Re: [AM] Database "refactoring" effort question

(responding to Scott)
>
>>My point is that you are creating two columns of *new* data.
>
> Where is the new data coming from?  If I had one column with
> "Ambler Scott" in it and new have two columns, one with "Ambler"
> and one with "Scott" there isn't any new data. This is the extent
> of the split column refactoring.  Anything you do beyond that is
> something else.

IF, and it hasn't yet been decided, that's what you do, then
the new data is generated from the old data plus a bit (or
more like 7 bits as I recall my Information Theory) of 
extra information.  It takes 1 binary digit to say whether
the front or back part of the original name goes into the
Surname field, and it takes around 6 binary digits of
information to say where to split the original data field.
This information needs to come from somewhere.

>> It is possible that there is some arbitrarily complex rule that
>> can derive the new data from the old data that you already
>> have, and that this may turn out to be an acceptable way
>> to create the new data, with an acceptable level of dirty
>> data created in the process.
>
> Yes, and it won't be pleasant I would think.

True, either the algorithm or the results would be nasty,
possibly both.  A good algorithm could derive the information
from that available, based on knowledge of names.  Not
all names will succumb to such treatment.

>>   Yet there is nowhere in the
>> system that contains all the data needed to create the new
>> data from what is already there.
>
> Why not?  Source data + rule ==> refactored data.

The rule isn't in the system yet.

> Yes, you may also decide to go in by hand at some point
> and clean up the dirty data, but that's above and beyond
> the refactoring.

You can still only get the data so clean before you need to
contact the individuals whose names you are trying to
decipher.

>>   In populating the new
>>columns, one needs to add information that is not yet
>>available.
>
> Like what?

Like the 7 binary digit's worth of information mentioned above.



>> For the
>> 10,000,000 records quoted, you need to create one new
>> piece of information each - at minimum.
>
> What information?

The 7 binary digit's worth of information mentioned above.


Paul Oldfield

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
www.aptprocess.com

any opinions expressed herein are not necessarily those of
Mentors of Cally or the Appropriate Process Movement
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

For more information about AM, visit the Agile Modeling Home Page at www.agilemodeling.com
--^----------------------------------------------------------------
This email was sent to: [email protected]

EASY UNSUBSCRIBE click here: http://topica.com/u/?bUrKDA.bWnbtk.Z2NtYS1h
Or send an email to: [email protected]

TOPICA - Start your own email discussion group. FREE!
http://www.topica.com/partner/tag02/create/index2.html
--^----------------------------------------------------------------