Hello Steven,

 MvdV>> Yes, what's the problem? The nodelist isn't meant to be read by
 MvdV>> humans like a novel. It is meant to be read by /machines/.

 SH> The nodelist is an information source for both machines and
 SH> persons.

No, it is not. read this form the header of the latest nodelist:

;A  .. The FidoNet NodeList is compiled so that computer systems within FidoNet
;A     may communicate with each other.  Use and intra-FidoNet distribution

Note: *computer systems*

And also note:

;A Other use without express written consent is forbidden.  For other use,
;A please contact Fido Software, Box 77731, San Francisco, CA  94107  USA.

 SH> I regularly read the raw nodelist to see who is
 SH> where, what their status is, and so on.

Did you obtain the permission of Fido Software for this "other
use" that is explicitly forbidden? 

 MvdV>> It is as accurate as it needs to be. The machines understand it
 MvdV>> perfectly. Once again: it is not English, it is computerese.

 SH> Again, it is read by persons who normally know what the
 SH> various abbreviations mean.

That they /think/ they knwo what they mean and hence arrive at a wrong
conclusion when they use the tool of the nodelist for a purpose that it was
never intended for is their own fault.

 SH> Now you can change "CM" from "continuous mail" to
 SH> "chocolate malt" but it will lose its meaning in real language.

That is only a problem for those that do not understand that the nodelist
is not written in English but in computerese.

 SH> What use a computer makes of it is not my problem.

What humans that do not undestand it is not English make of it is not the
problem of those that generate the nodelist.

 MvdV>> Meanings that are meant to be for machines are changed by
 MvdV>> programmers and that is precisely what happened. Ip mailers
 MvdV>> ignore the Pvt keyword and for POTS mailers it means "do not
 MvdV>> dial, reroute to the host".

 SH> Programmers serve people and not vice versa.

That's real life. In this Role Playing Game it is different. Here the
programmers can't be ordered around by the bosses to do things they
consider technically unsound.

 SH> The beauty of the nodelist was that a lot could be done with
 SH> the "raw" nodelist and a lot by the "compiled" nodelist.

I don't see how that changes by redefining the meaning of a keyword.

 MvdV>> For good reasons. The field is meant for /telephone numbers/ and
 MvdV>> so lots of programmes expect to find numbers or the word
 MvdV>> "-Unpublished-" anything else will confuse existing software.

 SH> Apart from the fact that you have already said that
 SH> anything could be put in there because it is a "meaning for
 SH> a machine",

I also added the provision of "downward compatible".

 SH> there is nothing that suggests that my Xlaxnode
 SH> would have any difficulty in compiling a nodelist with what
 SH> is in there now anymore than it has had with compiling any
 SH> nodelist since 1992.

Good for you, but since the specs say it should either be a nummeric
sequence or -Unpublished- one can not expect ALL nodelist processing
software to be so forgiving.

 MvdV>> That has been done. Put the domain name in the BBS name field -
 MvdV>> IP mailers can read it from there - and list the node with a Pvt
 MvdV>> keyword and an -Unpublished- telephone number so that POTS
 MvdV>> mailers will not attempt to dial it.

 SH> Have you not heard of the expression "dialing table"?

I have. How do I make a mailer not dial *anything at all* by merely
adapting a table that translates one number into another?

 MvdV>> Sorry, no deal. In the real world laywers and managers can order
 MvdV>> technicians to implement bad ideas. That does not fly in a hobby
 MvdV>> environment that is run by unpaid volunteers.

 SH> Had we been more professional, it is quite likely we would
 SH> have implemented better ideas.

De beste stuurlui staan altijd aam de kant. ;-)

 SH> Instead, we got bad ideas implemented in a mediocre fashion
 SH> while the brilliant stars simply left.

Is that so?

There is nothing like 20/20 hindsight you know. So maybe the choices that
emerged fifteen years ago weren't the best possible in retrospect.
Nevertheless those were the choices made and we can't just reverse that
without breaking a lot of things and leave people out in the cold.

 SH> Have you ever asked yourself why people like Randy Bush
 SH> or Vince Perreillo are no longer part of this hobby?

Because they got fed up by lawyers and managers trying to boss them around
just like they would if this was not a hobby and they were paid for it?

If that is the reason I can sympathise with them. I sometimes feel that way
when someone that can't programme himself out of a wet paper bag that is
open on three sides says "you guys should have come up with a decent
solution long ago".

Oh, c'mon Steven, what line of questioning is this?

I don't know why Vince or Randy left. I never met them in person and their
goodbye messages were to short to be conclusive. Maybe it was for the
reasons you suggest. Maybe not.

I do know however why Henk Wevers left. To fresh up your memory Henk was
the founder of the first net outside the Americas (net 500) and he is also
the author of the Ducthie mailer.

I was one of the 8 people present at the net 500 hub meeting of Nov 11,
1997 when he announced his departure. Actually I was the one to write the
minutes:

 > 0) Evert opent de vergadering om 19:02.
 >
 > 1) Vertrek Henk Wevers.
 >
 > Henk Wevers kondigt aan te stoppen met FidoNet. Hij verwacht
 > vanwege zijn verhuizing begin komend jaar onvoldoende tijd en
 > gelegenheid te hebben. Voor zover een gelegenheid passend kan
 > zijn voor een afscheid en overdracht van de fakkel, is deze het dan.

