[aprssig] DAYTON or other Event planning

[Rick Green]

On Mon, 23 May 2005, Robert Bruninga wrote:

> Here is how things like Dayton should be done:
>
> 1)  Support users as-is.  Its impossible to get everyone
>     to change just for an event.
   I agree.  That's the goal.

> 2) Set up a local IGate within RF range of the event
     Agreed.  Within RF range of the pockettrackers especially.

> 3) Set up an alt-input digi from 144.99 to 144.39
     I disagree.  Don't crowd .39 with more traffic.  Set up a .99 input 
port on the igate instead.  If you want to get more complete coverage, add 
input ports on all the 2m simplex freqs, looking for mic-e packets as 
well.

> 4) Set up ALL digis that can hear the event direct
>    to digi ONLY once.  That is:
>    - Set UIFLOOD and UITRACE OFF (no WIDEn-N)
>    - Set UIDIGI or ALIAS list to WIDE1-1,WIDE2-2,WIDE3-3
   When channel loading gets near saturation, the digis should be turned 
OFF, and allow the igates to send the traffic OUT of the area, but not add 
congestion by bringing any traffic into the area.

   I still think that digis should be smarter.  They should be sensitive to 
the position in the packet, and should refuse to digi any packet that 
originates outside of the 'service area' as defined by the sysop.  With a 
little more work, this 'service area' could be variable, depending on the 
current channel loading.  As the channel gets busier, the digi gets more 
provincial about what it chooses to repeat.
   With this type of smart digi, we could truly have a universal path spec. 
Each packet would travel as far as the network would allow, and be 
truncated when it got to an already congested area.

-- 
Rick Green

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