Discussion List for Freenum/ITAD Subscriber Number (ISN) Project

[John Todd <>]

> A few things.
>
> I have also been troubled by the "012" "suggestion" as being too North
> American specific.  On the MIT PBX our current plan is to use "*0" .

The cookbook and other existing documentation lists "012" is a 
suggestion, not a part of any specification.  Each site can use 
whatever prefix they choose, just as MIT is planning on "*0".

> Jonathan, you use "694 200 2233" as an example - is 694 a country code?
>
> >                         Also padding at the start, for a number at the end is
> >  counter-intuitive to any users with experience of the regular PSTN. Once
> >  ITADs get longer this gets unwieldy, effectively adding an extra 0 for every
> >  extra ITAD digit:
> >
> >  123 * 329 = 000 123 329
> >  123 * 56789 = 000 00 123 56789
> >  5551212 * 234567 = 000 000 5551212 234567
> >  5551212 * 2345678 = 000 0000 5551212 2345678
> >  5551212 * 23456789 = 000 00000 5551212 23456789
> >  .. etc
>
> Does this work any better?
>
> 123 * 329 = 000 123 329 3
> 123 * 56789 = 000 123 56789 5
> 5551212 * 234567 = 000 5551212 234567 6
> 5551212 * 2345678 = 000 5551212 2345678 7
> 5551212 * 23456789 = 000 5551212 23456789 8
>
>                      Dennis

Well, trailing runlength digits I suppose will work just as well as 
leading runlength digits for those instances where dialing a * is 
impossible.  The use of an "ISN-only" prefix would work, such as the 
"000" suggestion.  Howver, it would preclude any other non-ISN 
numbering system from being used after the "000" without the addition 
of even more complex dialing rules.  Here's the logic:

IF the prefix is 000
   then
     IF the sequence following contains <number>*<number>
       THEN treat as normal ISN call. End.
   ELSE take last digit as runlength encoding digit and parse as 
number-only ISN

This implies that any number that does NOT contain "*" would be 
digested by the runlength logic, thereby making any other number plan 
here other than ISN difficult, if not impossible.  I'm not very 
frightened by that prospect, but that is a local dialplan decision.

Question: With runlength encoding, we're limited to a single digit to 
describe runlength. What happens if/when there are more than a 
billion ISNs?  Or is "0" the top of the scale instead of the unused 
"bottom"?   We could also consider special use of "1" and "2" since 
those are currently digit ranges that are "reserved for private use" 
and do not overlap with the allocated ITAD range.

JT