| |
ZL3AI > APRDIG 30.09.03 10:49l 218 Lines 9150 Bytes #999 (0) @ WW
BID : 2029-ZL3AI
Read: GUEST
Subj: TAPR Digest, Sep 27, 5/7
Path: DB0FHN<DB0RGB<DB0MRW<DB0SON<DB0ERF<DB0FBB<DB0GOS<ON0AR<WB0TAX<7M3TJZ<
F6CDD<EA5RQ<VK7AX<ZL2BAU<ZL3VML
Sent: 030929/1353Z @:ZL3VML.#80.NZL.OC #:9307 [Chch-NZ] FBB7.00i $:2029-ZL3AI
From: ZL3AI@ZL3VML.#80.NZL.OC
To : APRDIG@WW
Subject: RE: Clueless Users? How's that <g>...
From: "Scott Miller" <scott@home.miller.org>
Date: Sat, 27 Sep 2003 09:37:12 -0700
X-Message-Number: 21
First, I think the network should be designed such that loss of Internet
access doesn't prevent it from working - there still needs to be a backup
plan, whether it's UHF backbones or traditional WIDE flooding.
I think taking different categories of traffic off 144.39 might be a good
idea - weather and telemetry stations, to begin with. Put them on another
frequency, and maybe set up a digi to queue traffic from that channel and
retransmit it on 144.39 as a (dupe-filtered) burst every 10 minutes - marked
as local-only traffic. That way you preserve your net cycle time, weather
data's still visible to everyone, and the weather stations aren't competing
with trackers.
Scott
N1VG
----------------------------------------------------------------------
Subject: Re: APRS Grass growing...
From: Steve Jones <s.jones@rogers.com>
Date: Sat, 27 Sep 2003 12:57:28 -0400
X-Message-Number: 22
on 9/27/03 11:20 AM, Bob Bruninga at bruninga@usna.edu wrote:
>Wow, MORE complexity, MORE things that a USER can screw up, MORE
>things for a USER to do...
Same thing can be said for your new LINKn-N routing. More complexity for
the enduser. He must now decide what path to use. Isn't LINKn-N similar to
the directional digipeating?
You speak of having stationary stations up their beacon time to 10 minutes
during nets. Now they must change this option in their software. If they
want to see stations timeout due to leaving the net, they must make sure
their station timeout timer is set low. More complexity on the part of the
enduser.
Now sysops must run up to their digi's to change software options (more
complexity). You keep assuming we all run KPC3's for nodes. Some of us run
digned which can easily be configured using a rules based system.
Sounds more like any improvements to the aprs system are dead unless you
deem them simple enough for Joe Ham to implement. Lets give the guys credit
as they did at least have enough knowledge to pass the exam. I guess any
improvements to the system are out if they require a little education on the
part of the enduser.
What's better? A system that might give you warning that a station has left
the net but is simple to operate or a system that gives you much better
warning but requires the net control station to enter a list of callsigns
participating in the net?
I'd go for more reliability but I guess I consider myself and the hams in
the area intelligent enough to figure things out.
Myself, I think nothing will change with regard to the current WIDEn-N
network and that LINKn-N will never get off the ground.
--
Steve <s.jones@rogers.com>
----------------------------------------------------------------------
Subject: Re: Another thought on reducing congestion in busy metro areas...
From: "Scott Miller" <scott@home.miller.org>
Date: Sat, 27 Sep 2003 10:16:04 -0700
X-Message-Number: 23
Yes, TNC dumb trackers need to go. They're a total waste of bandwidth when
even very simple trackers are capable of using compressed formats. Now if
we could just get all the TT3 users to take their cutesy comments out of
their beacons...
What exactly is sad about resistor-based D/A converters and PICs? (For the
record, mine uses a PWM-based D/A converter and a Motorola HC08 processor.
=]) If you're building a new system from scratch, the CML chips might be a
good option, but I don't see how they're really superior to what we're using
now. Yeah, it's easy to interface, but you're still using a half-duplex
audio channel with very limited bandwidth.
And hams like cheap stuff. A CMX882 costs how much, and still requires a
microcontroller? My design uses a single I/O pin and requires a resistor
and a capacitor. The RX-capable version uses a $0.90 NJM2211 FSK decoder.
Yeah, it's a pain to write the software, but you only need to do it once.
And by changing software I can completely change the modulation format -
even send recorded voice, if the MCU's got enough memory for it.
