The Whiteboard - 2013-03-22

« first day (931 days earlier) ← previous day next day → last day (4078 days later) »

12:01 AM

Has the possibility of being used in enterprise environment, if other devs don't know Haskell (remote possibility), then if generated code is clean they can pretend it was human written. Of course as long as you want the real source to be Haskell they can't touch it, but still.

Judicious implementation of show and an ast creating monad could probably do some interesting stuff in that vein

There's mumps in enterprise??

12:19 AM

MUMPS is mostly in health care, followed by financial. I would call both very much enterprise.

You could code generate some C# yourself.

1 hour later…

1:23 AM

@psr yeah, I was pondering how I would do that on my drive home from work.. might try something really simple because it sounds pretty crazy complex

user20683

@JimmyHoffa There's a "roll you a scheme in 48 hours" for Haskell

@WorldEngineer yeah, people tout it for learning parsing, but call me crazy, I found it jumped in way too deep without the up front "here's parser combinators, here's how you use them" bit. It's really hard to find the initial beginning explanation of parser combinator usage, I mostly ended up just guessing with hackage references how to do stuff, actually fparsec documentation was the first time parser combinators made sense to me because all the haskell docs online completely skip the basics

user20683

yeah

user20683

I'd argue that the SICP is better as far as that kind of learning goes but they are both very heavy

maybe if you had a formal education with training on grammars and such you wouldn't need that initial bits, but I was definitely lost at every intro to parser combinators stuff I found online. after working with them for a day though they click pretty quick, they're such a great API

but the roll you a scheme is about parsing which isn't particularly difficult (even writing parser combinators from scratch isn't that hard with a functional language), code generation on the other hand is a bit trickier in the fashion I'm thinking about

I'm imagining a monad that takes:
do
x <- 4
increment x

and returns:
Block [(Assignment x 4), (Exec increment x)]

which you could trivially fold to get a result of:
{
int x = 4;
return increment(x);
}

a monad that constructs that abstract syntax tree...I'm not certain how one would go about that..

user20683

1:33 AM

@JimmyHoffa I've covered a good deal of theory, I wasn't able to take the Compilers class when they offered it. It was in Java anyway...

user20683

not that I'd mind that but I don't know

it would have been a state machine class then, the only interesting part would be details about the difference between front/middle/back ends and when you don't need a middle and what the consequences of that are

user20683

between Programming Language Concepts (covered lexers and parsers to a degree along with formal BNF) and Automata, I could probably pull one out of my ass

@WorldEngineer parsing that do block into the abstract syntax tree would be easy enough, but I'm not talking about parsing it, rather referring to an implementation of >>= that builds it, so the final return in that block (from the call to increment) returns that abstract syntax tree instead of executing the code. The actual execution in haskell of that do block doesn't execute the code; but returns an AST describing the code so from the AST you could generate other code

and that, I do not know how to approach heh

user20683

hmm

1:40 AM

@WorldEngineer imagine this haskell evaluating to True for instance:

(foldl1 (+) [1,2,3]) == "Int[] nums = new Int[] { 1,2,3 }; int res = 0; for (int i = 0; i < nums; i++) { res += nums[i]; } return res;"

that could be a custom implementation of foldl1 perhaps or who knows, it's just a strange concept.

2:19 AM

Woo, I wrote a bunch more on my combinator blog post. Maybe I'll actually finish it tonight. @WorldEngineer you want a blog post on programmers about combinators? I wouldn't normally bother, but I read the current blog post there and I'm not a fan, so I'd be glad to unseat it as the current post heh

user20683

@JimmyHoffa sure, I'll put it into the queue

3:31 AM

Woot. Part 1 is done, save for a little saving-face blurb at the bottom I need to hack up for anyone who reads the article and actually knows how LINQ is implemented under the covers.

I'll write Part 2 either some time or never, basically it exists in the Maybe monad for now.

3:42 AM

agh crap crap crap, stupid blogger ate my code...

3:53 AM

@WorldEngineer let me know what ya think, Combinators Part I: The fall of society Ignore the last part

I haven't finished the final saving-face snippet yet, was just trying to get things formatted properly

10 hours later…

user41796

1:48 PM

0

Q: What's are some great monad transform tutorials on the net?

