[aprssig] are write-only APRS-IS clients valid?

[Pete Loveall AE5PL Lists]

Tom, you are correct.  RX-only IGates break the -current- intent of APRS-IS: to interconnect amateur radio APRS RF networks.  They break this intent because they do not support messaging.  They also can break messaging for nearby bidirectional IGates using filtered feeds from the upstream server.  The reason this last fact is true is because of dupe elimination present in APRS-IS servers.  If the RX-only IGate is quicker at getting packets to APRS-IS, the upstream server of the bidirectional IGate may never see packets gated by the bidirectional IGate.  Therefore, the bidirectional IGate will never see messages for APRS stations it sees on RF.

The statements of Steve "An rx only IGate is better than no IGate in almost all cases." and "If nothing else, all ARISS IGates should be receive only." are invalid because they do break messaging, even for bidirectional IGates.  Unlike "fill-in" digipeaters which provide RF coverage for areas that are dead spots to local wide-area digipeaters, IGates normally depend on local wide-area digipeaters for their "ears".  A receive-only IGate provides no added coverage and can interfere with the proper operation of a bidirectional IGate.

I highly discourage receive-only IGates as they break APRS messaging to the RF area they cover.  If your belief that APRS-IS solely exists to support database servers providing tracking services, the RX-only IGates are fine.  However, if you believe that APRS-IS is to support and interconnect amateur radio RF networks, RX-only IGates are not fine.

73,

Pete Loveall AE5PL
pete at ae5pl dot net


> -----Original Message-----
> From: Tom Hayward
> Sent: Sunday, December 01, 2013 11:00 PM
> 
> On Sun, Dec 1, 2013 at 8:37 PM, Steve Dimse <steve at dimse.com> wrote:
> >> This breaks two-way messaging. The APRS-IS assumes that the igate
> >> that gated a position packet for a station can also relay a message
> >> to that station.
> >
> > The APRS-IS assumes no such thing.
> 
> I didn't mean to imply that the server software is making routing decisions.
> We rely on each igate software (each part of the APRS-IS) to gate
> appropriate messages to RF. If an area is served by an igate that doesn't do
> this, then it breaks two-way messaging for out-of-area messages (including
> messages that originated on an altnet, which is common in my area).
> 
> >> In RX-only igates, this assumption is false and the message/ack may
> >> not reach its recipient.
> >
> > If an area is ONLY served by an rx-only IGate, then of course internet
> originated traffic will not appear on RF - both messages to the local RF
> stations as well as ACKs for their messages to internet stations. But that is in
> no way related to any assumption by the APRS-IS.
> 
> No need to break my paragraph apart. This was my original point.