[Haskell-cafe] GSOC and "scripting the inference process" .Was: Re: Haskell compilation errors break the complexity encapsulation on DSLs

[Jurriaan Hage]

Hello,

I currently have a PhD student who is working on exactly this topic: extending the work of our
Scripting the Type Inferencer paper to full Haskell. His name is
Alejandro Serrano Mena. He started last November.

I believe the complications of being able to deal with type system
extensions currently available and in use in Haskell to be so invasive that it will 
take a few years to get the complete picture (considering type classes with extensions,
GADTs, type families etc.). 

A Google Summer Of Code for the Haskell 98 subset (ie. implementing that in GHC)
should be possible, I guess. Whether that implementation later scales to the full
language may be something of a concern.

 We, ourselves, shall be prototyping the new things in the Utrecht 
Haskell Compiler, and the plan was that after Alejandro has finished his PhD and
given a positive outcome of his work, that something should be arranged to make
it available to the general Haskell programming audience by implementing it into GHC.

best,
Jurriaan



On 17Mar, 2014, at 12:50, Alberto G. Corona <agocorona at gmail.com> wrote:

> Now that GSOC comes again, I call you into attention one of the biggest problems that Haskell has to solve before being fully accepted in the IT industry: The (unnecessary) complexity of the DSL error messages.
> 
> There is some work being done: 
> 
>  http://dl.acm.org/citation.cfm?id=944707
> 
> 
> The idea was to implement it in GHC after refining and testing it if I remember well. If some of this work can be carried out in a GSOC project, IMHO it should be given priority.
> 
> In case that this work is not going to be available in the short-medium term, 
> there is a ticket with some suggestions that can be carried out in a GSOC project, just in case any of you are interested into mentoring/carrying out this or any other solution to overcome this problem:
> 
> http://hackage.haskell.org/trac/ghc/ticket/7870
> 
> 
> 
> 2013-04-27 22:14 GMT+02:00 Ozgur Akgun <ozgurakgun at gmail.com>:
> Hi,
> 
> On 27 April 2013 10:07, Alberto G. Corona <agocorona at gmail.com> wrote:
> I created a ticket for the feature request:
>  
> Ticket #7870
>  
> Teachers, newbies and people working in Industry: Please push it!
> 
> A link to the ticket may be helpful for the lazy.
> 
> http://hackage.haskell.org/trac/ghc/ticket/7870
> 
> I quite like this idea, and I think this is one thing people use TH for now instead.
> (Using quasi-quotes, you can produce any compilation error you like...)
> It would be great if we didn't have to pull in the full power of TH (or QQ) for this.
> 
> 
> Cheers,
> Ozgur
> 
> 
> 
> 
> -- 
> Alberto.