# Stadef and stacst questions: am I missing Skolen constants?

**URL:** <https://discourse.ats-lang.org/t/stadef-and-stacst-questions-am-i-missing-skolen-constants/613>\
**Category:** Google Group\
**Created:** [December 13, 2015, 3:06am UTC](https://discourse.ats-lang.org/t/stadef-and-stacst-questions-am-i-missing-skolen-constants/613 "2015-12-13T03:06:45Z")\
**Posts on this page:** 1\
**Page:** 2

<div class="post-metadata">

**Author:** ![Yannick\_Duchene](https://avatars.discourse-cdn.com/v4/letter/y/b77776/32.png) [@Yannick\_Duchene](https://discourse.ats-lang.org/u/Yannick_Duchene)\
**Post date:** [December 13, 2015, 8:18pm UTC](https://discourse.ats-lang.org/t/stadef-and-stacst-questions-am-i-missing-skolen-constants/613/21 "2015-12-13T20:18:32Z")

</div>

> […]
> 
> But the proof function was saying the other way: given all values in a  
> character range, they can always be represented with this form of UTF-8  
> string. And for this, a relation between b1 and b2 was required, as c  
> imposes this relation. So your way was the only one possible for this proof  
> functions:  
> […]

I tried to continue to investigate on my initial idea, and there is more. I  
won’t tell all of of what get trough my mind, but it can be summarized.

Given this:  
dataprop P =  
| P1

dataprop Q =  
| Q1 of P

One can say “if you have a P, you can get a Q”:  
prfn pf (p:P): Q = Q1(p)

But although it could intuitively make sense, there is no way to say  
something like “if you have a Q, you could got a P”:  
// For the picture, not possible!  
prfn pf (q:Q): [p:P; q == Q1(p)] void

Investigating more, I believe that’s also this kind of impasse, I was in.

[Previous page](https://discourse.ats-lang.org/t/stadef-and-stacst-questions-am-i-missing-skolen-constants/613.md?page=1)
