Subject: Re: [boost] [Boost.Functional][GSoC] Boost.Functional / Monad project proposal
From: Vicente J. Botet Escriba (vicente.botet_at_[hidden])
Date: 2014-03-17 17:13:58
Le 17/03/14 06:09, Benjamin, John C. (MU-Student) a écrit :
> Project Proposal
> The proposal is not too creative, just what is described in Boost.Functional / Monads
> Please provide a description of your proposed work.
> I will work to make sure the Monad implementation meets the actual description of a monad. One thing I think is important is to be able to have recursive monads that are efficient. The IO monad for example is particularly amenable to recursive structures that allow for easy sanitation of input and output.
> Proposed Milestones and Schedule
> Rough draft of milestones and schedules:
> May 19 - June 2 : review Monads and Category Theory. Understand the current state of Boost.Functional / Monad. Read through Pure and discuss how our implementation is different.
> June 2 - June 30 : Propose, discuss, and finalize working prototype which implements all the required existing instances of monads and meets syntactic conventions. Begin coding
> July 1 - July 21 : Implement finalized proposal and prepare for testing
> July 21- end of summer : test and document my work
> Please provide any feedback. If you need more information or a more specific proposal please let me know.
> John Benjamin?
If the proposal is not too creative, why do you want to spend all the
summer on it?
The monad project can be short or very large. All this depends on what
you want to do.
Your project proposal do not contains enough information. Even if monads
are something well known, you need to state what do you plan to
implement, solve, ...
How C+11/C++14 would help you?
In addition you can not spend the first 1/4 of the project understanding
the problem domain and start testing and documenting on the second half.
A more iterative schedule would be more appropriated.
Have you installed Boost, are you able to build the libs, run the test?
Have you tried to do a probe of concept? using the Boost tools?
See also the other proposal thread.