Hi, Michiel!

20 Nov 02 09:21, Michiel van der Vlist said to Alex Shakhaylo:

 MV>>> anyone is upon you. You prove to me that there is no software
 MV>>> (oter than makenl) that has a problem with it.

 AS>> But you know there is no way to prove it. I should test all
 AS>> the existing software and that is impossible.

 MV> Ah, you are finally coming to see the light.

This light was seen from the very beginning, but it also is seen that
futher improvement demands some price to be paid.

 MV> Indeed we can not test all exiting software for the simple reason
 MV> that we have no idea what's out there. The fact is that when we make
 MV> a change that is not within present specs, we have no idea what and
 MV> who will be affected by it.

But according to present specs -Unpublished- does mean an absence of a
dialable phonenumber.

 AS>> But I have a way to avoid the problems, this way is the patcher.

 MV> So yiou claim to have a solution. What was the problem again?

 MV> O, yes, some people have a problem with the "stigmata" of the Pvt
 MV> keyword which indicates that they are ION.

I see the different problem. Pvt keyword should indicate that node is not
contactable in any way, not only to POTS nodes.

 MV> Well, I have a much better solution for that. One that is /guaranteed/
 MV> not to break any legacy software.

 MV> Let's make a patcher that removes Pvt from all nodes that have one or
 MV> more valid IP flags.

 MV> So everyone that feels he needs compensation for whatever he feels
 MV> needs to be compensated can run this little programme to have his
 MV> stigmata removed from his private copy of the nodelist. How's that?

The answer depends on the number of needers. The nodelist should be patched
for those who are less in number. 

 AS>> It will be written in plain ansi ‘, I hope we will find a
 AS>> person to compile it for MAC.

 MV> And the Atari? And the Lisa? And the Acorn Atom?

I like DD and SL idea about running the patcher further upstream.

 MV>>> 2) How are you going to handle the case where there is already
 MV>>> something in the first field, like Hub or Host?

 AS>> This is more interesting question. It seems Hub and Host
 AS>> conception must be thought out, but till then we should not
 AS>> have Hubs and Hosts without the real phonenumber.

 MV> But in that case, your operation is purely cosmetic. It serves no
 MV> purpose whatsoever except to create an extra step that will make
 MV> nodelist distribution more complicated with the inherent chance of
 MV> extra errors.

As far as we are talking about the  patcher for a POTS mailer we can
safely change first field of all the -Unpublished- entries with pvt.
Anyway POTS mailer cannot make use of -Unpublished- entry whether it
is Hub, Host or anything else.

 AS>> It is not useless as far as we don't have anything else except
 AS>> pvt, hold and down in the first field followed by -Unpublished-
 AS>> phone in the current nodelist.

 MV> Yes, it is useless. You are staring yourself blind on those three
 MV> characters. Removing them in the distribution list and then later
 MV> inserting them in the working copy is a purely cosmetic operations
 MV> that serves no purpose whatsoever except to satisfy the genitally
 MV> challenged.

I don't agree. Theoretically we can have ION hosts and ION hubs and
new software may utilize this information.

 AS>> It is not universal solution, but it is step forward.

 MV> Adding unneeded conversion steps in the nodelist distribution is not a
 MV> step forward.

For IONs it is step forward, for POTS it is not a step at all. But
number of IONs will increase and thus profit will grow.

Bye, Sinc, Alex

--- GoldED/W32 3.0.1