TriFunc meetup programming ideas

[Brian Adkins]

We had a great turnout of eight people at the July TriFunc meetup. Thanks to all who took the time to join us this evening. It was great seeing old timers and meeting new folks.

One of the things we discussed was the possibility of working through a book, or set of functional programming problems, individually, but concurrently with others in the group.

This would allow people to use their own functional programming language of choice while providing some cohesiveness, structure, common points of discussion/instruction, etc. Even if someone is the only one using a particular language, they can still benefit from others working on the same problem.

Some ideas proposed were:

* Structure and Interpretation of Computer Programs (SICP) 

  https://mitpress.mit.edu/sicp/full-text/book/book.html

* Project Euler 

  https://projecteuler.net/

* ICFP Programming Contest problems 

  http://icfpcontest.org/

* TriFunc member submitted programming problems

Let's use this thread to propose other ideas, then I'll create a survey monkey (or similar) to measure the interest level of proposals. If the votes cluster, we may have concurrent "tracks".

I don't think this was proposed, but one of the member's brought Chris Okasaki's "Purely Functional Data Structures" book, so I'll add that as a possibility.

Thanks,

Brian

[Matthew Maravillas]

I couldn't make it this time, but I'm hoping to make it out in the future.

All of those problem sources sound great, especially Project Euler, since it's language agnostic but still provides answer checking. A couple other sources are the ACM ICPC finals archive (http://icpc.baylor.edu/worldfinals/problems) and the UVa judge (http://uva.onlinejudge.org). The latter only accepts Pascal, C/C++, and Java submissions for online grading, unfortunately. 

Matthew

--
You received this message because you are subscribed to the Google Groups "TriFunc" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trifunc+u...@googlegroups.com.
To post to this group, send email to tri...@googlegroups.com.
Visit this group at http://groups.google.com/group/trifunc.
For more options, visit https://groups.google.com/d/optout.

[Brian Adkins]

I've created a simple survey to vote on programming ideas:

https://www.surveymonkey.com/s/PKJL3CS

I'll let it run for a while, and then post the results here. If you're not planning on attending the meetups, but you'd still like to work through the same material, feel free to vote. I expect we'll post questions/comments/tips on the mailing list as we go.

Thanks,

Brian

[Brian Adkins]

Interim update. We've had 4 responses thus far. I don't want to bias the folks who haven't voted yet, so I'll anonymize the choices but still give an idea of the distribution:

One choice is at 7

Three choices are tied at 6

One choice is at 3

One choice is at 1

"Interested" is worth 2 points, "Neutral" is worth 1 point, "Not interested" & no answer is worth 0 points.

I know there are a few others who had expressed interest, so I'll leave the survey up for a while longer, but if you haven't voted yet, and want to do so, please do so soon.

Thanks,

Brian 

[tom_b]

I mentioned it my survey response, but I've had fun playing with the problems at www.hackerrank.com using Clojure.  They support a ton of programming languages (e.g, you submit code for a problem rather than just an answer).  Might be worth checking out, especially for either comparing solutions across different functional languages or a source of problems to grind on at meetings.

Glad to see the group is reviving.  Not sure if I'll be able to attend until the September timeframe, but I'm lurking around the forum watching.

-Tom

[Brian Adkins]

We're up to 7 responses now. Each response strictly alternated between a tie and a winner. It's now a three way tie :) Forty eight hours is probably a sufficient length of time for voting, so the results are:

ACM ICPC: 1 interested, 5 neutral => 7

ICFP Contest: 4 interested, 3 neutral => 11

Project Euler: 4 interested, 3 neutral => 11

SICP: 4 interested, 3 neutral => 11

TriFunc submitted: 2 interested, 5 neutral => 9

Other: 4 neutral => 4

Since ICFP made it to the finals, and they'll announce the problems in less than a day, we might as well wait and see what the problems are before moving forward in case that changes any opinions.

Beyond that, I'll just throw it out for discussion re: the best way to proceed. I'm personally fine with any of the top three, and if others feel that way, it may just be matter of ordering them; then we could go through one after the other.

Thoughts from the participants?

Brian 

[Brian Adkins]

ICFP Programming Contest spec is online now: http://icfpcontest.org/specification.html

[Brian Adkins]

David Joyner popped into the IRC channel today, and we discussed this a bit and came to the conclusion that starting out with Project Euler may be the best approach. Of the three ideas that were tied, it's the finest grained which will allow us to make some decent progress without investing a lot of time. After we make some progress, we can discuss how it's working out and how long to continue before switching to the ICFP problem and/or SICP. Folks are welcome to propose individual problems that we could mixin during the more "formal" group project also.

We already had a github repository for TriFunc: https://github.com/lojic/trifunc

So I just created a members directory at the top level and a placeholder for my Project Euler code. If you email me directly, I can add you as a collaborator to the repo, then you can add your own directory.

Also, if there is clearly a better way to organize this, I'm open to suggestions.

Thanks,

Brian