Dear Gyesik,

Thanks for your interest!
We are just in the process of updating the contribs so that corn compiles
with trunk again.
I will check again about the status. However, I believe the current version
should compile with 8.4pl4.

Please note that we have made a partial transition to the more modern use
of type classes, the math-classes library.
http://math-classes.org/
It is a submodule of:
https://github.com/c-corn
The older parts of the code still work, but you will not get the type class
magic.

There are many interesting projects. Personally, within constructive
analysis, I am very interested in combining proving with computation inside
Coq. If you are interested in this direction, as a very initial experiment,
one could try to port some more algorithms from Reals/Fast to Reals/Faster.
The latter is indeed faster and uses type classes.

Another project I thought about was fast numerical computation.
Here is a possible development in constructive analysis:
http://arxiv.org/abs/1202.3460

Another possible direction is the correctness proof of Newton iteration.
I can suggest some more depending on your taste.

Best,

Bas

PS: We have a small meeting planned with Nicolas Magaud, just after TYPES,
to connect the corn library with their library for non-standard analysis.
PPS: The homepage http://corn.cs.ru.nl/ is not up to date, we should really
do something about it.






On Tue, Apr 29, 2014 at 5:57 PM, Gyesik Lee <[email protected]> wrote:

> Dear C-Corn developers/users,
>
> I and my students are strongly interested in formalizing constructive
> mathematics.
> Although we are originally mathematicians and have some experience in
> using Coq, dealing with pure mathematics in Coq is new for all of us. So we
> are wondering whether the C-Corn project is a good starting point for us,
> which we hope strongly.
>
> Currently we are in the phase of thinking of and looking for some well
> known and interesting, but hopefully not so difficult mathematical theorems
> which we could set as our first target.
>
> We also looked at Freek Wiedijk's "Formalizing 100 Theorems" website, but
> we are not sure which parts of the C-Corn library provide a good base for
> some of the theorems mentioned there.
>
> We haven't yet checked all the contents of the C-Corn library. So we will
> be highly grateful when you can give some hints about or information of the
> current state of the library.
>
> Thanks in advance
>
> Gyesik
> _______________________________________________
> C-CoRN mailing list
> [email protected]
> http://mailman.science.ru.nl/mailman/listinfo/c-corn
>
_______________________________________________
C-CoRN mailing list
[email protected]
http://mailman.science.ru.nl/mailman/listinfo/c-corn

Reply via email to