"Korleone" wrote in message
...
David Hare-Scott;797991 Wrote:
"Korleone" wrote in message
...-
Hello.
I have some simple questions about fertilizers.
For the last few days I've spent a lot of time figuring out ideal
profiles for fertilizers for indoor tomatoes.
I wrote a program that figures out how to mix separate fertilizers
together to better match your custom fertilizer profile. Which is
great (and bloody genius), but..
-
Why? What is the need for this?
Well..
If I wanted to experiment with a particular fertilizer profile but it
wasn't available commercially I could mix my own using separate
fertilizers. Figuring out how to balance them with pen and paper would
take a very long time to do accurately. So now, I can take any
fertilizers and find out the best possible way to combine them to get
the profile I want.
If you know what the deficiencies of your soil are and the needs of your crop
are element by element and you are going to align the two using only your
custom mix then I can see that your software may be useful. In practice none
of those three conditions are likely to be true.
One key thing you have left out is the existing nutrients available in
the
soil. Plants don't get only what is added as fertilser, nutrients come
from
many sources, some plants even have little mates that make nutrients
for them.
Also different soils bind nutrients to various degrees. This means
there is
no such thing as "ideal" tomato fertiliser or anything else fertilser.
And
texture and drainage make a huge difference. And lots more, this matter
is
much deeper than it first looks.
Yes, you're right. Soil is complex stuff. However, when I say ideal,
I don't mean a-cure-for-all-ills fertilizer, just the best ballpark
figure for my particular setup.
I want to be able to decide on a great N-P-K-Ca-Mg-S ratio, and then
only have to adjust the ppm as the plant grows.
I don't understand. You seem to be talking about two processes, is that
right? What is the first one where you get "great ratio"? How do you
subsequently adjust the "ppm", how would you know what it is at any time? Do
you think that the elemental content of your soil as determined by some assay
method is what is available to the plants?
I won't be able to use
it all of the time of course. I know that. But, I will be able to use
it most of the time in a controlled setup.
Controlled setup? Are we talking hydroponics here? That's rather different.
Writing your own software is admirable. May I suggest that whatever
you write
will be far more useful and you will learn far more from doing it if
you
understand the subject matter that you are modelling or work from a
specification written by someone who does. There is much more to good
software than kool kode and a fAnCy iNTerFAce.
It's definitely not about interface. 
Just a bog-standard number
cruncher.
It simply figures out the best matching ratios. It's not about
modeling at all, really.
Yes it is by my definition of 'model'.
It's not like I'm building a soil model. That would be completely
unnecessary and altogether useless.
I wasn't hinting that you ought to be modelling soil composition in relation
to growing plants and adding fertiliser in its entirety. However, any
software that supports decision making about a real-world situation is
modelling that situation in some way. It embodies certain data that are
collected out of a much larger conceivable set, it makes simplifying
assumptions, uses selected formulae, produces a set of results out of a much
larger possible set, etc. Modelling is about choosing to deal with a
manageable assembage of what is important to you out of a larger system and
showing (or trusting) that what you ignore will not mess it up too much.
I am suggesting that, for example, you may not have looked carefully at what
your assumptions are and what affect they will have on the real-world
usefulness of your results.
David