It handles bigger data tables, but it's still DOS. Yes, customers may ask for it, so you may be stuck for a while supplying it, but that's no reason to embrace it as The Solution or spend a lot on it. >Y2K was a bust but I made lots of money giving lectures telling >people that I had no idea what was going to be happening but that >all organizations needed to plan for emergencies and have back plans >whether it was Y2K, a fire, an Ice storm, or a tornado. > >Same goes with IPV6. I am not sure what will happen or if it even >will. But I need to have a plan to be ready no matter what >comes. The federal government has set a directive to make all their >networks IPV6 compliant by next year I believe. So if I want to be >able to service their traffic then I have to have it. I remember the 1985 GOSIP requirement too. Government procurements had to be OSI compatible. So yeah, people made money selling it. But nobody actually used it... -- Fred Goldstein k1io fgoldstein "at" ionary.com ionary Consulting http://www.ionary.com/ +1 617 795 2701 -------------------------------------------------------------------------------- WISPA Wants You! Join today! http://signup.wispa.org/ -------------------------------------------------------------------------------- WISPA Wireless List: wireless@wispa.org Subscribe/Unsubscribe: http://lists.wispa.org/mailman/listinfo/wireless Archives: http://lists.wispa.org/pipermail/wireless/
