| |
ZL3AI > APRDIG 06.03.04 14:00l 228 Lines 9630 Bytes #999 (0) @ WW
BID : 2968-ZL3AI
Read: GUEST
Subj: TAPR Digest, Mar 04, 10/14
Path: DB0FHN<DB0RGB<OK0PPL<DB0RES<ON0AR<LZ3NP<KD7HAH<GB7LGS<GB7ESX<ZL2BAU<
ZL3VML
Sent: 040306/1112Z @:ZL3VML.#80.NZL.OC #:20457 [Chch-NZ] FBB7.00i $:2968-ZL3AI
From: ZL3AI@ZL3VML.#80.NZL.OC
To : APRDIG@WW
Subject: RE: stir up the path pot!
From: "Scott Miller" <scott@opentrac.org>
Date: Thu, 4 Mar 2004 15:41:12 -0800
X-Message-Number: 74
>Then I suggest putting up more Igates in that area! With this mind set if
>there aren't any I-Gates in my area, I should increase my path until I do
>get into one just so I can be seen on the internet and to heck with anyone
>else that it affects.
Since it's obvious from the packet payload where the transmitting station
is, how about only restricting excessive paths for your local users? That
way people in the middle of nowhere can still reach you...
Scott
N1VG
----------------------------------------------------------------------
Subject: RE: stir up the path pot!
From: "Christensen, Eric" <CHRISTENSENE@MAIL.ECU.EDU>
Date: Thu, 4 Mar 2004 19:03:52 -0500
X-Message-Number: 75
Steve... Take a deep breath... I'm not forcing my values on the rest of
the APRS community. Physics is going to limit APRS not me. I'm just trying
to show how to get as many people into the same amount of time and space.
----------------------------------------------------------------------
Subject: RE: stir up the path pot!
From: Jeff King <jeff@aerodata.net>
Date: Thu, 04 Mar 2004 18:08:19 -0600 (CST)
X-Message-Number: 76
>Then I suggest putting up more Igates in that area!
Well, taking the politics out of this, this is a technically sound idea. Why?
Simple.
Lets say you have a digipeater. Lets say it hears and retransmits 9 out of
every 10 packets it hears (and many will be far less then this). So, it has
a 90% success rate.
But if the path is 4 (N) hops long to the closest Igate, the path success
rate is no longer 90%, but the individial success rate to the N power. In
the case of 90% success per digi, it would be 0.9^4 or 65.%. If each Digi
has a 80% success rate, and the path is 4 hops, the success rate to the
IGATE drops even lower, to about 41% success.
Tom Clark did a paper for either PSR or Packet Radio magazine about this in
the mid 80's as to why long digipeater paths are not very reliable for
packet radio. Granted, the flood method APRS uses may make this better or
worse, but in the case of a path just long enough to reach the first Igate,
the formula should stand on its own.
If your goal is getting the packet into the APRS-IS system, the quicker you
can move it off RF and onto the internet system, the better for both
yourself and other users of the channel.
----------------------------------------------------------------------
Subject: RE: stir up the path pot!
From: Steve Dimse <k4hg@tapr.org>
Date: Thu, 4 Mar 2004 19:29:33 -0500
X-Message-Number: 77
On 3/4/04 at 7:03 PM Christensen, Eric <CHRISTENSENE@MAIL.ECU.EDU> sent:
>Steve... Take a deep breath... I'm not forcing my values on the rest of
>the APRS community. Physics is going to limit APRS not me. I'm just trying
>to show how to get as many people into the same amount of time and space.
Here is another example, the Florida Keys. In all 130 miles of the Keys,
there is one full-time APRS user, me, and one occasional user, KQ4AZ. The
county has installed weather stations every 20 miles or so, on the highest
available towers (which because of environmental laws are very low, and of
course the highest point in the Keys is 10 feet above sea level) so the
range is very limited. These weather stations are digis, and allow the data
to make it to IGates in Miami, but most of them require more than two hops
to do so.
So under your proposal I would need to install IGates in three more places
in the Keys in order to be sure the weather data made it to the internet
and on to the NWS. I might be able to place an IGate on KQ4AZ's computer,
and the only expense would be the DSL internet connection or a second phone
line, the other two locations would require computers as well as the
connections. Even if the locations for these machines are rent-free, are
you willing to pay the $2,000 startup costs and $70-$150/month for the
additional IGates you recommend?
The long paths do not cause congestion in the Keys, and are not stunting
the growth of APRS here (it's because there are few hams and lots of other
things to do here). Four digi hops do not cause problems here, yet you plan
on forcing your path band-aid on the Keys, and are making recommendations
that would cost others money.
