RS> Well you would be wrong. You appear to be thinking that you 
 RS> can just apply an accurately measured amount of the 
 RS> ingredients and thats all there is too it. In many 
 RS> situations you cant actually do that. What you do is add 
 RS> about whats needed and then finely adjust whats added extra 
 RS> to get the result you want.

And so you describe the result you're looking for, and an approximate
amount of ingredients.

 RS> For example most pastry cooking needs the pastry before 
 RS> cooked at a particular consistency. And the flour varys a 
 RS> hell of a lot more than you might think. So you add the 
 RS> liquid, water or milk, at less than will be needed, then 
 RS> gradually add a bit more in steps until the runnyness of the 
 RS> result is right after mixing it in. Particularly important 
 RS> with things like pancakes.

So describe the runniness (viscosity I think is the correct term), perhaps
with a description "put a blob on a metal tray, and see if it takes 10
seconds to travel 10 cm".  Also, put a caveat that it is difficult to
describe the exact runniness required, you will have to experiment, too
little runniness cause the pancake to come out flat, too much runniness
causes it to get burnt (e.g.).

 RS> With some things there are sophisticated techniques used to 
 RS> work out when the consistency is right too. For example when
 RS>  whipping egg whites for meringue type stuff, you have got 
 RS> where you want to be when removing the spoon leaves little 
 RS> stable peaks in the whipped stuff, they dont collapse under 
 RS> their own weight. The amount of whipping and amount of sugar 
 RS> required to get that depends on subtle things like the how 
 RS> old the eggs are and what temperature they were before they 
 RS> were whipped.

You've just done it again, described scientifically the result you're
looking for, and use that as the number of ingredients required.

 RS> OTOH I dont see thats a problem with a computerised form of 
 RS> receipe tho. Those problems are surmounted somehow in 
 RS> printed receipe books, so why not in an electronic one. In 
 RS> fact you can go much further with an electronic one. With 
 RS> the egg whipping for example you can have a popup window 
 RS> which explains the subtletys of deciding when its right.

Exactly, I've never heard of a recipe that couldn't be put into a cookbook,
so I figured I was pretty safe.

 RS> There is your problem. Its not just a matter of getting the 
 RS> measurements right. On many occasions the final result 
 RS> depends on the detail of the ingredient used. For example 
 RS> the amount of liquid needed depends on the detail of the 
 RS> flour used. And its not just simple stuff like plain and
 RS> self raising flour. There is a big variation in just say 
 RS> plain flour. Makes a hell of a difference for some cooking 
 RS> like breads.

 RS> I still dont think its an insurmountable problem. But does 
 RS> need to be allowed for.

The theory that it can only be handed down, not put into computerized form,
is total myth IMO.  Incidentally, the recipe was verbally translated going
from mother to girlfriend, and written down on a piece of paper too.  Both
of these things are easily computerized.  BFN.

Paul

--- msgedsq/2 2.2b