Display only digipeaters
barrie
Is it possible to create a map for specific areas that would display
only the active APRS digipeaters.
73 Barrie VE3BSB
Heikki Hannikainen
> Is it possible to create a map for specific areas that would display
> only the active APRS digipeaters.
Everything is possible, given the time and sufficient peak of motivation.
It's been on my mind (like a ton of other things), and it's been requested
before, too.
I'd like to do a larger, more general set of improvements in the map
filtering. Show (or not show) stations based on the destination callsign
(for ALTNET support), the symbol, callsign glob (hide CW* or only show
OH7*), source callsign match, only show igates or digipeaters (or both:
only show network infrastructure). And of course the old
ais/object/item/aprs station filters in the Preferences should be merged
to the same place. Anything else? Filter by tracker vendor / software?
Altitude - only show flying things? Minimum speed?
The harder part is figuring out the user interface. How to fit it in the
already crowded pane on the right? Another tool button to open a pop-up
window containing the controls? An expandable box within the pane?
Another thing I'd like to do is an advanced search tool with quite exactly
the same search terms as the filter. It'd be cool if I could manage to
write a single implementation of the UI and use it both for the advanced
search and the filtering.
- Hessu
Cap Pennell
<snip>
> ais/object/item/aprs station filters in the Preferences should be
> merged
> to the same place. Anything else? Filter by tracker vendor / software?
> Altitude - only show flying things? Minimum speed?
I'd like to be able to still view VHF WX stations while filtering out the
"internet-only" WX stations. I'd like the gradient colored squares
designating activity (when the map is zoomed out too far to show individual
stations) to _not_ include AIS data (but instead to show only APRS RF
activity).
I'd like the info page to not proclaim Good Path! for WIDE1-1,WIDE2-2 (three
hops in all directions is not so good).
You asked.
73, Cap KE6AFE
James Ewen
> I'd like the info page to not proclaim Good Path! for WIDE1-1,WIDE2-2 (three
> hops in all directions is not so good).
Only for you Cap!
Three hops is within the proclaimed proper operations...
There are some areas in the wilds that need 3 hops to get to another
human being, let alone i-gate.
Hessu would need to have multiple "good path" recommendations, based
on what area of the planet, and even then there will be people
complaining about it not being "proper".
Besides which, we've been over this a lot already... fix your
digipeater network. The coverage area is too large for the number of
users you are trying to support. Don't blame the network, fix it! We
haven't done this in a long time. 8)
Hope your new year is great!
James
VE6SRV
Cap Pennell
Yes, our digis are still on "hills" in California, but luckily more of them
"trap" or ignore WIDE2-2 now. And I know it's tougher to convince hams to
add enough IGates where there are fewer people living (it needs at least a
full-time internet connection and a VHF receiver). <grin>
Proclaimed proper operations? Around here it's suggested that most
commuters better use only plain WIDE1-1 to suit their audience's needs,
while helping to share the busy VHF channel. Seems most of our ham
stations' audience members are watching from Hessu's aprs.fi nowadays
anyway. hi hi
Our international wiki,
http://info.aprs.net/index.php/Paths
includes:
MOBILE Stations in North America (VHF)
WIDE1-1,WIDE2-1 :Two hops in all directions with potential to use fill-in
digis. Best path for mobile use in most all areas.
WIDE1-1,WIDE2-2 :Three hops with potential to use fill-in digis. Use only
if truly needed while very far from any mountain digis or cities
Happy New Year to all. See you on APRS!
73, Cap KE6AFE
James Ewen
> Spoken like a true Central Albertan Flatlander, or a Texan, James.
I look forward to the days when a little VHF lift helps out, and I can
see stations from Grande Prairie when I'm driving around Bonnyville.
It's only 550 km a the crow flies... a little further on digipeater
routes. When that happens I have someone to look at finally!
> Seems most of our ham stations' audience members
> are watching from Hessu's aprs.fi nowadays
> anyway. hi hi
Hey, when you build a product that does what everyone needs and wants,
puts it in front of them in an attractive and simple to use manner,
what do you expect?
My biggest beef with Hessu is that he doesn't have a version of
aprs.fi that I can put on my laptop, interface to my radios, and take
it with me where ever I go...
