In reply to G3VQO:
I think the behaviour is correct according the RBN info page. There again he’s still on the same summit, just the other side of the border.
Andy
MM0FMF
In reply to G3VQO:
There were two relevant alerts showing
10:00 I/F6HBI/P on I/LG-309 - [edit] 10.122-cw
then move to F/AM-512 (Posted by F6HBI)
12:00 F6HBI/P on F/AM-512 - [edit] 10.122-cw
Time approx, also FFF-188 and other Bands (Posted by F6HBI)
Checking RBN, the change corresponds to a change in callsign used. When the call’s from “I/F6HBI/P” it matches the first alert, and when the call’s from “F6HBI/P” it matches the second…
73, Rick M0LEP
In reply to G3VQO:
The final spot is the one that confuses me. I believed that, with two
or more valid alerts, RBNgate will associate the spot with the nearest
alert. Therefore, any RBNgate spot after 11:00 should include the
12:00 alerted summit F/AM-512, and certainly the spot at 12:18 (after
the second alert time) should be for the French summit.Is this a glitch, or is it another “feature” that we need to be aware
of?
Hi Les,
Your understanding about how RBNGate does a time-based match to the nearest alert (in time) is correct. And, “I/F6HBI/P” and “F6HBI/P” are the same callsign “F6HBI” from RBNGate’s perspective since it ignores portable designators when performing matches of spots to alerts.
The event that caused the confusing outcome you described was F6HBI’s self-spot at 11:22:
Thu 11:22 I/F6HBI/P on I/LG-309 24.902 cw
- (Posted by F6HBI)
That self-spot cancelled both of his SOTAWatch alerts and created a new virtual one (within the RBNGate software only) for the summit he spotted himself on (I/LG-309). This alert extended for 3 hours from the time the self-spot was received, and for 3 hours he would have been spotted on I/LG-309 regardless of which summit he was actually on.
The solution for this case would have been for F6HBI to later spot himself on F/AM-512 once he arrived there. Or, he could have refrained from sending in any self-spots during his trip, and his SOTAWatch alerts would have resulted in him being spotted on the correct summit at the correct time. He obviously plans his trips carefully and meets his ETAs with great precision!
73,
Eric KU6J
===========================================
Free SOTA Spot Monitor Software + RBNGate FAQ:
http://www.ku6j.com
Edit: Spotted summit was I/LG-309 and not I/F6HBI/P as I incorrectly had in the parentheses
In reply to KU6J:
And, “I/F6HBI/P” and “F6HBI/P” are the same callsign “F6HBI” from RBNGate’s perspective since it ignores
portable designators when performing matches of spots to alerts.
…which is certainly the cause of my confusion. Ignoring “/P” makes not a lot of difference, but the “I/” is significant, and shouldn’t have been ignored.
73, Rick M0LEP
In reply to M0LEP:
Well I’d always strip the junk people prepend and append to callsigns before matching them. We won’t go into the complete farce of matching UK callsigns with their inwards changes of secondary locator letter.
Eric, a suggestion which may appease those who expect 100% accuracy from a system trying to guess what is happening. Irrespective of the alert, get RBNgate to always spot the summit as ??/??-???. That way RBNgate spots that the RBN has heard a callsign matching an alert and chasers know an activator is on frequency calling CQ. Only give the summit ref in an unambiguous way if you see a self-spot from the activator.
I can live with the occasional inaccurate spot. The above tweak should make the system still as useful as now but with fewer complaints/questions.
Andy
MM0FMF
In reply to KU6J:
Hi Eric,
Many thanks for explaining the situation with the RBN cancelling the SOTAwatch alerts that have been placed once it receives a self-spot from an activator, something that I had not previously realised. Presumably this is for alerts within the 3 hour window from the receipt of the self-spot, or does it take out everything alerted for that day? If it is the latter, then I must remember not to fall foul of this on multi-summit days (usually greater than 3 hours between summits).
73, Gerald G4OIG
In reply to MM0FMF:
Eric, a suggestion which may appease those who expect 100% accuracy
from a system trying to guess what is happening. Irrespective of the
alert, get RBNgate to always spot the summit as ??/??-???. That way
RBNgate spots that the RBN has heard a callsign matching an alert and
chasers know an activator is on frequency calling CQ. Only give the
summit ref in an unambiguous way if you see a self-spot from the
activator.
That implementation is possible Andy, but one downside of always spotting the summit as ??/??-??? is that it would eliminate our ability to filter spots and trigger audio alerts based on the geographic region of the summit. SOTA Spot Monitor does this, and I imagine that people are doing this using other software as well, now that you are providing a DX-cluster-compatible telnet feed of SOTAWatch spots. As more and more associations around the world are added to the program, I believe that capability will only grow in importance.
