Peter Shaw wrote:
Thanks for the example Peter. Thought I would ask (in case I've missed
something simple!) about a couple of scenarios where this algorithm might be
used.
Scenario: I am a call centre operator taking an account number over the
phone.
Use: When we informed customers of their Account numbers we gave them
6-digit numbers instead of 5, this allows us to check (when we try to
retrieve their details) that the account number we typed in is a) valid, b)
most likely to be the one they told us.
This is actually very useful. I should point out that you can place the
check digit in any position in an account number. You just have to know
at what position it is when you verify it. Currently, the Verhoeff
check digit CFC used the last position as the check digit location.
Would anybody find it useful to have a default argument to place it else
where? To specify where to put it on compute() and where it is on
verify()...
Scenario: I am a data entry operative with a whole pile of forms to key in
with survey information about people's height and weight.
Use: Check digits cannot be used for this type of information.
Correct, unless the original information contained the computed check
digit. However, if that was the case - why would we be keying in the
information a second time? ;-)
Scenario: I am a data entry operative with a whole pile of forms to key in
with encoded survey responses.
Use: Because we were clever when we set the survey, the responses already
contain a calculated check-sum digit which allows validation on entry.
To me at least, these all seem slightly different from a MOD10 check on a CC
number where the check sum is always 0 rather than an extra digit that is
included with the number. Am I following along correctly?
Yes Peter, you are correct. This is a slightly different schema that is
more robust than the IBM check. Before the introduction of the Euro,
German Banknotes used a modified Verhoeff check digit system in the
serial number. They also used alpha characters that had a numerical
equivalent - this did reduce the effectiveness of the Verhoeff check a
little, but gave them more unique serials.
Thanks,
Peter
As I wrote the other night, I am going to release an number encoder CFC
- which will take alpha characters and convert them to numeric
equivalents in order to compute() a check digit. Also, it will have the
ability to compute a date (ex. 31/12/2005) into a 4 digit alphanumeric
equivalent. This is most useful to encode extra information into a
number - invoice and order numbers especially.
.Peter
--
Peter J. Farrell :: Maestro Publishing
blog :: http://blog.maestropublishing.com
email :: [EMAIL PROTECTED]
phone :: 651-204-0513
I've learned that I don't suffer from insanity, I enjoy it.
--
----------------------------------------------------------
You are subscribed to cfcdev. To unsubscribe, send an email to
[email protected] with the words 'unsubscribe cfcdev' as the subject of the
email.
CFCDev is run by CFCZone (www.cfczone.org) and supported by CFXHosting
(www.cfxhosting.com).
An archive of the CFCDev list is available at
www.mail-archive.com/[email protected]
