Call for Awesome

28 views
Skip to first unread message

Matt Youell

unread,
Aug 21, 2012, 8:21:27 PM8/21/12
to pdx...@googlegroups.com
Hey y'all. 

Been a while. Probably time for another meeting. Interested?

I'm aiming for late September / early October. Let's try a split format this time: one longer talk (1 hr) and a cluster of lightning talks (10-15 mins). 

This is your Call for Awesome! If you're interested in presenting please post to this thread and tell us what fascinating/interesting/quirky/etc. programming language you'd like to share. Format and presentation style are flexible. Interpretive dance is always welcome. Conventional slide decks and code are also acceptable, I guess.

(If you offered last time but weren't chosen, send me a note and we'll put you at the front of the line for a lightning talk. If you offered before and want the long format, pitch it to the group.)

Oh, and feel free to recruit someone interested in talking, as Bart suggested earlier.

Thanks!

Matt

Jesse Cooke

unread,
Aug 22, 2012, 8:06:20 PM8/22/12
to pdx...@googlegroups.com
Sounds like a great format. Psyched for another meeting!

--------------------------------------------
Jesse Cooke :: N-tier Engineer
jc00ke.com / @jc00ke

Bart Massey

unread,
Aug 23, 2012, 2:39:08 AM8/23/12
to pdx...@googlegroups.com
My schedule is starting to fill up pretty hard for this time frame, so
let's get it scheduled ASAP. Hope I can make it. If I can, I have lots
of choices for what I might want to talk about:

* Language design and programming in Nickle (http://nickle.org)
* M4 is useful for DSL prototyping
* FGHC execution and implicit parallelism
* The Z notation and its uses [note: not technically a programming language]

Seems like there were other topics I'd considered as well, but this is
at least a start.

--Bart

On Tue, Aug 21, 2012 at 5:21 PM, Matt Youell <ma...@newmoniclabs.com> wrote:

Igal Koshevoy

unread,
Aug 27, 2012, 9:35:57 PM8/27/12
to pdx...@googlegroups.com
On Wed, Aug 22, 2012 at 11:39 PM, Bart Massey <ba...@cs.pdx.edu> wrote:
My schedule is starting to fill up pretty hard for this time frame, so
let's get it scheduled ASAP. Hope I can make it. If I can, I have lots
of choices for what I might want to talk about:
* Language design and programming in Nickle (http://nickle.org)
I'd like to hear a full meeting's worth of Nickle content because there's so much going on in there that I don't think doing a short segment on it would be enough. Maybe not at this next meeting, but at a future pdxlang (or pdxfunc) would be lovely.
 
* M4 is useful for DSL prototyping
I remember those dark years well, wandering the nightmare corpse-city built upon measureless KLOC of m4. Within its non-Euclidean halls staggered the lost souls, their emaciated bodies lurching blindly towards anything that their rotting minds mistook for a "divert", gibbering madly to themselves, "I just need to add one more layer of metaprogramming and then it'll finally all be over!" I have screamed until nothing but "dnl" came forth from my lungs. I have stared deep into the tentacled void of nested "define" and made sordid deals with the grinning teeth within, feeding hapless interns into its star-shaped maw to appease it, albeit for just a brief moment. Oh the hell-wind, titan-blur, black wings ... K&R save me from the four-lobed macro preprocessor of the ancients ... gaaargh!! 

Um. Er. Yes, I would be interested in hearing about the best practices of applying m4 to practical DSL development in an enterprise environment. *twitch* *twitch*

-Iä! Iä!

Matt Youell

unread,
Aug 27, 2012, 10:42:55 PM8/27/12
to pdx...@googlegroups.com
On Mon, Aug 27, 2012 at 6:35 PM, Igal Koshevoy <ig...@pragmaticraft.com> wrote:

* M4 is useful for DSL prototyping
I remember those dark years well, wandering the nightmare corpse-city built upon measureless KLOC of m4. Within its non-Euclidean halls staggered the lost souls, their emaciated bodies lurching blindly towards anything that their rotting minds mistook for a "divert", gibbering madly to themselves, "I just need to add one more layer of metaprogramming and then it'll finally all be over!" I have screamed until nothing but "dnl" came forth from my lungs. I have stared deep into the tentacled void of nested "define" and made sordid deals with the grinning teeth within, feeding hapless interns into its star-shaped maw to appease it, albeit for just a brief moment. Oh the hell-wind, titan-blur, black wings ... K&R save me from the four-lobed macro preprocessor of the ancients ... gaaargh!! 


Best M4 description.

