Re: I-D Action: New draft - draft-bellis-enum-send-n-00

Otmar Lendl said:
> Your definition of the min and max fields are relative to the location 
> of the NAPTR record:
> 
>     The 'digitsmin' field of the data MUST correspond to the minimum
>     number of additional digits to be dialled, relative to the current
>     record, which might result in reaching a full ENUM record in the ENUM
>     database.
> 
> 
> Wouldn't it be easier to use absolute values here?

No.

(1) This record corresponds to equivalent functionality in ISUP, which uses
the relative case.

(2) Making it relative means that the record value depends only on records
lower down the tree, not where the record is in the tree. If it's absolute,
then moving a sub-tree could alter the required value.

For example, my home telephone exchange has changed over the years from
        +44 954 8xxxx
    to  +44 954 78xxxx
    to  +44 1954 78xxxx
The latter change also involved an overlap period.

Under this proposal, the send-n record corresponding to the xxxx is
"pstndata:send-n/4-4" in all three cases. With your proposal, it would have
to alter from:
        pstndata:send-n/10-10
    to  pstndata:send-n/11-11
    to  pstndata:send-n/12-12

(3) The present proposal matches what the device consuming the record
actually needs to do. The above send-n record could be reached after I dial
any of:
    78
    0195478
    0044195478
In these three cases, the device needs a total of 6, 11, or 14 digits, but
in each case your proposal means it gets a record saying 12 digits. The
original proposal says it needs to get 4 more digits, which is always
correct.

-- 
Clive D.W. Feather  | Work:  <[email protected]>   | Tel:    +44 20 8495 6138
Internet Expert     | Home:  <[email protected]>  | Fax:    +44 870 051 9937
Demon Internet      | WWW: http://www.davros.org | Mobile: +44 7973 377646
THUS plc            |                            |