Scott
----------------------------------------------------------------------
Subject: Re: APRS Grass growing...
From: Bob Bruninga <bruninga@usna.edu>
Date: Sat, 27 Sep 2003 13:30:08 -0400 (EDT)
X-Message-Number: 24
On Sat, 27 Sep 2003, Steve Jones wrote:
>on 9/27/03 11:20 AM, Bob Bruninga at bruninga@usna.edu wrote:
>
>>Wow, MORE complexity, MORE things that a USER can screw up, MORE
>>things for a USER to do...
>
>Same thing can be said for your new LINKn-N routing. More complexity for
>the enduser. He must now decide what path to use. Isn't LINKn-N similar to
>the directional digipeating?
Yes, but it puts the onus on the DX person that wants to send long
distance to configure for it... Right now we have a network that allows,
or outright encourages people to FLOOD the network with their default
settings. This is what I am trying to do. Get people to use the default
paths that work everywhere, and plan on local activity.
If they NEED to get packets long distances from point A to Point B, then
I want to make sure they have that capapbilty, even if it takes a little
more "understanding" of the local net to accomplish it.
.. >
>You speak of having stationary stations up their beacon time to 10 minutes
>during nets. Now they must change this option in their software. If they
>want to see stations timeout due to leaving the net, they must make sure
>their station timeout timer is set low. More complexity on the part of the
>enduser.
That is the FAULT of client software. In APRSdos, (and part of the
fundamentals of APRS) the user software decays the packet period to 30
minutes if the user is using a path longer than 1 hop. If he is operating
direct, or through a single local digi, then it AUTOMATICALLY changes the
decay algorith to the 10 minute final period.
I never intended for USERS to be able to dink with these fundamental
assumptions about network timing and data latency. Problem is, some other
client software never implemented these fundamentals, thus users have all
kinds of ways they can dink up their seettings and totally screw up the
reliabliity and fundcionality of APRS...
>Now sysops must run up to their digi's to change software options (more
>complexity). You keep assuming we all run KPC3's for nodes. Some of us
>run digned which can easily be configured using a rules based system.
Yes, I thnk DIGI_NED is an ideal engine at a digi... I assumed it has
remote sysop features. Are you saying it doesnt?
>Sounds more like any improvements to the aprs system are dead unless you
>deem them simple enough for Joe Ham to implement. Lets give the guys
>credit as they did at least have enough knowledge to pass the exam. I
>guess any improvements to the system are out if they require a little
>education on the part of the enduser.
Yes. That is the basis of it. If we are building a network who's
reliability is VERY MUCH dependent on the nuances of the fundamental
assumptions and settings, then we cannot let the common user dink with
it. It can only screw it up...
>What's better? A system that might give you warning that a station has
>left the net but is simple to operate or a system that gives you much
>better warning but requires the net control station to enter a list of
>callsigns participating in the net?
>
>I'd go for more reliability but I guess I consider myself and the hams in
>the area intelligent enough to figure things out.
Yes, the greater reliability is the KISS approach. Just let everyone
beacon their status, and you just cannot get any more reliable than that
in the aggregate of all users.
>Myself, I think nothing will change with regard to the current WIDEn-N
>network and that LINKn-N will never get off the ground.
It shouldnt get off the ground unless it has a local application. If
locals see the value in it, then they will use it. If they dont, they
wont. All I ask is that ALL AREAS support the fundamental LOCAL APRS
paths of RELAY and WIDE. All I am doing is helping them to understand
why this is so important, and I am giving them OPTIONS that they can use
locally to reduce the DX packets that may be flooding ther system from
afar.
No more... no less.
Ah, and then as Hasan says, we can get back to USER education, because
APRS will then work the way we designed it AND there is ONE answer
everywhere as to what is the proper path... Then only local specialty
options are needed for those instances whree someone needs to make packets
go farther from point A to point B but without screwing up neighboring
reliability.
Bob, WB4APR
----------------------------------------------------------------------
Subject: National WIDE5-5 day
From: "Spider" <spider@rivcom.net>
Date: Sat, 27 Sep 2003 10:34:41 -0700
X-Message-Number: 25
Wanna prove your local network?
How about a national WIDE5-5 day where everyone runs WIDE5-5 for a day.
How about October 5th?
Jim
WA6OFT
----------------------------------------------------------------------
Read previous mail | Read next mail
| |
