Hello Michael.

05 Nov 02 23:45, you wrote to me:

 RR>> That is rather foolish, tell me what IP software doesn't read the
 RR>> protocol flags and ignore the Pvt flag at the beginning? What IP
 RR>> software is going to screw up because of the Pvt flag?

 MG> Tell me what use is a nodelist flag if a mailer is supposed to ignore
 MG> it? Tell me what mailer works more efficiently, one that recognizes
 MG> /all/ the flags and acts on them, or one that ignores flags?

One that uses all that are described in the FTSC standards. Any intelligent
IP mailer that is written to follow these standards will ignore the Pvt flag
if an IP protocol flag is within the same entry.

 RR>> You have developers out there both past and present whose
 RR>> software is specifically designed to operate under these
 RR>> conditions yet all you can do is condemm it and promote listings
 RR>> that no software supports.

 MG> Who exactly decided that Pvt, -Unpublished- was an acceptable solution?
 MG> Did Charles Cruden, author of Internet Rex, have any say on that? Did
 MG> the Russian author of BinkP and BinkD have any say? Did the developers
 MG> of Argus, Radius, Olegon, or BeeMail have /any/ say on the use of the
 MG> Pvt -Unpublished- kludge?

Yes. Don't you read the FTSC docs which clearly state the members of the FTSC
who approved these standards and had them documented as such.

 MG> I'd be very suprised if /any/ of those developers thought it's a good
 MG> idea to have their mailers ignore nodelist flags, and I'll bet they also
 MG> think it's a bad idea to have dual meanings for nodelist flags. The FTSC
 MG> made the kludge before most of these IP mailers even /existed/, and made
 MG> it at a time when POTS nodes were the vast majority in this network. The
 MG> kludge was designed to protect POTS nodes and place their concerns
 MG> before the concerns of Internet Only Nodes. The developers of the later
 MG> mailers had to accomodate the FTSC documentation whether they liked it
 MG> or not, because the FTSC refused to or was unable to change the
 MG> standards. Thus the developers were /forced/ into making their IP based
 MG> mailers ignore nodelist flags.

Wrong again, IRex for example did not come into existence as binkp and ftp
capable until after the committee proposed the standards (back in 1997).
Charles was also a member of the FTSC when this proposal was issued as a
standard.

 MG>>> probably be so much easier to just put it in the phone number
 MG>>> field, why is it so *blamed* hard for POTS nodes to make a small
 MG>>> accomodation by agreeing that /any/ 000-* listing in the
 MG>>> phone number field will indicate IP only node, and thus agree
 MG>>> that all POTS nodes will block it out???

That usage is not a problem. However the use of the 000- prefix without a
static IP following it is a problem.

 RR>> Then why were you using 1-000 prefix and yet now advocate a 000-
 RR>> prefix?

 MG> I've now changed it, but IMO, either one works just as well. All that's
 MG> needed is a simple agreement. Why is it so hard to agree to reserve a
 MG> certain numerical sequence to mean non-connectivity of POTS mailers?

Good to see you changed it, I hope your IP is listed after the prefix. I
can't tell if it is as that node does not appear in this week's nodelist
here. If you remember thats all that was asked originally, to get rid of the
1-000-000-000 and replace it with the standard, of which 000- is
one.

 RR>> No, its time to activate a central organisation of developers who not
 RR>> only can propose standards but also work towards adjusting and
 RR>> creating software to support those standards. It is useless promoting
 RR>> a self-styled method if not only the majority don't accept it but also
 RR>> no software supports it.

 MG> At the best of times, the FTSC was a political body, and excluded many
 MG> bright developers of Fidonet software. An exclusive club of a few
 MG> Fidonet members who have developed a few programs is of no help, and
 MG> causes more problems that it solves. Fidonet needs to move to a far more
 MG> open approach; the open source movement has proven it's value, and
 MG> Fidonet ought to adopt similar development methods. A small group
 MG> forcing their opinions on the rest of this network is counterproductive.

I hope you inform Charles Cruden, Tobias Ernst and a host of others that you
consider them politcal people and the few odd bits of software they wrote are
of no consequence. I am sure they would appreciate your views.

However I do agree with you that the FTSC should look towards becoming an
open source movement which allows the input of various developers on various
projects without relying solely on one or two to keep a section going.

 RR>> I prefer to travel the path and listen to the developers rather
 RR>> then telling everybody what I think the ideal method should be.

 MG> The problem is that the FTSC /stopped/ listening to the developers a
 MG> /long/ time ago, and /insisted/ on continuing to promote the use of
 MG> ancient kludges that have caused more problems than they've solved, and
 MG> they also failed to look for any /real/ solutions to the problems that
 MG> IP connectivity poses to the nodelist. That is the main reason why the
 MG> FTSC is dead in the water today.

Actually the problems stemmed from part of fidonet which refused the changes
the FTSC wanted to make. Too much opposition stifled the committee. The
problem lies in that the basic old software is stretched to its limit and
theres little or no way to keep expanding the possibilities of fidonet
without breaking all that software. That is the problem, the developers know
that to make things better a line is going to have to be drawn, and if your
software is too old and behind that line, then it will become useless.

 RR>> I am quite prepared to use the standards they have proposed and
 RR>> agreed upon and use software that supports those standards. With
 RR>> those standards then it is possible to connect to anyone with no
 RR>> manual intervention.

 MG> Standards are of no use when they do not recognize the realities of the
 MG> network and the realities of the software it's users are using to
 MG> participate in it today. A "standard" that was designed
for old software
 MG> that any given node may never use or may never even have /heard/ of, is
 MG> no "standard" for that node.

But what harm has it done to the software you use to change your IP phone
number to the accepted 000- prefix method. Actually, what affect would it
have on your IP node if you did use the Pvt kludge in the entry?

 RR>> Btw, I notice your /10 address dissappeared from the last
 RR>> nodelist. You gone back to being just a POTS node?

 MG> No. It probably was related to Janis' nodelist compiler problems.

Most probably, well at least next week you'll have a standard entry.

Cheers,
                Rick

... Was Jimi Hendrix's modem a Purple Hayes?

--- GoldED+/LNX 1.1.5 - Debian/GNU