The only reason I run other software is because I can't use aprs.fi
offline. When we chase balloons, we set up a mobile wifi hotspot, and
tether to the cellphone to be able to use aprs.fi for chasing. We feed
the data received from the balloon with our mobile radios, feed that
to the internet over the wifi/cellphone tether, and then pull it back
off the internet on the aprs.fi site. How's that for dedication to a
medium? 8)
James
VE6SRV
Heikki Hannikainen
> My biggest beef with Hessu is that he doesn't have a version of
> aprs.fi that I can put on my laptop, interface to my radios, and take
> it with me where ever I go...
In case you are actually not joking...
If I went on and started working on *that*, you'd have no updates for
aprs.fi for the next 3-4 years. :) Or I'd have to quit my day job, but
that's currently not an alternative.
I'm hoping for Lynn to fill that gap in desktop software!
- Hessu
Lynn W Deffenbaugh (Mr)
my fingers can enter the code! In case anyone doesn't know what Hessu's
talking about, check out the APRSISCE/32 Wiki at
http://aprsisce.wikidot.com/start and then join the Yahoo group at
http://groups.yahoo.com/group/APRSISCE/ to download your free copy!
Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32
PS. Thanks for the opening for the shameless plug! I wouldn't stand a
chance of developing this beast without aprs.fi to help with the debugging!
PPS. Yep, and I still have the day job too that keeps interfering with
my "work" on APRSISCE/32!
Bill V WA7NWP
> Yes, our digis are still on "hills" in California, but luckily more of them
> "trap" or ignore WIDE2-2 now. And I know it's tougher to convince hams to
> add enough IGates where there are fewer people living (it needs at least a
> full-time internet connection and a VHF receiver).
I wonder what the Amateur Radio RF network would evolve to if folks
put some time and energy into making it work smarter and better. Than
it could really support a reasonable level of traffic. Instead we
use old and slow Amateur technology with more and more reliance on the
Internet.
73,
Bill - WA7NWP
Heikki Hannikainen
> Hessu would need to have multiple "good path" recommendations, based
> on what area of the planet, and even then there will be people
> complaining about it not being "proper".
This was what I was thinking of while implementing the path advisor's
recommendations. I tried to be quite conservative and stick to Bob's
general universal recommendations which now have some IARU backing, even.
For some areas 3 hops might not quite make it to an igate, but naturally
that network won't work too well in any case. For many areas 1 hop is the
right choice, especially if you just need to hit one digipeater and an
igate listening to it. But how can aprs.fi know?
The path advisor's main intent is to clearly flag the Really Evil settings
of WIDE7-7 (and more) as bad paths. Those are doing the worst damage to
the network, and for themselves, since really long paths also result in
duplicates and out-of-order packets on the APRS-IS. Here's the real wall
of shame: http://f5vag.eu/php/find_digi.php?BadDigiPath
I expected to get more feedback about the path advisor, but I've only had
a couple of complaints about it requesting shorter paths than what the
user thinks is required. So I guess it's working pretty well after all,
since I didn't receive any letter bombs.
I've had two improvement ideas:
(1): combining the path advisor (WIDE4-4 is too long!) with the transmit
rate advisor (it's not good to send a packet every 20 seconds all the
time!), so that it'd actually calculate some fuzzy guesstimate of the
user's radio channel usage, by multiplying the mean packet rate (packets
per minute) by the mean amount of digipeaters requested, and then doing
the complaining based on thresholds of that. Or maybe an exponential
punishment based on the digi hops, or something in between. But you get
the idea:
It'd be OK to send pretty often if you have an empty path and no
digipeaters, and OK to send with 3 or 4 hops if you only transmit rarely.
That would also account for proportional pathing schemes properly - some
people transmit quite often with 0 digi hops requested (empty path), and
every 4th packet asks for 2 or 3 hops. Or something. That makes sense to
me, but currently trips the transmit rate advisor on aprs.fi.
(2) Implementing a new tool which could be used to determine the amount of
digis needed for a given route. The user would first drive from place A to
place B with WIDE1-1,WIDE2-2 (3 hops), then go to aprs.fi and ask for path
statistics for his callsign between the two spots (or timestamps). aprs.fi
would then calculate, from the received points, that 99% of the points
were received after passing through 0 or 1 digipeater hops, 1% required 2
hops, 0% used 3 hops. That would serve as technical evidence that WIDE1-1
or WIDE2-1 is good in that area and additional hops didn't do much
additional good.
