Interviews: Alexander Stepanov and Daniel E. Rose Answer Your Questions

[Rikki Cattermole via Digitalmars-d]

On 25/02/2015 9:28 p.m., Vlad Levenfeld wrote:
> On Wednesday, 21 January 2015 at 22:21:57 UTC, H. S. Teoh wrote:
>> ...
>> T
>
> Stepanov's book only goes so far as to describe coordinates for
> traversing a binary tree, but the extension to n-trees is pretty
> straightforward. Have you given any more thought to these topics? I've
> built a few of the solutions you mentioned over the last few months but
> now I'm working on putting it on more rigorous foundations.
>
> So far its come down to defining a notion of spaces (distinct from
> ranges though many structures are both ranges and spaces), classifying
> data structures by their topologies (e.g. traversal by links, by
> modifying a coordinate and passing it to opIndex, by incrementing and
> dereferencing an iterator, etc), and characterizing the various
> topology-transforming functors on ranges and spaces (like flattening a
> tree into a singly-linked-list or calling .array to transform an input
> range into a random-access range), but its pulling me in a lot of
> different directions and I'm having a hard time culling the options
> down. I do have a nearly complete space package that I use for general
> projects, but otherwise nothing too concrete yet; just pages of
> scribbled notes.
>
> Anyway this is the topic of my thesis (senior research project for
> discrete math) and, since I'm implementing it in D, it would be cool if
> some of the work I do can contribute to the ecosystem. If there's
> interest I'll keep you apprised of updates and if you're willing I'd
> like to pick your brain about use cases and feedback sometime.

If you are able to, create a e.g. github repository with your code / 
notes ext.
That way anyone that wants to keep updated can follow it and read the 
changes as they happen.

After all, version control isn't just for source code ;)