Foodcodes, Was: Re: [Babase] FPSAMPS Kidcontact and Kidsuckle blanks

Karl O. Pinc babase@www.eco.princeton.edu
Sun, 7 Nov 2004 18:31:00 -0600


On 2004.10.27 06:10 Susan Alberts wrote:
> This is a very belated reply, sorry.

Likewise, likewise.  :-)

> 
>> What does "-  If "Other social" is recorded in the point sample, the 
>> nature of the social  interaction is recorded..    -  If the food 
>> being eaten is Other or Unknown, notes on the food are recorded."
>> mean?  That is, what's recorded where?  Is this a demography note?
>> I didn't think Psions took comments.
> 
> 
> There is an option to enter an extended comment. They press *, then 
> O, then describe the interaction. There was no way to get around this 
> one relatively free-form type of entry.

Ah, right.  These are the ADLIBS table.  Not a problem.


> 
>> 
>> Which reminds me, arn't foodcodes single character codes that
>> are 'lumped' together as a muti-character string?
> 
> 
> Yes basically although it is not the case that each column within the 
> field is equivalent across food codes.
> 
>> This will make them hard to validate and I'd imagine odd
>> to query.
> 
> 
> They are odd to query, it is not at all trivial. However, the codes 
> themselves are part of a rather deeply entrenched system of food 
> codes that were started decades ago by Stuart so i don't think 
> changing is a good option at this point.

I'm making them into a table.  This makes them easy to validate and
you can add new ones when you like.  We can make additional columns
in the food code table to make things easier to query if you've
got catagories, or rather alternate-catagories, as I've added a Ftype
column and another support table (FOODTYPES) to support the
foodcodes.  This should help with querying.

I'm using your attached document to populate the FOODCODES and
FOODTYPES table for the conversion.

Karl <kop@meme.com>
Free Software:  "You don't pay back, you pay forward."
                  -- Robert A. Heinlein