Plans for next meeting (11 Sep)

2 views
Skip to first unread message

Sean Leather

unread,
Aug 18, 2009, 4:36:31 AM8/18/09
to dutc...@googlegroups.com
There were some complaints about lack of warnings and planning for last week's meeting. I just saw a reminder for the Boston HUG, and I thought I'd start getting the work rolling for next month's meeting.

Tom mentioned wanting to present his project. Do you still want to do that? Note the date and time: 19:00 on Friday, 11 September. Are you going to want an overhead projector?

Does anybody have a place we can meet for a talk? Pedro, do you want to look into the UU library?

Lastly, do we have any volunteer to send out meeting reminders to this mailing list and Haskell Café? I think it would be nice to have the reminders come from the same person, but we can rotate the responsibility monthly if no one is willing to take it on long-term.

Regards,
Sean

Tom Lokhorst

unread,
Aug 18, 2009, 6:09:35 AM8/18/09
to Dutch Haskell Users' Group
It's a good idea to have more advance reminders for the DHUG meetings.
Especially for planning the somewhat more complicated things like
presentations and rooms with beamers.
I'd be willing to volunteer for sending the reminders, unless someone
else really, really wants to ;-)

How many presentations do we want per meeting?
Maybe one big one that's planned ahead, with the possibility for a few
5 minute lighting talks by random people?


- Tom Lokhorst

Chris Eidhof

unread,
Aug 18, 2009, 10:11:39 AM8/18/09
to dutc...@googlegroups.com
That sounds great!

-chris

noknok

unread,
Aug 19, 2009, 7:13:06 PM8/19/09
to Dutch Haskell Users' Group
Dear Sean, Dear Tom,

Thank you for your kind initiatives!

Friday 11 September is fine and the positive reactions only increased
my motivation to give this talk (about this propositional logic
system). I am a pedantic and slow developer and yesterday, I started
all over again, once more. Nevertheless, I am quite convinced that I
can meet the deadline.

I thought, I'd print out copies of a Haddock-style summary of the core
functionality and maybe another sheet with the core definitions and
results. It could be useful to have a blackboard available. If there
is an interest in the core algebra of the subject (where things like
the "P=NP problem" come along), then it would be nice to explain and
construct some "Karnaugh-Veitch diagrams" to visualize things on the
board. And it would be really great, if I could start Hugs or ghci,
load my modules and exercise some in- and output.

According to the interface concept of modern languages, the only
necessary notions to explain are "propositional formula" and
"propositional algebra". Since this is really one of the most basic
structures in formal sciences, it probably is just a reminder of what
you all know. My whole talk is essentially done in 5 minutes. But from
there, we could have a presentation/discussion that goes into various
directions:
(*) The algorithm I use does make such a propositional system feasible
and implicitly provides a fast "SAT solver". Default propositional
logic is of exponential complexity and this is probably why
programming languages have established say list structures and number
systems, but no propositional logic systems (yet).
(*) One of my original motivations to implement propositional logic is
a re-definition of the computability concept. Instead of having
several layers of languages with say Haskell at the top, I try to turn
the perspective around and "implement" abstract concepts directly in
propositional (i.e. digital) logic. In other words, I like to replace
the lambda calculus by the propositional calculus as a programming
paradigm. I find the new perspective very exciting and promising, but
sure that is a large project and still in progress.
(*) There are some things that I seem not to be able to implement
nicely in Haskell. For example, each single "normalizer", i.e. a
function that takes formulas and returns an equivalent normal form of
some sort, induces an entire algebra. In Standard ML I was able to
take a normalizer as an argument of a higher order "functor" and that
returned the whole according structure. But in Haskell, I cannot have
structures (here: instances of my PropAlg type class) as function
results. Right? Are these kind of situations why some of you augment
Haskell to become "generic"?
(*) ...
Well, you see, 5 minutes can easily become a whole night, if it's up
to me. But I wouldn't want to dominate the whole agenda. There also
seemed to be volunteers for "Generic Haskell for Dummies (like
me)" ...

Thanks again,
groetjes Thomas B.

José Pedro Magalhães

unread,
Aug 20, 2009, 9:09:41 AM8/20/09
to dutc...@googlegroups.com
Hello all,

On Tue, Aug 18, 2009 at 10:36, Sean Leather <lea...@cs.uu.nl> wrote:
Does anybody have a place we can meet for a talk? Pedro, do you want to look into the UU library?

In principle it would be possible to meet at the Booth Hall of the University Library at the Uithof [1]. It has room for 125 people, which means it's quite big, but the other rooms are too small. Getting an overhead projector and microphones shouldn't be a problem. There's also free car parking at the library [2], and it's easily accessible by bike and bus [3].