So far I found these three: http://blog.sigfpe.com/2006/05/grok-haskell-monad-transformers.html http://en.wikibooks.org/wiki/Haskell/Monad_transformers#liftM And the chapter from Real World Haskell. They're all pretty good, but thus far on the journey I've found that there's always some obscu...

user41796

^^^^^ @JimmyHoffa - you should be a little less obvious in trying to convert everyone to haskell and monads.

hello

first time here .. do I just ask a question?

user41796

@chibro2 - scroll back through the chat log and you'll see a conversation between WorldEngineer, JimmyHoffa, & psr about various Haskell resources.

user41796

@chibro2 - and yes, just asking a question is kind of how it works. Not many are active in the chat room at the moment. The US is still waking up.

oh hey oh great thanks ok

Thanks for posting my question here too :)

user41796

1:54 PM

You can also call out users with the @ sign. Which is why I pinged Jimmy regarding your question. @WorldEngineer is pretty good about picking up on Haskell conversations as well. Personally, I'm just a simpleton who sticks with OO. :-)

Ah great thanks feel like a 7 year old in a new elementary school again and you're that nice dude

user41796

yes, we've all had that moment. Chat is a much better place for polling questions. The flipside is that your question may not get as much attention.

ah it's ok no different than other times in my life. Thanks again!

user41796

You know it's going to be a long day when you spend 15 minutes trying to understand why your new WCF service isn't available and you realize that you forgot to add the operation contract decorator. <sigh>

@chibro2 Wish I could help but I haven't frankly bothered working with transformers yet myself. That was going to be my next tackle as well. I have a general understanding of them, what they're for and how they work but actually implementing or working with one iduno.

@GlenH7 heh oops. Look at the bright side, you're not getting null values on datacontract members because you forgot a datamember attribute, that's a nasty one because it gives you no inkling why other than you start getting null reference exceptions somewhere down the stack

2:11 PM

@Jimm

@JimmyHoffa It's all good thanks for replying though! If I find anything I'll post it here

I think I remember the one on the haskellwiki being fairly well explained, but then I like the writing style on the haskellwiki in general. Very terse and authoratative

user41796

@JimmyHoffa - that was a few days ago.... ;-) ironically, I had the same type of problem with advertising an enum since I wasn't using it with a data member yet. <grrrr>

user41796

Clearly I'm not the only one having challenges today

user41796

1

Q: How do you deal with pain from coding too much?

This is probably not the right place for this question, but I figure it is a board visited by thousands of programmers, and I am a programmer, and it is related, so I would try anyway. This is not a technical question, but rather a physical one - targeting mostly programmers, who have, as we all...

ergonomics lifestyle

@JimmyHoffa yup Im taking a look

2:47 PM

Great, I just implemented the maybe monad in C# to actually use it... It was fun when I did that in the past playing around, but actually doing it in a real code base feels really dirty because it is sooo non-idiomatic...

LOL someone reads your code and goes: digitalscrapbookingshop.com/wp-content/digital-scrapbooking/…

3:12 PM

@GlenH7 Where would you suggest asking that question? And can it be edited to be more on topic?

@Phil For the medical issues, you should really consider talking to a doctor (don't take medical advice from the internet, it never turns out well). For the productivity issues, you should take a look at our sister site, Personal Productivity.

@YannisRizos how do you know I don't have my M.D.? Perhaps internet medical advice from me is fully backed by the united states medical board, did ya ever think of that?

@JimmyHoffa That's irrelevant. If you have a M.D. and are giving medical advice over the internet, you're a charlatan.

My name's not charles

Oh charlatan, I was thinking charlemagne

:)

@Phil Don't edit it - it's gonna end up rejected anyway. It was judged off-topic by enough high reppers to close it, and has attracted quite a few "me too!" answers for the short space of time it's been up. But it got a very decent answer anyway, so hopefully the OP got out of it what they needed.

3:43 PM

I'm trying to build character but Eclipse is really confusing.

Wouldn't this throw a StackOverflowException?

@maple_shaft that depends.

It looks like java, so; yes.