Always spotting ??/??-??? would also inconvenience chasers with limited time who are primarily interested in improving their Unique Summits and Mountain Hunter award scores. They would no longer be able to make the chase/don’t chase decision by glancing at the spots, and would instead need to slog down to the shack, turn on the radio and wait for the activator to send their summit reference before even deciding whether or not to enter the pile-up.
In this case the activator actually did give an unambiguous summit reference via a self-spot, and that reference was correct when sent. The design philosophy in the RBNGate software is that the activator best knows where they are and where they are going, and is thus given control of whether or not they get spotted, and if so, when and on which summit. However, any number of reasons could prevent the activator from sending (or the system receiving) a subsequent self-spot. The system will never be 100% accurate as you rightly point out, and as Mike KD5KC would say, ‘it is what it is’.
If the goal is to appease and reduce the number of complaints/questions, then I’d first observe that virtually all of the complaints come from Europe and are related to activations in Europe. RBNGate was actually developed to address a specific combination of circumstances that create issues for activators in North America (and possibly Australia), but this combination of circumstances doesn’t exist in Europe: low chaser population + poor SMS/APRS coverage = activators spending far more time on the summit than necessary, perhaps so much time that their health and safety could become endangered as a result.
Europe’s much higher population and density of chasers means that RBNGate provides less value on your side of the pond - your chaser population can locate and spot the activators much faster than our smaller population can. Therefore, if appeasement is really necessary, my counter proposal would be to simply exclude all European activations from the RBNGate system.
P.S. - Whenever I hear “appease” I think back to when I first encountered that word as a child, while watching one of the BBC’s World at War episodes. They were making the point that Mr. Chamberlain’s policy of appeasement didn’t work out so well back then. I’m not sure that such a policy would work out any better today. 
73,
Eric KU6J
===========================================
Free SOTA Spot Monitor Software + RBNGate FAQ:
http://www.ku6j.com
In reply to G4OIG:
Hi Gerald,
Self-spots only cancel the alerts that are currently active. The standard alert window begins 1 hour before the ETA and ends 3 hours after it.
For example, in this particular case F6HBI’s alerts were for 1000z and 1200z. The self-spot was received at 1122z. 1122z is less than 3 hours after 1000z so the first alert was active. 1122z is less than 1 hour before 1200z so the second alert was active as well. The self-spot therefore cancelled both of these alerts and created a new one to replace them. Had there been another SOTAWatch alert in place for 1300z (for example) it would not have been affected.
73,
Eric KU6J
===========================================
Free SOTA Spot Monitor Software + RBNGate FAQ:
http://www.ku6j.com
In reply to KU6J:
one downside of always
spotting the summit as ??/??-??? is that it would eliminate our
ability to filter spots and trigger audio alerts based on the
geographic region of the summit.
Why not just eliminate the last part of the ref? So all spots would be of the form xx/yy-???
This would satisfy both camps, people could tell at a glance if they were likely to work the station yet they would also have to get the complete ref from the operator.
Colin G8TMV
In reply to KU6J:
I forgot about Spot Monitoring. Yes that would make a vague spot less useful.
I like Mike’s expression! I don’t have a problem personally, it spots me when I’m out activating. That’ll do fine. I only play with other wasys of spotting because I’m not 100% CW and it’s still nice to be able to stir up a wee pile-up on the voice modes. If I was 100% CW I’d not need anything else.
Andy
MM0FMF
In reply to KU6J:
Thanks for the detailed explanation Eric. Now I have an additional fact in my quest to understand, and utilise, the various facets of RBNgate behaviour.
As has been stated numerous times, the logic is fine for the vast majority of SOTA activations where one summit and one callsign is involved. Where operations involve multiple summits and multiple variations of callsign (as with F6HBI yesterday), or part SOTA and part non-SOTA (as with me in January when I started this topic), things get difficult.
I’m of the opinion that, for multi-summit activations, alerting is not ideal as it only complicates the process. For such situations I will simply self-spot at the start of each separate activation and let RBNgate take care of frequency changes.
When I have another part SOTA and part non-SOTA trip I will either rely on being found and spotted by “real” chasers, or try to remember the RBNN tag in my self spots.
Thanks again Eric.
73 de Les, G3VQO
The current arrangements are just fine. Chasers must take responsibility for verifying the information in any auto-generated spot, just like they must take responsibility for making sure they are actually in QSO with the spotted activator. Some don’t/won’t, such is life.
