[Haskell & FP in Education] Welcome and introductions

[Chris Smith]

Oops, I accidentally dropped the mailing list.

On Mon, Dec 17, 2018 at 3:09 PM Chris Smith <cdsmith at gmail.com> wrote:

> On Mon, Dec 17, 2018 at 12:25 PM Fernando Alegre <falegre at lsu.edu> wrote:
>
>> I am also testing material for other courses with a couple of teachers
>> and also directly with
>> a small group of advanced students. I liked the CodeWorld environment so
>> much that I decided I would
>> try it with the advanced students, and the feedback I got has been very
>> interesting. Apart from
>> their dislike of the syntax and similar nuances, their main complaint is
>> that functional programs
>> are too dense, and that increases their cognitive load too much, so that,
>> even though they can
>> write the code, they do not enjoy doing it. I am struggling to find ways
>> to make it more enjoyable
>> for them, as it is really fun for me.
>
>
> Thanks for sharing this.  I've been thinking about similar things.  My
> observation is that composing expressions seems to be a very unfamiliar,
> and therefore uncomfortable, way of working for many students (perhaps for
> many non-students, as well).  It's natural for students to worry about what
> they need to do, but not about how to connect it to what's already there.
> Therefore, they want to just add the right thing in their code.  They can
> tolerate having to add it between some curly braces.  But thinking about
> how the new addition is related to the whole, where it fits, and how it's
> combined, seems to be a step too far for many.  Beginners in imperative
> languages have exactly the same problems when they are forced to program
> with nested blocks of control flow and other substantial structure.  But
> imperative code comes with a default aggregation on statements, while the
> functional environment mostly forces you to be explicit about how
> subexpressions are composed together, from the beginning.
>
> On the one hand, I'd like to argue that increasing students' familiarity
> with this kind of thinking, which I see as part of the essence of a
> mathematical world-view, is in fact the goal, and therefore worth doing
> regardless of student support.  But, if course, you can't really force
> anyone to learn anything against their will, so this answer is not
> sufficient.  In any case, I suspect we're seeing the same thing, and I
> think you're right that it's often a fight to get students on-board because
> the functional environment asks them for more careful thinking than they
> would like it to.  If you do find an answer to this, I'd love to see it.
>
> My best answer so far is to allow students to see the program's structure
> in a variety of ways.  The "Inspect" button in CodeWorld is an example of
> this.  When a student shoves something into a drawing at the wrong place
> and becomes frustrated, I often find it helpful to ask them to open up the
> Inspect dialog, and take a look at how different expressions in their code
> define their own pictures, and check that against what they expected.  It
> helps, but it doesn't make the problem go away.  Luite and Shae and I
> talked at ICFP this year about a more general-purpose system for examining
> the meanings of expressions in a running Haskell program, and I think it's
> doable with GHC 8.6 and a lot of work.  I'm still waiting on GHCJS to reach
> a stable 8.6 milestone before I give it a shot.  There's a brief write-up
> at https://github.com/google/codeworld/issues/741, but the idea is
> certainly not specific to CodeWorld (except for the UI portion).  I think
> having something like this at one's disposal would make it easier to give
> students a feeling for the idea that each of their subexpressions really is
> a meaningful value in its own right, and really understanding that as a
> tool for comprehending code seems to be the key for getting over the
> unfamiliarity and finding the power of this way of working.  But, that
> could be me just being optimistic.
>
> Chris
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/education/attachments/20181217/a7cec9ef/attachment.html>