AimThere is already quite good documentation available for users to learn how to interact with Isabelle and to use it for proving theorems (see here). However, the entry barrier for users to program on the ML-level of Isabelle is still unbearably high. In this project we want to change this state of affairs by providing a tutorial about ML-coding in Isabelle. At its centre we describe the implementation of a simple prototypical Isabelle package (we describe a simplified version of the inductive package). We hope by documenting every aspect of this package that in the future you will be able to modify, extend and adapt this package for your own purposes, see how developers code in Isabelle and understand better Isabelle's idiosyncrasies. As a result you will be able to quickly produce code for your own needs and will be able to participate in keeping Isabelle alive in the long run.
Specific topics of the tutorial we have not yet covered:
Please help us to expand on this list.
We need YouWe need help for accomplishing this tutorial. If you are familiar with the ML-level of Isabelle, please get in contact with us (see addresses below). If you are not familiar, but like to know more about the bits and pieces that make up the Isabelle code, then let us know what you are interested in or what project you like to implement. Above all we like to help future users and developers of Isabelle; we do not want to end up with some "artificial documentation" that is of nobody's help. Also proofreading and feedback would be most appreciated.
ContactI am trying to stay on top of things, but no promises:
RepositoryA repository is available and you are very welcome to contribute to it. The current draft of the tutorial is here.
Stable VersionThere is no real stable version of the Programming Tutorial yet. If you need a version that is guaranteed to work with a stable version of Isabelle, then please get in touch with me (christian.urban at kcl ac uk).
My path towards HOL was from HOL Light, to HOL4, to Isabelle/HOL and the Cookbook was essential for the last step (Piotr Trojanek, University of Bristol, 2014)
...and the Isabelle/ML Cookbook are valuable resources (Lars Hupel, 2014)
GREAT!!! That's exactly what I was looking for! (Thomas Genet, ISTIC/IRISA, 2012)
The tutorial really is quite nice. (Jeremy Avigad, CMU, 2011)
By the way, thanks for the Isabelle Cookbook, it's really helpful to start with Isabelle/ML. (Mathieu Giorgino, 2011)
Great! This seems to have a lot of the missing pieces I couldn't find in the tutorial and reference manual. (Kevin Van Horn, 2011)
I've been fighting for hours trying to derive some theorems automatically, until Florian told me I should try using conversions and then I read the Cookbook section and it solved my problem. The example with abs_conv in particular was particularly illuminating. Thanks! (Jasmin Blanchette, TU Munich, 2010)
The cook book seems really helpful, if not even exactly what I need. (Nils Jähnig, FU Berlin, 2010)
The Programming Tutorial has helped me tremendously to get familiar with Isabelle, which I used in my bachelor thesis about the type-inference algorithm in Isabelle. (Dmitriy Traytel, TU Munich, 2010)
I've already pointed a couple of PhD students at the programming tutorial, and they've already found it to be a great place to start with Isabelle programming. (Lucas Dixon, University of Edinburgh, 2009)
The Isabelle Programming Tutorial is great! I finally learned how to add new keywords and how to parse outer syntax! (Amine Chaieb, University of Cambridge)
I felt the programming tutorial was very clear and comprehensible. I'm toying with actually getting back to Isabelle again. Thanks to everyone involved in the documentation project. I really think it is worthwhile. (Tom Ridge, University of Cambridge, 2009)
I learned from the tutorial how to deal with fresh variables and also which function to use for applying substitutions that are generated by unification. (Lukas Bulwahn, TU Munich, 2009)
What a great resource! I wish I'd had this when I wrote my first parser in December 2008. (Timothy Bourke, UNSW Sydney, 2009)