Just across the library is the Basket [4], where we could move to for drinks after the talks. It's open until 23h.

Now the questions...
1) How many talks will we have?
2) How long should talks last?
3) Apart from Thomas, is anyone else interested in giving a talk?
4) Do we have enough participants for September 11, or would it be better to postpone the talks to October 12?
5) Does anyone have other suggestions for a possible meeting place for the talks? Is everyone ok with this suggestion?

As for my opinion...
2) I think we should try not to exceed one hour in talks, in general. The length for each talk would depend on how many talks there would be, but I guess aiming for 15 minutes talk + 5 minutes questions for each talk would be fine. Lightning talks could be nice as well, and wouldn't need to be announced in advance (just on the day itself).
3) I'm sure an introductory talk about generics would be suited for this meeting too.
4) I generally don't favor Fridays much, but if we would settle for this date for the presentations then I would most likely attend.


Cheers,
Pedro

[1] http://www.uu.nl/EN/library/contact/university_library/zaalverhuur/Pages/default.aspx#booth
[2] http://www.uu.nl/EN/library/contact/university_library/Parkeren/Pages/default.aspx
[3] http://www.uu.nl/EN/library/contact/university_library/plattegrondenrou/Pages/default.aspx
[4] http://utrecht.thebasket.nl/

Sean Leather

unread,
Aug 20, 2009, 9:24:23 AM8/20/09
to dutc...@googlegroups.com
In principle it would be possible to meet at the Booth Hall of the University Library at the Uithof [1]. It has room for 125 people, which means it's quite big, but the other rooms are too small. Getting an overhead projector and microphones shouldn't be a problem. There's also free car parking at the library [2], and it's easily accessible by bike and bus [3].

Great! Thanks for the research.
 
1) How many talks will we have?

I think we have as many as the number of people interested assuming it's a small number.
 
2) How long should talks last?

I'm happy with the people who want to talk deciding the length of time for which they want to do it. I would limit the total time to 1 or 1.5 hours. Whether we have one person who wants the whole time or five people who want to split it, both work for me.

3) Apart from Thomas, is anyone else interested in giving a talk?

I'm not ready to commit to anything specific, but I can probably put something together.

4) Do we have enough participants for September 11, or would it be better to postpone the talks to October 12?

I don't think we need to fill up slots for this. Let's just go with whoever volunteers.
 
Sean

Tom Lokhorst

unread,
Aug 20, 2009, 9:56:50 AM8/20/09
to Dutch Haskell Users' Group
I've setup a doodle to track participants and prospective presenters.

Please submit your name if you plan to attend or talk:

http://doodle.com/a7vx3a6imcue4ghr

José Pedro Magalhães

unread,
Aug 29, 2009, 8:04:32 AM8/29/09
to dutc...@googlegroups.com
Hello all,

I've just received an email confirming the reservation of the room at the University Library for the 11th.

I'll be giving a talk about generics, so make sure you come if you want to hear more about this fascinating subject! ;-)


Cheers,
Pedro

Andres Loeh

unread,
Aug 29, 2009, 7:42:31 PM8/29/09
to dutc...@googlegroups.com
Great that we'll have a meeting with talks, even if I'll miss it
because I'll be out of country for the weekend. Let's hope it is a
success and we'll have another meeting later, then I could also give a
talk, for instance on lhs2TeX.

Cheers,
Andres

Tom Lokhorst

unread,
Aug 31, 2009, 8:25:16 AM8/31/09
to Dutch Haskell Users' Group
I want to send out a reminder for this meeting to several lists.

Sean and Pedro, what will your talks be about?

Sean Leather

unread,
Aug 31, 2009, 8:55:55 AM8/31/09
to dutc...@googlegroups.com
I want to send out a reminder for this meeting to several lists.

Great! Thanks for doing that, Tom.
 
Sean and Pedro, what will your talks be about?

I suppose, since Pedro is introducing generic programming, I could follow up with an example of a special-purpose library designed with GP principles. I will talk about xformat, a library for extensible and type-safe formatting with scanf- and printf-like functions. I will discuss why I went with functional dependencies over associated types. I would especially like to get feedback on the library to see how it should be further developed.

Related links:

http://github.com/spl/xformat
http://hackage.haskell.org/package/xformat
http://splonderzoek.blogspot.com/2009/06/rfc-extensible-typed-scanf-and-printf.html

Sean
Reply all
Reply to author
Forward
0 new messages