If I was 100% CW I’d not need anything else.
Not 100% true Andy. Try getting an RBNgate autospot on 24MHz during the local mornings!
Tom M1EYP
In reply to KU6J:
Many thanks for the clear explanation Eric which is very much appreciated.
Personally it looks as though there are situations where I could be affected by a self-spot placed at the end of an activation on one summit causing issues for the RBN on my next one. I just wonder whether the 3 hour window is too long and whether it could be reduced, maybe to 2 hours.
Whatever the period set, it would not have affected the outcome for Gerald F6HBI as he only had a very short distance to cross the border and reset his station. We don’t have this scenario here in the UK.
73, Gerald G4OIG
In reply to M1EYP:
The current arrangements are just fine. Chasers must take
responsibility for verifying the information in any auto-generated
spot, just like they must take responsibility for making sure they are
actually in QSO with the spotted activator. Some don’t/won’t, such is
life.
And there we have it in one!
When RBN goes wrong, blame the chasers.
73 Mike
2E0YYY
In reply to 2E0YYY:
Not at all Mickey. Just the same as using the DX Cluster properly. You need to be sure that what a 3rd party (or a computer) has spotted is correct and is what you are hearing.
Good practice. It isn’t just the RBNgate that gets it wrong occasionally. Chasers sometimes spot the wrong reference or callsign, and even activators have been known to self-spot the wrong reference.
Any good op will listen on the frequency to hear for him or herself all the necessary details and not just rely on the spot, be it human or RBN generated.
Tom M1EYP
In reply to KU6J who wrote among other things:
“Always spotting ??/??-??? would also inconvenience chasers with limited time who are primarily interested in improving their Unique Summits and Mountain Hunter award scores. They would no longer be able to make the chase/don’t chase decision by glancing at the spots, and would instead need to slog down to the shack, turn on the radio and wait for the activator to send their summit reference before even deciding whether or not to enter the pile-up.”
Relying on the spots, whether on RBN, the DX cluster, or SOTAwatch is not something I would recommend anyway. Too many sources of errors, except in the case of self spots, and if you haven’t got a computer in the shack, they are of little use, too. It is also possible that there are several active stations around the same frequency, some may be spotted, others not … Whom do you hear, whom can you work? The answer may be completely different from what you have read on the computer screen!
In the end, spots are just an indication of some activity on some frequency. They do not relieve the chasers from listening carefully, or the activators from identifying themselves in accordance with the rules. Stating the activator’s location every few QSOs also helps both sides - chasers won’t have to wait so long for a confirmation, activators are not inundated with as many questions.
73, Jan-Martin
In reply to M1EYP:
The current arrangements are just fine. Chasers must take
responsibility for verifying the information in any auto-generated
spot, just like they must take responsibility for making sure they are
actually in QSO with the spotted activator. Some don’t/won’t, such is
life.
Playing Devil’s Advocate here, I’d bet that chaser accuracy would improve immensely if chaser points were only valid when that elusive * was present!!!
73 de Les, G3VQO
In reply to G3VQO:
Playing Devil’s Advocate here, I’d bet that chaser accuracy would
improve immensely if chaser points were only valid if that mysterious
- was present!!!
Indeed. But chasers’ total scores would drop considerably too, because some activators are less than meticulous in uploading their logs to the SOTA database. Having said that, it would be the same for everyone, so it might not be a bad thing.
BTW, more than 90% of my activations are on CW but I do not use RBN-gate … and I never seem to have any difficulty in attracting attention by calling CQ on one of the recognised SOTA frequencies (or self-spotting if convenient).
73,
Walt (G3NYY)
In reply to KU6J:
It seems most of the people who do not like RBNgate are activators. Chasers can just decide to disregard the RBNgate generated spots if they wish so.
The activators can avoid to be spotted by the RBNgate by not posting an alert or simply adding a few characters to it.
For the activators who do not want to use these two options, couldn’t RBNgate implement an opt-out mechanism so that the activators who choose to opt-out are never spotted, maybe unless they specifically add a new RBNgate command in their alert to request to be spotted during that activation ?
Wouldn’t that be an answer to most of the complaints ?
Christophe
ON6ZQ
In reply to ON6ZQ:
For the activators who do not want to use these two options, couldn’t
RBNgate implement an opt-out mechanism so that the activators who
choose to opt-out are never spotted?
It already exists, Christophe. You just send an email to KU6J and he adds your call to the exclusion list. I have already done this, so I am never spotted by RBN-gate.
73,
Walt (G3NYY)