Text 27785, 106 rader
Skriven 2012-04-10 00:52:10 av Michiel van der Vlist (2:280/5555)
     Kommentar till en text av Robert Bashe (2:2448/44)
Ärende: Ghost nodes
===================
Hello Robert,

On Monday April 09 2012 18:21, you wrote to me:

 RB>>> The choice is whether you spend every minute of your free time

 MV>> Nonsense. With nets having no more than a hand full of nodes,
 MV>> checking them just takes a few minutes.

 RB> Nonsense yourself. I've done it and know what I'm talking about.

So do I. When I took over from Jan Vermeulen as NC280 some three or four years
ago, I cleaned out the dead wood. I had voice numbers and/or e-mail addresses
for most of the sysops, so it was not all that hard to track most of those
who's systems were no longer pollable. There were a few that took a bit more
effort, but it did not take all that long to decide which one were really gone
and which one needed some more probing.

After the initial weed out, maintaining a clean nodelist segment is not much
work at all. I have all the data needed to quickly inquire what the problem is
when there is a problem.

The only exception was when a node dropped out last year because the sysop
died. It was not unexpected, I knew he had serious health problems, but the
confirmation took a bit longer because his e-mail address and voice number were
dropped within days.

 RB>  I spent several days more than once trying to locate someone (ONE
 RB> node) to ask whether his system was still active or not. You can't
 RB> just poll the system, not even several times - plenty of people go
 RB> offline fo repairs or changes, and if you're unlucky, you'll poll them
 RB> exactly then.

Granted a single poll won't do.

 RB>  Anyone who has tried to poll our RC in the past few weeks would,
 RB> using your reasoning, have taken him out of the nodelist. And then?

I have mail waiting for him in the outbound since 27 March. My system tries to
deliver it every 30 minutes. It is not the first time this happens. It is
obvious that there is a problem that goes beyond a mere power failure, or down
for maintenance.

Had this been a leaf node in my net, I would have flagged him as "Hold".

 RB>>> running after people who don't give a damn

 MV>> I don't do it for /them/, I do it for all the sysops all over the
 MV>> world that are still active and who expect a nodelist that is as
 MV>> up to date as possible.

 RB> Your choice, then. If you were in R24, I'd offer you the NC2448.

I am not in R24.

 RB>>> or simply let them exist as dead wood.

 MV>> If you let them exist as dead wood you are derelict in your NC
 MV>> duties.

 RB> Michiel, you do what you think is right and I'll do what I think is
 RB> right, OK? If Ward or our invisible RC wants to remove me as NC, that
 RB> would overjoy me. The only problem would be that they't have to find a
 RB> replacement, and there aren't that many around nowadays.

Then maybe it is time that you guys in R24 do what Jan Vermeulen did in R28
some ten years ago: a complete regroup. In 2000 we had 13 nets, some of them
with just one or two nodes. Some with inactive NC's and no one willing/able to
replace them. The solution was simple but drastic: disband all nets, except
one. Throw all remaining nodes in one net.

It was drastic but it worked. And surprisingly it met with little resistance.
Most could be convinced that a complete regroup was the only viable long term
solution.

 MV>> Then perhaps you should consider to hand over the hat to someone
 MV>> who is willing and able to maintain the nodelist as an NC should.

 RB> I find that remark offensive and presumptuous.

Sorry if you feel that way, but I really think that if by your own estimate
only 30% of the listings in your segment reflects reality, then you are doing
something wrong.

 RB> He is NOT sending a regional segment, and my net segment is NOT
 RB> included unless I send it directly to Ward and ask him to manually
 RB> include it, which I personally consider a great imposition for him.

I concur. A ZC should not have to do the job of an RC.

 RB> I think I'm doing more than most others would do under the
 RB> circumstances and find your criticism exaggerated and very unfair.

Again: sorry if you feel that way, but doing more than most isn't always the
best strategy. maybe it is time for a major regroup. Think out of the box.

And don't tell me this would lead to a massive revolt in R24. It is what I
thought at first when Jan Vermeulen proposed his plan for R28. We had our share
of problems too in R28 you know. But times change and old hatches can be buried
when the need arises. It happened in R28 and it can happen in R24.


Cheers, Michiel

--- GoldED+/W32-MINGW 1.1.5-b20110320
 * Origin: http://www.vlist.org (2:280/5555)