That, of course, assumes that all the user wants to do is hit an igate,
not caring about the visibility to the larger RF network. I'm guessing
that's what most people care about these days, although that might not be
the dream world of some enthusiasts. But also, that's probably the best we
can do with our state-of-the-art 1200 bit/s ALOHA technology. :)
Maybe someone should make some statistics about how often packets have
made it to an igate after more than 2 or 3 hops in each country or state
(excluding duplicate packets which have also been seen with less used
digipeater hops). If it's a rare occasion, that's a clear technical proof
that long paths are not required in that area. The tricky part is figuring
out the country and state borders (or other sensible areas). Maybe grid
squares would work instead of political boundaries?
Actually, idea (3) would be that the people manufacturing trackers and
implementing tracker software (Lynn, that includes you now :) would make
it DEAD SIMPLE to configure a Good Path, and outright IMPOSSIBLE for the
end-user to configure a Really Bad Path. Something like a graphical
drop-down menu of 1, 2 or 3 hops, and maybe a checkbox for 'use TRACEn-N
instead of WIDEn-N'. Maybe an entry in the drop-down menu for 'ARISS' path
alone for ISS working. There could be an arbitrary path text field hidden
deep somewhere in the advanced settings (or in the config file), but even
then, the tracker software should stop working if the user tries to
configure a path with more than 4 digipeaters total. If we could get all
tracker manufacturers to commit to that, the world would become a better
place and some rainforests might be saved again. ;)
PS. If you wish to further discuss the advancement of the state of the art
(digi paths, improving your regional APRS networks and things like 9600
bit/s settings), I would suggest that you take the discussion to the
region's mailing lists and discussion groups, and keep this group
dedicated to aprs.fi specific things. Or APRSSIG for global general APRS
things. Thanks!
- Hessu
Cap Pennell
proportional pathing as you note. Displaying the product of a station's
digipaths and transmission intervals sure could better indicate relative VHF
courtesy. <grin>
Thanks for all your work.
73, Cap KE6AFE
> -----Original Message-----
> From: apr...@googlegroups.com [mailto:apr...@googlegroups.com] On
> Behalf Of Heikki Hannikainen
> Sent: Friday, January 07, 2011 12:16 AM
> To: apr...@googlegroups.com
Bill V WA7NWP
> of WIDE7-7 (and more) as bad paths.
Those are doing the worst damage to the
> network, and for themselves, since really long paths also result in
> duplicates and out-of-order packets on the APRS-IS. Here's the real wall of
> shame: http://f5vag.eu/php/find_digi.php?BadDigiPath
The problem is there is more than just the one universal network being
viewed here. I could use WIDE7-7 every 30 seconds on our alt-network
and it would be "a good thing" yet the universal advisor would
improperly flag it as Really Evil....
The network monitoring should be a local issue...
73
Bill - WA7NWP
Lynn W Deffenbaugh (Mr)
frequency and not just an AltNet (page 15 of aprs101.pdf) destination
address on a shared frequency.
Network monitoring really can't be done with APRS-IS unless it comes
with a page full of caveats about what is and is not implied by the
analysis. The dupe filters and unknowns about what frequency, baud
rate, and other stuff make such "load" analysis not worth the effort
IMHO. Heck, we can't even tell how many digis a packet went through
even if I hear that packet directly on RF! (no, please don't change the
topic, but path non-completeness may be germane to this discussion).
Lynn (D) - KJ4ERJ - Clarifying the ambiguous language...
dra...@gmail.com
Heikki Hannikainen
> Hey Hessu,
> A bump to this thread but do you think this "advanced search" feature might be released some day? In my specific want, I'd love to ONLY see all of
> the available digipeaters in a given area.
advisor, but:
You can use the filter feature to show the available digipeaters:
1. Tap the "funnel" filter button
2. Select "Digipeaters"
3. Deselect "All stations"
You can use the "Save as new list" to save it as a new filter list with a
custom name ("digipeaters" for example), after which you can use the
drop-down menu to quickly recall it.
- Hessu