Order Tray | Contact Us | Home | SIG Lists

[aprssig] New Digi Settings

Rick Green rtg at aapsc.com
Thu Jun 9 14:02:33 UTC 2005


On Thu, 9 Jun 2005, AE5PL Lists wrote:

> The amateur radio community has taken a link level protocol (AX.25) and
> tried everyway they can to abuse the path capability built into it.  In
> ALL cases, that abuse (attempting to make it act like a network level
> protocol) has ended in abject failure.
    /snip/
> If you really want to "support" the users, make it so they don't have to
> know _anything_ about paths.
   /snip/
> You have your "new" paradigm.  It has holes, as has been pointed out, so
> now lets turn our attention to how we can take APRS to the next level.
> And the way you do that is get rid of digipeating based on path.  Many
> "smart digipeater" authors are already implementing (or have
> implemented) the no-source-routing algorithm shown in the Spring 2005
> PSR.  It might be worthwhile for you to expend some of your efforts with
> Kantronics to get them to do the same.
>
  Hear! Hear!
  A while ago, maybe 6 months or a year, as Bob was ranting about some 
users abusing 144.39 by specifying ridiculously long paths, I suggested 
implementing a digi which is sensitive to the position in the packet, and 
simply digis packets within its sysop-defined 'service area' and drops 
others on the floor.
   While he was ranting about 'abusers' of the frequency, he also chose to 
attack my proposal, insisting that *he* have the ability to specify source 
routing so that his mother could track his progress over the mountains and 
thru the woods on the way to Thanksgiving dinner.

   I must admit that after experiencing this hypocrisy, my eyes almost 
totally glazed over, and my reading of this list has been extremely 
sporadic ever since.

   I heartily agree with you that source routing is evil, and should be 
eliminated from the APRS spec.  As we continue to build a real network on 
144.39, lets implement network routing.

   I'll add one other pet peeve to this rant:  APRS messaging must go.  It 
seems totally ridiculous to me to (ab)use what is essentially a broadcast 
protocol, for point-to-point communications, with retries and 
application-level acknowledgements just adding more clutter to an already 
saturated channel.  Simple AX.25 is the mode for point-to-point 
communications.  With the increasing sophistication of radio/TNC/software 
suites, isn't it about time to implement an auto-QSY off the 'calling 
channel' when you wish to have a private conversation?

-- 
Rick Green

"They that can give up essential liberty to obtain a little
  temporary safety, deserve neither liberty nor safety."
                                   -Benjamin Franklin




More information about the aprssig mailing list