Lyle Kopnicky

unread,
Aug 28, 2012, 12:47:32 AM8/28/12
to pdx...@googlegroups.com
Sounds great to me. I'm looking forward to another meeting!

Bart Massey

unread,
Aug 28, 2012, 2:10:54 PM8/28/12
to pdx...@googlegroups.com
On Mon, Aug 27, 2012 at 6:35 PM, Igal Koshevoy <ig...@pragmaticraft.com> wrote:
> On Wed, Aug 22, 2012 at 11:39 PM, Bart Massey <ba...@cs.pdx.edu> wrote:
>> * M4 is useful for DSL prototyping
>
> I remember those dark years well, wandering the nightmare corpse-city built
> upon measureless KLOC of m4. Within its non-Euclidean halls staggered the
> lost souls, their emaciated bodies lurching blindly towards anything that
> their rotting minds mistook for a "divert", gibbering madly to themselves,
> "I just need to add one more layer of metaprogramming and then it'll finally
> all be over!" I have screamed until nothing but "dnl" came forth from my
> lungs. I have stared deep into the tentacled void of nested "define" and
> made sordid deals with the grinning teeth within, feeding hapless interns
> into its star-shaped maw to appease it, albeit for just a brief moment. Oh
> the hell-wind, titan-blur, black wings ... K&R save me from the four-lobed
> macro preprocessor of the ancients ... gaaargh!!

Very nice.

> Um. Er. Yes, I would be interested in hearing about the best practices of
> applying m4 to practical DSL development in an enterprise environment.
> *twitch* *twitch*
>
> -Iä! Iä!

I don't quite know what an "enterprise environment" is
(sounds...enterprisey); I've used an M4 DSL for my MS thesis (assembly
generation from compiler intermediate code), for my PhD dissertation
(description of AI planning problems), and for the XCB project
(description of C API). All three were successful, IMHO. And none of
them looked like autotools or sendmail config, even a little.

What I'm trying to say, I guess, is that I have strode the night sky
wielding the mighty Hammer M4, smiting those problems which dare to
interfere; speaking the language of the Gods, and daring the Gods to
answer. The Gordian Knot of compiler output processing fell before me;
the Oracle of the Unseen Intelligence fell silent at my wisdom;
Promethean, I set free the fire of the X protocol. Let those who doubt
the strength of my weapon face it, and quail!

-Cthulhu Fhtagn!

--Bart

Matt Youell

unread,
Aug 29, 2012, 12:43:53 PM8/29/12
to pdx...@googlegroups.com
So, Bart's interested in presenting. Nickel or M4, thoughts?

Jesse Cooke

unread,
Aug 29, 2012, 12:46:40 PM8/29/12
to pdx...@googlegroups.com
As long as Nickel isn't followed by Back, I'm fine with it. Both sound interesting. Coin flip?


--------------------------------------------
Jesse Cooke :: N-tier Engineer
jc00ke.com / @jc00ke


Igal Koshevoy

unread,
Aug 30, 2012, 10:15:18 AM8/30/12
to pdx...@googlegroups.com
Those all sound interesting, and are reasonable uses of m4. I'd be interested in hearing more.

My particular trauma-inspired flashbacks were related to managing lots of computers at large budget-strapped organizations running spiteful and/or ancient releases of OSes like Asupex, A/UX, Domain/OS, DYNIX/ptx, AIX, HP-UX, etc. These decrepit systems cost so much, and had so much legacy software on them, that retiring them was not an option. Doing devops-style system automation or writing custom software was a pain because buying an expensive C compiler wasn't in the budget (these could cost tens of thousands of dollars per license), or if a C compiler was available then it was either so non-standard that we couldn't use it nor could compile something like Perl or gawk with it. However, m4 was often bundled or available as a binary, or GNU m4 would often compile with only a little elbow grease. 

m4 could do miracles, albeit sometimes at the price of one's sanity. Writing complex code in m4 inspired a kind of Stockholm Syndrome where the loathing melted away as working with it became an engrossing intellectual challenge of figuring out how to get stuff done with such a deceptively simple language and such an odd paradigm, since it's a Turing-complete general purpose programming language.

-igal

Igal Koshevoy

unread,
Aug 30, 2012, 10:17:35 AM8/30/12
to pdx...@googlegroups.com
On Wed, Aug 29, 2012 at 9:43 AM, Matt Youell <ma...@newmoniclabs.com> wrote:
So, Bart's interested in presenting. Nickel or M4, thoughts?

I'd be good with either. 

However, if it's Nickle, I think it'd be good to dedicate pretty much all of a meeting to it since there's a lot to it in terms of features and philosophy that I'd like to hear.

-igal
Reply all
Reply to author
Forward
0 new messages