For those that can not read Dutch: Henk announced his departure from
FidoNet. He expects to have insufficient time to fulfill his duties due to
him moving into another house early next year.

Of course that is only the part that made it into the record. Off the
record we talked about it for a while. It simply ammounted to Henk loosing
interest and finding other priorities in life. My guess it that that goes
for most of those that have left us.

 MvdV> It means nothing. He can not possibly test /every/ piece of
 MvdV> software in use in FidoNet if only because no one knows what
 MvdV> is out there.

 SH> Fidonet itself rests on standards.

He isn't testing against the standard. On the contrary he is testing how
that software behaves /out of standard/.

That way he may get answers on how that particular software behaves, but it
will tell you nothing about the untested sofware.

 SH> If he or others test a number of pieces of software which
 SH> comply with the standards and they work, that may be the end
 SH> of the story.

It can not be because there is no way to know what is all out there, let
alone test it.

 MvdV>> That is assuming there is any way at al to do it properly.
 MvdV>> There are mailers where this is not possible. They can be
 MvdV>> configured to dial something else but they can not be
 MvdV>> configured to not dial anything at all.

 SH> Michiel, Michiel.  They can be configured to not dial
 SH> "-000".  This is not rocket science.

Wrong. Rocket science deals with the possible. There are mailers that can
NOT be so configured.

(3207)  Tue 3 Dec 02  8:21
By: Ward Dossche
To: Michiel van der Vlist
Re: bogus arguments shot WRT 000-, 411 and 911
St: Rcvd                                                Reply chain 3030  3256
-----------------------------------------------------------------
Michiel,

> 2) I am still not convinced that it is actually possible with ALL
> mailers. The fact that your Frontdoor can do it does not mean that ALL
> POTS mailers can do it.

My mailer can not blocj a call to a "000"-number

 \%/{at}rd

-+- DB 1.58/001877
 # Origin: ** MANY GLACIER ** Preserve - Protect - Conserve (2:292/854)

 MvdV>> But even if /my/ nodelist compiler can handle it, how do
 MvdV>> we know that ALL nodelist compilers can?

 SH> Do you need to try every wrench in the world on a 13 mm
 SH> bolt to know a wrench will turn the bolt?

No. That is because bolts and spanners are standardized and turning a 13 mm
bolt with a 13 mm spanner confirms to those standards. So any 13 mm spanner
should fit any 13 mm bolt.

Putting a FQDN in the phone number field however is NOT according to
standard. It is using the tools for something that is not in the specs. It
is like the spanner that was designed for a hexagonal bolt to be used to
crank up the weights of a grandfather clock. It may work or it may not. And
if it works on my clock it may not work on yours. To know if it works on
all clocks one does indeed have to try it on all different types of clocks.

 MvdV>> It will attempt to make the call if I set the status
 MvdV>> of the message to "crash".

 SH> Point taken but you know who you are writing to.

Do I? Yes /I/ do. I do now because I have digged into the matter. In
general I have no other way of knowing if a node like you is POTS callabel
or not than to fire up an ascii editor and look at the raw nodelist. And
that is not how it was intended to be.

 SH> Crashmail to me would make no sense even if I were still POTS
 SH> capable.

My software however has no way of knowing that as you refuse to carry a Pvt
keyword. My editor does not warn me that sending a crash message to you is
not possible. So iyt will happily accpet the message and offer it to the
mailer for ddispatching.  My mailer will only discover it can not be send
when it starts to actually try it.

 MvdV>> Strictly speaking policy does not allow routing to RIN's.

 SH> Policy is now so far out of date that people may simply
 SH> decide to say to hell with it.

That's anarchy.

 SH> And sometimes common sense does prevail -- I was receiving
 SH> routed netmail

Common sense says that if one is able to send and receive routed netmail
one no longer has an excuse not to be in a net.

 SH> and routed echomail

Echomail is never routed.

 SH> since I first became a RIN in 1991.

If you were able to receive routed mail you should never have been a RIN in
the first place. You should have been a member of the net you got your
routed mail from. THAT is common sense.

 MvdV>> I need it in order to make a direct connection to you.

 SH> And why do you need to make a direct connection to me?

I may want to send you a message that I consider confidential and so I do
not want others to be able to read it.

 MvdV>> It is new to me and that is what matters.

 SH> There are thousands of software programs out there that I
 SH> do not use.  If I decide to use one, it may be new to me
 SH> but it's far from the cutting edge.

Smoke screen.

What is relevant is that you force me to replace and/or update my software.

 MvdV>> No it is not. It is not directly reachable for over half of
 MvdV>> FidoNet and it is not reachable by host routing. You don't
 MvdV>> have a host.

 SH> I have a host.

No you do not. Now who is bastardizing the language?

 SH> Now if someone did some serious work on the nodelist,
 SH> they'd figure out how to insert that information.

If someone would not insist on driving his car upside down, there would be
no need for a technical monstrum as a car with wheels on the roof.

 MvdV>> Fidonet technology isn't fading at all. It works
 MvdV>> just as it did five years ago.

 SH> So does my 1992 Honda Accord but the time to upgrade the
 SH> technology does come.

Why?

Actually I have a car that is even a year older. It is a Volvo 345. It just
had its annual service. It works fine and suits my needs, so why should I
spend a lot of money to replace it?

Cheers, Michiel

--- InterMail 2.29k