Text 3867, 215 rader
Skriven 2009-11-21 19:58:00 av Alexey Vissarionov (2:5020/545)
Kommentar till text 3865 av Michiel van der Vlist (2:280/5555)
Ärende: Some thoughts after the election...
===========================================
Good ${greeting_time}, Michiel!
21 Nov 09 12:37, you wrote to Ross Cassell:
MV>>> Hmmm.. That would be a way to get rid of old farts that oppose
MV>>> all change by blocking the road with history books.....
RC>> Those whom disregard history are doomed to repeat it.
MV> History does not repeat itself.
Yes - the history is repeated by people...
MV> It just looks that way to those with tunnel vision. Other than that:
MV> all we learn from history is that we do not learn from history.
... who don't know of this fact.
MV>>> But I think it is a bad idea. It would force experienced people
MV>>> out.
RC>> But you already expressed concern over Ivory Towers..
MV> In case FTSC members and only FTSC members may nominate and vote.
MV> When FTSC members can not vote or nominate and they have to go
MV> through the election process every x years, I see no danger of an
MV> ivory tower.
So, the necessary and sufficient condition (== the criterion) for avoiding the
Ivory Tower effect is keeping politicians away from our technical swamp. QED.
RC>> I had a node that was overusing existing flags within their
RC>> specified usages that broke nodelist processing and he very much
RC>> wanted the system name intact
What prevented the use of DNS name in "system name" field? I was quite happy
having "Fido.Gremlin.RU" there, and
gremlin@home:~ > host Fido.Gremlin.RU
Fido.Gremlin.RU has address 89.250.1.33
- it's resolved exactly as expected.
RC>> and offerred POTS, so the pots field was used..
I hope our colleagues from R2:40 don't read this, as "pots" in Hebrew (or
Yiddish?) means "the prick" :-)
#include <FTA-1006>
Anyway, I insist that PSTN field MUST contain either phone number or special
"-Unpublished-" string. Nothing else.
RC>> IBN:<hishostname>:<port> and so on for all the other one million
RC>> protocols he offerred.
MV> An idiot breaking the nodelist with such a stupid move is not a
MV> technical problem. No more than puncturing a tyre with a knife is.
IBN|IFC|etc flags SHOULD carry one optional argument - the port number.
Everything else (host address) SHOULD be in INA flag.
Personally I'd like to have s/should/must/ig there - it could make the
generation of DDN NS zone much more simple. Well, R2:50 already uses this
approach, but it's only one region.
RC>> In fact I had to replace his hand picked NC over the matter.
RC>> But he wasnt the only one doing this, hence my suggestion that
RC>> gained some traction and whats more important, it works.
MV> Are you sure that it works? Do you know how Irex handles the INA
MV> flag?
I don't. And never will, unable to look at the source code, as Irex is a
proprietary shit.
MV>>> Putting the host name or IP address of an IP node in the system
MV>>> name field is a technically sound solution.
Not the best, but... Just give your node a perfect DNS name :-)
RC>> Whats wong with having other options and why do you think
RC>> surrendering the system name field will suffice just because
RC>> you dont mind it?
2 RC: who said that word - "surrendering"? Don't you have a personal domain?
MV> What is wrong is that it breaks existing software and forces the few
MV> authors that are still around to update their programs to support yet
MV> another way of listing IP nodes.
That's normal.
MV> What is wrong that to appease a relatively small group of whiners,
MV> it forces EVERYONE ELSE to upgrade their software or be stuck with
MV> something that is broken.
Bwa-ha-ha... The word "forcing" may mean: break into their systems, FUBAR it
(preferrably ruin all hard disks) and then tell them "try this new thing - it's
very easy to set up" :-)
I did it once, and it had the predicted effect. For now... I'm not sure the
effect nesessarily will be the same.
MV> What is wrong is that it shows total disregard for the programmers.
Agree. But if the project is supported - it usually adopts that, and if it is
not - it dies.
MV> I once suggested that everyone launching a new flag should first
MV> write a parser that does something useful with it. That would stop
MV> the introduction of frivolous ideas that put an undue burden on the
MV> authors of software. Didn't make it...
ww ww hh hh yy yy ?????
ww ww hh hh yy yy ?? ??
ww ww ww hhhhhhhh yy yy ??
ww ww ww hh hh yyyyyy ??
ww ww ww ww hh hh yy
www www hh hh yyyyyy ??
