[Pdx-pm] Fwd: [pdxfunc] PDXfunc meeting summary for 2007-11-15

Kevin Scaldeferri kevin at scaldeferri.com
Fri Nov 16 09:18:37 PST 2007


Perhaps some perl mongers will also be interested in this new group,  
dedicated to functional programming.  The first meeting was last  
night, the next is probably Dec 10th.  Older messages and subscription  
info at: http://groups.google.com/group/pdxfunc

-kevin

Begin forwarded message:

> From: Igal Koshevoy <igal at pragmaticraft.com>
> Date: November 16, 2007 3:24:05 AM PST
> To: pdxfunc <pdxfunc at googlegroups.com>
> Subject: [pdxfunc] PDXfunc meeting summary for 2007-11-15
> Reply-To: pdxfunc at googlegroups.com
>
>
> PDXfunc - Portland Functional Programming User Group
>
> Our first meeting was a great success. We had 12 people show up and
> had some great discussions. This email summarizes group business
> issues, educational resources for learning more about FP (functional
> programming), and the topics discussed at the meeting.
>
> GROUP BUSINESS:
>
> - NEXT MEETING: We've TENTATIVELY agreed to meet every second Monday,
> so the next meeting is TENTATIVELY scheduled for Monday, December 10
> at 7pm. Igal Koshevoy will contact Don Stewart to see if he's
> available to give a presentation and reserve room at CubeSpace --
> http://www.cubespacepdx.com/directions
>
> - PHILOSOPHY: We want this group to welcome everyone, regardless of
> their level of experience with FP or the language they use. Attendees
> come from radically different backgrounds, so it's important that
> meetings provide content that caters to different skill levels, that
> way everyone can learn something. We use many different terms,
> approaches and tools, so it's vital that everyone be open, tolerant
> and willing to rephrase themselves to help others understand and
> participate.
>
> - CALL FOR PRESENTATIONS: We want you to be involved, so please post
> your ideas to the mailing list. Are you doing something interesting,
> come across a tip, or have a tough question? Consider delivering a
> formal 1 hour talk, or giving a casual 5 minute demo, or just bring up
> an interesting question that can encourage good discussion, or do
> something in between. Although we'll probably have a main presentation
> for the next meeting, it'd be great to have some smaller talks that
> are ready to go if we have time.
>
>
> EDUCATIONAL RESOURCES:
>
> - VIDEO: Simon Peyton Jones "A Taste of Haskell" at OSCON --
> http://blip.tv/file/324976 AND http://blip.tv/file/325646
> - CLASS: Portland State University's "Introduction to functional
> programming" in the spring. Taught by award-winning Professor Mark
> Jones using Haskell -- http://web.cecs.pdx.edu/~mpj/
> - PAPERS: "Functional Pearls" papers describe elegant and instructive
> functional programming examples -- http://www.haskell.org/haskellwiki/Research_papers/Functional_pearls
> - CONFERENCE: Commercial Users of Functional Programming held a
> conference in Portland recently and their site provides some
> interesting abstracts and industry links -- http://cufp.galois.com/
>
>
> MEETING DISCUSSION TOPICS:
>
> - INTRODUCTIONS: The diversity of the group was incredible. We had a
> professor that taught and created FPLs (functional programming
> languages), PhD students doing sophisticated research with FPLs, full-
> time commercial FP developers working with Haskell at companies like
> Galois, a developer doing mobile phone synchronization with Erlang,
> some hobbyists exploring FP for fun, and some imperative/OOP
> developers whose experience was limited to using the FP features of
> Ruby, Python and Perl.
>
> - WHAT IS FP?: We had a big, non-conclusive discussion about how to
> define functional programming. Is it the presence of good things, like
> first class functions? Is it the absence of bad things, like state? Is
> it the purity? Is it the natural relationship to math? Different folks
> that used different languages had different preferences about what was
> important. The group thought it might be good to spend a few hours
> together reviewing and trying to improve the Wikipedia article on the
> topic.
>
> - FP PROPAGANDA: Professor Tim Sheard (Portland State University),
> Iavor Diatchki (Galios R&D Engineering) and others spoke
> enthusiastically about the benefits of functional programming,
> including....
>    - ELEGANCE: Provides clean solutions to tricky problems.
>    - ABSTRACTION: Makes it possible to create reusable design pattern-
> like features. For instance, monads as an abstract data type offer
> unique benefits difficult to obtain with less-capable languages.
>    - INFERRED TYPING: Catches many errors at compile time, simplifies
> development and improves code quality. This typing approach is asier
> to work with than strictly typed languages (e.g., Java) and provides
> assurance that the code is free certain bugs that are difficult to
> eliminate in dynamic languages (e.g., Ruby) even with excellent tests
> and code coverage.
>
> - HARDWARE OPTIMIZATION: FP has interesting ties to hardware design,
> and many ideas behind purpose-specific FP hardware created in the past
> are now present in the micro-architectures of modern commodity
> processors. Applying FP to hardware design can also provide intriguing
> benefits, such as quickly-producing programs compiled to expensive,
> small run FPGAs -- and later recompiling to target inexpensive high
> volume manufactured ASICs.
>
> - CHALLENGES TO FP's UPTAKE: The concepts and benefits can be hard to
> describe, especially to those without a theoretical background. FP
> discussions often use academic, specialized vocabulary to explain
> concepts that are well understood by mainstream programmers under more
> general terms. Failure to understand or express the benefits of FP has
> limited uptake of FP by mainstream businesses, and kept it mostly to
> academia and a few niche industries, such as financial modeling and
> telephony. However, interest has been growing recently.
>
> - CHALLENGES OF DESIGNING FPs: Many small languages, few programmers
> and radically different needs. Languages have evolved and even
> splintered in attempt to cater to these niches. Different FP languages
> have very different goals -- e.g., Haskell GHC's "everything but the
> kitchen sink" vs. Haskell Hugs' minimalistic footprint vs. OCaml's
> high performance requirements. The communities behind FP languages are
> also very different -- Haskell seems to welcome anybody vs. ML is lead
> by a small set of experts vs. other languages communities that are
> tied to certain businesses, nationalities, locations, etc).
>
> Thank you for reading these notes, I look forward to seeing you at the
> next meeting.
>
> PS: Tom stayed after the meeting to help me craft this summary,
> thanks!
> PPS: All errors in this posting are my fault, please post corrections.
> --~--~---------~--~----~------------~-------~--~----~
> You received this message because you are subscribed to the Google  
> Groups "pdxfunc" group.
> To post to this group, send email to pdxfunc at googlegroups.com
> To unsubscribe from this group, send email to pdxfunc-unsubscribe at googlegroups.com
> For more options, visit this group at http://groups.google.com/group/pdxfunc?hl=en
> -~----------~----~----~----~------~----~------~--~---
>



More information about the Pdx-pm-list mailing list