Anyone have - Top Ten Stupid Networking Ideas

possibly knows the answer.

--Frank

Apologies, didn't read with brain switched to "on".

--Frank

:)

Weird thing is that I distinctly recall reading it years ago, probably on usenet, but there is no trace of it now.

BTW, I don't want the published paper just the raw list.

Go to

Use "top ten stupid network ideas seifert group:comp.*" as the search term (I think that will go directly) and the dig. It may not give you the entire list in one place, but it will at least give you three of the entries.

rick jones

Thanks I found it now. Not exactly sure why I failed to find it previously. As you say not the entire list but I got what I wanted after reading the thread.

I'll do you one better--send me an email to "usenet at richseifert dot com" and I will send you the original PowerPoint presentation from that talk. It was given as the keynote speech at the 2001 IEEE Local Computer Networks Conference at the University of South Florida.

Looking back on it today, I was right on some (e.g., ATM, ISO Protocols, etc.) wrong on others. FYI: When the talk refers to "VoIP", it really means "Voice over the Internet", i.e., using the Internet as a replacement for the telephone network. I still think that is a dumb idea (and has not really taken off); the use of VoIP within *private* networks (including those operated by the traditional telephone companies) has been fairly successful, but that wasn't really my emphasis back then.

-- Rich Seifert Networks and Communications Consulting 21885 Bear Creek Way (408) 395-5700 Los Gatos, CA 95033 (408) 228-0803 FAX

Send replies to: usenet at richseifert dot com

My interest was piqued when someone asked about getting a home DSL IPv6 ready and I recalled that you had commented on IPv6 a while back:) Not really sure where things are now but my impression is that IPv4 and NAT will be with us for a very long time to come.

NAT has turned out to be a very fine security feature but it does of course limit communications, for example if you want your fridge to be on-line then NAT is a real nuicance.

I have the idea that the major ISPs are all gearing up for IPv6 on their backbones but I am not sure.

I work for an unnamed wireless ISP and can tell you that we've had a team dedicated to IPv6 since about 2002/2003 and they're barely any closer to deploying anything now than when they started. It's not as easy as it might seem. As usual, the devil is in the details.

One of my previous coworkers jumped ship a few years ago and now works for a giant cable ISP. He reports basically the same thing from his side of the fence.

Out of curiousity, can you comment on some of those devils?

rick jones

I need to remain as vague as possible, so I'll just say that as we peel back each layer of the onion, we find more and more special cases that need individual consideration and handling. All of this is quite unlike a pure IPv4 environment, for us anyway, where data is data and it's all basically treated the same.

We're finding that each vendor (of ALG's, for example) has their own ideas of how their product should work, where it should be placed in our network, etc. Accommodating one product causes a ripple effect, and in turn more ripples as more changes are considered, and it's been a never ending loop. *Something* would have been done by now, if we were indeed fully exhausted on our IPv4 address allocation, but we keep getting new slivers allocated, so things drag on.