[aprssig] IGate Registering for ANSRVR (Messaging failures)
steve at dimse.com
Fri Jul 15 21:44:17 CDT 2016
> On Jul 15, 2016, at 7:03 PM, Pete Loveall AE5PL Lists via aprssig <aprssig at tapr.org> wrote:
> Please reread both cases. #1 Yes, it is more broken if the HT user doesn't know why they are unable to message. #2, this has nothing to do with javAPRSSrvr. It has to do with IGates connected to local (usually home-based or collocated) servers which are connected to the APRS-IS via a restricted feed. This includes integrated IGates but is not exclusive to integrated IGates. The RO and RW IGates are connected to different servers as originally stated.
> Nothing needs to change and I purposely did not mention your private email to me.
It was not meant to be private, thanks for pointing out the mistake. Included here
> On Jul 15, 2016, at 12:55 PM, Pete Loveall AE5PL Lists via aprssig <aprssig at tapr.org> wrote:
> Whoever came up with the "receive-only" IGate as a good thing did not think through that APRS is a -communications- protocol, not a tracking protocol and they did not fully understand the potential for breaking the communication mechanisms that APRS-IS depends on.
I came up with it, and I did think it through.
First of all, satellite IGates should be receive only, they cause QRM otherwise.
Second, there are two different uses for the APRS-IS. The first is to get RF packets to the internet. This came first, and numerically is responsible for a far greater percentage of the use of the APRS-IS. Two way messaging is the other use, and one-way IGates are indeed not useful for this purpose. However they do not break the communications mechanisms. The only potential adverse effect is that someone considering installing a two way IGate in an area already served by a one-way might not realize their service would still be valuable (since local stations already appear on the APRS-IS), and not set up the needed service.
> Why not configure your home station for direct-only bidirectional and then you can support messaging instead of only sometimes being able to communicate reliably?
It is interesting that this suggestion still runs afoul of the one reason I give above why one way IGates may hurt the functioning of the system. A ham wanting to set up a full, multi-hop two way gate might be discouraged by a zero hop IGate, since all the traffic, regardless of number of hops, is sent to the APRS IS.
Third, I have always believed strongly, and acted on the belief, that the choice of what goes on a local RF network is the choice of the local hams and must not be dictated by any national or international group no matter how well-intentioned. If a local area decides for any reason it does not want 2-way messaging on its network it should not then be forced to remove all its traffic from the APRS IS.
Fourth, IGates are illegal or require special permissions in some countries. We should not exclude people who live under less enlightened rules from placing their traffic onto the APRS IS if the rules allow RF to INet but not the reverse.
Fifth, and very importantly, is the risking of our ham licenses. The deliberately weak authentication mechanism built into the APRS IS was designed only to provide cover for IGate operators from FCC rules. Once that mechanism was published by aprsd in the name of open-source purism that cover evaporated. Every IGate operator puts their license at risk by opening a two-way IGate. The risk is small, but again, the only alternative should not be non-participation.
> Satgates are purposely receive-only and are not part of this discussion as the satellite frequencies are purposely not gated to from APRS-IS. You have failed to give a single -communication- reason why receive-only IGates are worthwhile,
If you define communication as necessarily two way, true, one way IGates are not communication-worthwhile. But I don't define communication that way. The APRS-IS was not two way for its first two years, and even today two-way comms represent a small minority of the traffic. If you really believe this why not ban everything except the two-way messages? Given my definition that the gathering and aggregation of APRS activity from around the globe and distributing via the internet does constitute communication, then every one of my five reasons above DOES improve communication.
> especially since they can and do interfere with APRS messaging (as shown by my 2 examples).
Again, common sense says delivering a message that otherwise would not cannot be called "interfering" with communications; your first example actually shows a circumstance where a one-way IGate improved communication as I define it.
Your second example makes no sense as you have described it. Nothing "upstream" should care about which IGate passes a packet to an intermediate hub, nor should what an upstream hub knows about who passed a packet to the APRS IS affect what any downstream hub or IGate does. Maybe there is something that has not been explained clearly, but if I don't see it I doubt anyone else can either. You failure to explain speaks volumes.
> APRServ was 16 years ago and didn't provide the necessary filtering as demanded by today's users, clients, and Internet connections. Your intent was good but it did not survive the demands of a greatly expanded APRS-IS. Roger's user-defined filter software in conjunction with javAPRSSrvr's restricted feed does and has been expanded and improved over the past decade both with javAPRSSrvr and aprsc.
I long ago admitted filtering was useful. But it needs to be done right, and if something in the filtering and or restricting code is causing a problem that did not exist as the system was originally designed, then the answer is to fix the problem in the new code, not to remove one-way capability from those users that do find it useful.
> I can't say it any clearer than I have already put forth. Do as you will but understand that receive-only IGates can and do interfere with APRS messaging as shown in my post. If they make no sense to you, then you are either unwilling or unable to comprehend the issue. To continue to promote receive-only IGates on the standard APRS frequencies is to knowingly promote IGates that can and do interfere with APRS messaging.
Your post DOES NOT SHOW ANY PROBLEM. And the scenario strongly implies that if there is a problem that you haven't explained clearly enough then the problem is in likely in your own design/code. You seem to hope that people will just accept your great knowledge and bow to your pronouncements, but when challenged you talk about having other examples, but you won't show them, you offer one example that is wrong by common sense alone, and another that makes no sense to the guy that designed the system, and when he asks for you to explain clearer, you run away.
I hope what is going on is as clear to everyone else as it is to me. I call shenanigans!
More information about the aprssig