Text 2817, 75 rader
Skriven 2008-04-03 14:10:01 av Michiel van der Vlist (2:280/5555)
   Kommentar till text 2771 av Dieter Ringhofer (2:2476/14)
Ärende: Two related questions plus one to the candidates
========================================================
Hello Dieter,

On Wednesday April 02 2008 12:47, you wrote to me:

 DR>>> ICANN started to implement IP v6 last days.

 MvdV>> Good! So why do you expect this to be a problem for FidoNet?

 DR> Threre might problems with DNSs pop up all over the world next time.
 DR> This meets Fidonet as well as internet as medium.

Ok... I guess we will just have to wait and see then. But if te problems are
not specifically FidoNet related , there is A) not much that we can do about it
and B) others will jump in to solve it....

 MvdV>> problems for many systems. But I do not see any problems for
 MvdV>> FidoNet. It is my understanding that every internationalised
 MvdV>> domain also has an aka that conforms to the tradional rules.

 DR> AFAIK this is no automatic implementation (see example mentioned by
 DR> Carol in this thread). One either uses punycode or might have a
 DR> problem.

Only if someone insists o using the unencide UTF8 characters in the nodelist.

But frankly I do not see why this should be a problem at all. I can see why
poeple would want "müller" instead of "mueller" in their domain name for a
website. website addresses are for use by humans. Bu the nodelist prmarily for
machine use. Why not settle for a CNAME in ASCII that bears a reasonable
aproximation to the "real"one?

 DR> Flags alone are of help but, they might be insufficient.

 DR> Since long time I try find an optimal solution to reflect my own
 DR> system: - Connectable via POTS (ISDN, analog) - binkd, multiple
 DR> machines via multiple ports

I think the best solition for multiple machines that use the same type of
server is to use more than one node number.They would be the exception rather
than the rule and in thsoe cases an extra nodle number wuld be justified I
think. It is the easiest solution.

 DR>  - telnet (one machine only for security purposes) - ssh, multiple
 DR> machines via multiple ports - http, multiple machines via multiple
 DR> ports

Frankly I do not understand why anyone who is binkp capable would want all that
/for FidoNet/. FTS-1/EMSI over telenet is lousy, the timing constraints were
optimised for use over a phone line, over InterNet it works only with many
repeats. I suppose ssh is not much bettter and http, what is the added value?

 DR> You might want to say http is not important but, there is chance to
 DR> download fidonet mail. It can work like a file request. When being
 DR> identified you get your mail and your system can toss it.

Sorry, I see no added value when adding it to a binkp capable system.

Actually I think this phletoria of protocols is a liability rather then an
asset. It is detremental to communication. The more protocols, the less that
change that one has one in common. Unlees everyone carries them all or unless
there is at least one that is mandated. In either case, everyone would be
better of by agreeing to only use the best and dump the rest.

It is like languages. Different languages are a barrier for comunication.
Everyone would be better of in the long run if we copuld all agree on one
commomen language. Not going to happen in real life, I know, but with a
relatively small group of technicians there is a chanve.

In fact we see a tendency that binkp is a de facto langua franca...


Cheers, Michiel

--- GoldED+/W32-MINGW 1.1.5-b20070503
 * Origin: http://www.vlist.org (2:280/5555)