Yes, I think you are trying to force your values on others!
There is not now, and likely never will be, a single path that works in all
areas. I think acceptable local networking behavior need to be defined
locally, not on a world-wide basis.
Steve K4HG
----------------------------------------------------------------------
Subject: RE: stir up the path pot!
From: "Robert Bruninga" <bruninga@usna.edu>
Date: Thu, 04 Mar 2004 19:35:55 -0500
X-Message-Number: 78
Steve said:
>No, it only solves 2 and part of 4. There is still no impact on the amount of
>data appearing on the local RF network.
Not directly, but VERY MUCH indirectly.
I still think the majority of users want to be seen on the IS
Hence, they will see the error messages and will fix itt.
>this does nothing to educate the user, because with a position,
>when they go to the find.cgi page they will see nothing but
>a message saying no position, the weather page will say
>no data, the message page no messages.
No, it will display his original packet like this:
LID>APRS,WIDE7-7,RELAY7-7:Packet Truncated due to too-big-path in
area...
>It does nothing to adress that this is enforcing your will
>upon others, even areas that accept these paths, and
>are denying access to a resource that belongs to all of
>amateur radio.
No, it is only enfocing the will of the IGate sysops in each local area who
hopefully are in the best position to take action for their local area. And
this is fed back to the user when he looks on the APRS-IS... That is what
we are looking for is a feedback path to the user.
And we DONT WANT the feedback path on RF where it just adds to the load.
(and doesnt get to abusvie trackers anyway...)
>Finally, while it would give me a way to see what is wrong,
>it still places me in the position of getting these emails,
>investigating them, and answering them.
No this is not an Email thing at all, It is simply a truncating and
marking of excessive paths at the source of entry into the APRS-IS. The
end user then sees the feedback when he checks to see his packets on FINDU.
>I'm still very much opposed, you are taking a heavy-handed
>approach in a part of the system not being adversly affected.
I dont see it that way. I see it as an easy way to get good LOCALLY
derived feedback to users of abusive paths even if they are using trackers
and not listening on RF without adding any further load to the already
challenged RF. It is trivial to implement and thus has very good return on
investment. The ideas to replace ALL 1000 digipeaters to solve the same
problem WILL NEVER HAPPEN.... Hence, this idea to implement the filter when
it can be done quickly and easily...
>Instead of just wiping his data out, we could just shoot him
>or cut his coax. I'd rank the proposed solution just below
>these two alternatives.
Those are good alternatives too, but not as easy to implement overnight in
software.
>What is wrong with dealing with these as they become
>known, with a courteous email or snail mail, or a personal
>visit? This is the way ham radio used to address operator
>problems...
Because in 12 years of APRS it hasnt worked and with more and more newbees
getting on APRS without having evolved through APRS, they just dont get it.
>And as a last resort, after a ham has been educated,
>if he still refuses despite knowing he is causing interference,
>it can be bumped up to the FCC. I suspect that if the ham
>is approached in a friendly educational manner this will
>vitually never be needed!
This never works. No ham likes being challenged. That is why this
truncation at the IGate is such a good idea. It says "you may use that
path, but you are not going to get into the APRS-IS with it." THus it is
his choice.
Bob
----------------------------------------------------------------------
Subject: RE: stir up the path pot!
From: Steve Dimse <k4hg@tapr.org>
Date: Thu, 4 Mar 2004 19:38:56 -0500
X-Message-Number: 79
>If your goal is getting the packet into the APRS-IS system, the quicker you
>can move it off RF and onto the internet system, the better for both yourself
>and other users of the channel.
True enough, but there can be factors that make the placement of IGates
impractical or expensive. Certainly an area with a large active population
of hams would likely have a location that could be used, but there are
areas in the US and the rest of the world where the activity is low and
full time INet access rare, and to exclude those areas from the APRS IS is
not fair. They are not causing QRM, are not suffering QRM, and are doing
nothing wrong. Why make them suffer?
As I say, MAYBE there would be justification if the proposed measure
actually fixed the problem, but it does not, the offenders will still be
transmitting and causing the same amount of QRM. This proposal is based on
"punishing" them by booting them off the APRS IS, hoping they will notice,
and adjust there behavior. Will probably work for some, but certainly not
for all.
Find a solution that factually ixes the problem, and does not cause innocent
people to suffer, even if it is harder to implement.
Steve K4HG
----------------------------------------------------------------------
Read previous mail | Read next mail
| |
