Subj : proposed new nodelist                                    [2]
To   : Malcolm Miles
From : Jasen Betts
Date : Tue Jul 16 2002 01:25 pm

Hi Malcolm.

15-Jul-02 16:16:42, Malcolm Miles wrote to Jasen Betts

 MM> On Jul 14, 2002 Jasen Betts wrote to Peter Knapper:

 PK>>> My current messages have been aimed at something completely new
 PK>>> JUST for IP nodes, leave the PSTN side of things alone as they
 PK>>> are

 JB>>> Leave them broken? why?

 PK>> The current Nodelist is NOT broken to me.

 JB>> you are not all of fidonet...

 MM> It is not broken for me either. Seeing that the current nodelist
 MM> format has kept Fidonet running for at over 10 years, including
 MM> supporting IP nodes, I don't think you could say it is broken.

It's brokren because it provides bogus contact information in some
instances.

 MM> Granted, although the current format does support IP nodes, it is
 MM> a kludge and what we are looking for here is a more ideal
 MM> solution. If we can use standard Internet protocols and services
 MM> then I believe that is the correct way to go. No point in
 MM> inventing our own protocols and services which, in reality, nobody
 MM> is going to code up

 JB>> That's a good thing, but you're still relying on a nodelist of
 JB>> sorts, all you've trimmed from it are the capability flags. if
 JB>> you're a dedicated anarchist you may want to to a less structured
 JB>> model like gnutella for instance. - something like that could
 JB>> probably be modified for ftn

 MM> Gnutella-type models are based on content, not nodes.

What I mean is some sort of host-less peer-to-peer information service.
practically indestructable.

 JB>> Is this about some *Cs disallowing certain nodelist entries
 JB>> (mainly those notconforming to FTS0005) they claim to do this
 JB>> because those "bad" entries cause problems with some systems.

 MM> And why is this a bad thing?

I dodn't say it was (and I'm not convinced that it is).
I was asking peter if that was one of his motives.

 JB>> This is why I want to transform the nodelist into something that
 JB>> can contany any useful information and is fututre proof and easy
 JB>> to extend. then the won't have a technical argument against those
 JB>> nodes.

 MM> Totally disagree, a technical argument should be the only argument
 MM> used to disallow nodes. If a node is causing problems on the
 MM> network, they should be denied access to the network until they
 MM> fix the problem

that makes sense...  what I mean is whole debate over 000- "phone numbers"
and other related topics.

 PK>> Of course it does. If the node is NOT listed as crash, then its
 PK>> the callers fault for attempting to do so when they are not
 PK>> supposed to. If I am unable to send mail direct to a node (for
 PK>> whatever reason), then I Host Route it via the NC/HUB

 JB>> There are HUBS (and NCs and atleast one RC) with no published
 JB>> POTS connection.

 MM> I believe that every Host node (and a host node does not have to
 MM> be the NC)

Yeah the NC just cracks the whip right (if he can convice some other node
to be host that's fime) as I read the "rules" the NC is "resposible" for
seing that netmail moves...

 MM> should have a POTS connection and should have the
 MM> capability to get a netmail to any of the nodes in the net.

yeah, haing nodes in a net that can't get mail from the host would be a
problem.

 MM> RCs don't have to route mail to their nets

They probably should if they have ION region independants...
(I expect you don't like them either)


There's a guy in zone 1 called Gregory_Deyss who seems to have found a way
to tunnel V32 over IP if the nodelist can be believed ;)

 -=> Bye <=-

---
 # Origin: Misery loves company, but company does not reciprocat (3:640/531.42)
 * Origin: Baddog BBS (1:218/903)

.