Why?
It could be a really good requirement to publish a reference implementation.
If not reference - ok, let it be an example.
Personally I support it. And I guess (being sure, ha-ha) at least Nick, Andrew,
Stas and Egor - being active developers - would support it.
MV> In the end I complied and adapted my software, But it was not a
MV> trivial matter. the INA flag upset my carefully designed system of
MV> grouping flags into classes.
Why??? Flags are quite selfish...
MV> Modem capability flags, IP capability flags, e-mail flags, ect, etc.
MV> That way I could specify what flags may carry what parameter ect,
MV> without having to hard code the flags.
Hmmm... I haven't seen the source, but I afraid you've missed something.
MV> But the INA flag screwed up that system, I had to introduce an entire
MV> new class of flag: the IP address carrier flag. A flag that contrary
MV> to normal IP capability flags MUST carry a host name, may NOT carry a
MV> port number and may not be used by itself, but must be used in
MV> combination with one or more IP capability flags. It could be done of
MV> course and it has been done, but it was not an exercise in
MV> triviality.
Perl (PCRE library for C) allows doing such a parsing in a couple of lines.
MV> The INA flag set me to WORK!
My apologies.
MV> And for what? So that politicians could appease a small group of
MV> whiners and so gain points for the next popularity contest.
Not only - it really simplifies the parsing of nodelist line.
MV> It would have been one thing had I been a paid employee. A job is a
MV> job isn't it? But to be forced to work as a volunteer is a different
MV> kettle of fish.
Michiel, you know I mostly agree with you - but here I see only your mistake.
RC>> Let them all eat cake!
MV> I associate that with a French queen that lived some 200 years ago.
MV> Ancient history, I fail to see the link to Fidonet.
You [at least] both want historical references? I have enough... :-E
MV>>> If that is such a good idea, why is there no limitation on the
MV>>> term of an RC? Or ZC? How long have you been RC? When will you
MV>>> step aside to make room for fresh blood? You keep harping Ward
MV>>> for being ZC for life, but Janis has been in the velvet seat for
MV>>> ages in my perception. When will the Z1 RC counsel tell here it
MV>>> is time for fresh blood?
RC>> You are the one whom expressed concern for the FTSC, I merely
RC>> tossed a suggestion.
MV> You are avoiding the question. if limitations of terms is such a
MV> good idea, why not start with the *Cs?
When I read this, I feeled there must be a historical reference - and here it
is: all Soviet leaders with the only exception of Khruschyov had left their
chairs being carried with their feet facing forward.
MV>>> As I see it, limitation of terms is much more needed for th *Cs
MV>>> than for FTSC members. Technicians that are around for a long
MV>>> time, just have their reflexes slow down and they become less
MV>>> productive. They just become old technicians, they do no harm.
Disagree. They become old farts who are braking the progress.
One more historical reference to our Soviet era: when an old respectable
engineer, who most likely was an inventor of blleding-edge technologies,
approached the age of 50...60 years, he had no chances to "climb up" his
professional ladder - he was kicked off the hierarchy and could either to
continue working as an ordinary specialist or find another job.
MV>>> Politicians OTOH gain power and as we all know power corrupts.
RC>> Deja-Vu, you arent related to Lesley-Dee Dylan are you?
MV> The Lesly-Dee Dylan affair illustrates the drawbacks of the
MV> incestious system where RCs elect the ZCs and the ZC appoints the
MV> RCs. THAT is the kind of system that gets you a ZC for life.
MV> Something that you have been harping on when it pops up somewhere
MV> else...
OMFG... One more reference to our Soviet era - exactly shows the internals of
Communist Party and their leadership.
We all know how they've finished. And I don't like the possibility of Fidonet
finishing in similar manner... :-/
--
Alexey V. Vissarionov aka Gremlin from Kremlin
gremlin.ru!gremlin; +vii-cmiii-ccxxix-lxxix-xlii
... god@universe:~ # cvs up && make world
--- /bin/vi
* Origin: http://openwall.com/Owl (2:5020/545)
|