Though I don't know for certain if any of the JVMs have tail call optimization or not

He may be good at Physics, but his Java class design needs some work

though if it were tail call optimized (which it could be), it would just be an infinite loop anyway

because there's no exit case

@JimmyHoffa I think that NOT throwing a StackOverflow in this case would be a bad thing

It would, because then you'd have an infinite loop

stackoverflow is preferable to that

3:50 PM

right... thats why I think that tail call optimization, even if it exists, doesn't apply to this scenario

now a better way to do this would be to start a new thread of execution on aim

but just saying.. JVM probably does have tail call optimization because it has scala, and it's possible for a java compiler to use those IL codes for that scenario since the parameters are known quantitys. Just saying, not all recursive functions can be tail call optimized, it depends on how the recursive call is given values, but that one could be

and a better way still would be to not use catch statements for your application logic flow

I suddenly realized he only created that comic specifically to make people have conversations like this

Touche Randal Munroe... you win this round.

@JimmyHoffa Hmm... If I had a ton of free time on my hands I would try to modify the JVM source code to see what would happen ;-)

@JimmyHoffa damnit... your right... he got us again lol

user41796

4:22 PM

@JimmyHoffa & @maple_shaft - you two are aware of "What If" XKCD, right?

user41796

what-if.xkcd.com

user41796

I should probably add a "highly addictive" warning to that URL....

@GlenH7 did you just steal my entire morning? In the case of maple_shaft you're liable to get him in some crap, he needs to become a photoshop expert within a week

user41796

erm, don't click through then. God save you otherwise.

@JimmyHoffa @GlenH7 Don't worry I discovered this months ago and since then have been hitting F5 repeatedly every Tuesday until the new one loads

The first one, Light Speed Baseball Pitch is probably the single funniest thing I have ever read on the Internet

user41796

4:29 PM

I need to add What If to my RSS feedlist. Unfortunately, I need to find a new RSS reader package since Google is axing Reader. <sigh>

@GlenH7 sucks right?! RSS has been around for over a decade

and I still have yet to find a decent quality RSS feed reader

user41796

yeah, I'm annoyed. I use a mac at home and Winders at work. Reader made it easy to jump back and forth between the platforms

Its even harder if you have a Blackberry

They're killing reader? Ugh. That is indeed the only RSS reader I found worth a damn.

user41796

There's the typical internet outrage about it, but they've dropped reader off of the menu bar when you're logged in

user41796

4:31 PM

that's pretty much a kiss of death

user41796

and I'm pretty happy with my feature phone

Does anyone anywhere know why google does these things?

I kid you not... the only passable free RSS reader on the Blackberry app marketplace keeps throwing RIMJobException

2

HAHAHA

user41796

although I've been eying some of the newer androids

4:32 PM

dont google that!

@maple_shaft make it do that, take a picture of it, and e-mail that image to the daily wtf

btw, I'm 3 back into what-if now, thanks. Jerk. I'm going to close that tab to avoid hurting myself..

user41796

too late. Image googled that term. Thanks....

If only capturing a screenshot on a blackberry was that simple

user41796

I warned you @JimmyHoffa

@maple_shaft it's much more fun to take pictures of things that have cameras, the irony is tantamount.

ahhh..closing that tab was smart, I had to larf at the method I just wrote and couldn't reasonably continue immediately, and went looking for that tab. That would have been another 20 minutes lost..

1 hour later…

user41796

5:52 PM

@YannisRizos - I think a non-friend of the community is back.

goma in his head thinks he's a friend, so I can only presume you mean leslar

@GlenH7 Is it Leslar? I missed that guy!

user41796

Me thinks, yes. <sigh>

user41796

MO is a little different but not by much. Rather sporting of him to give us some clues about actual identity

user41796

perhaps it's not as much fun if there's no chance of getting caught

5:54 PM

> You've earned the "Tumbleweed" badge for Haskell network connection graceful handler. See your profile.

@GlenH7 Nothing Leslar-y about this one. But certainly something trollish (oscillation is the latest MSO meme...)

18

A: The Many Memes of Meta

Meme: Oscillation Originator: asdf_enel_hak First Heard: February 3rd, 2013 Cultural Height: Oscillates between then and now Definition: The repetitive variation, typically in time, of some measure about a central value (often a point of equilibrium) or between two or more different states (W...

user41796

perhaps I should hang out at MSO more often. Or not. P.SE eats up enough of my time as it is. :-)

user41796

the comments in that MSO link are hilarious

user41796

poor unicorns

@GlenH7 poor people feeling sad due to the treatment of imaginary things ;P

user41796

5:58 PM

@JimmyHoffa - don't you know? N. Korea has actual, physical evidence of unicorns having existed.

user41796

Or so they say. I'm not aware of any outside verification of those claims. But N. Korea has proven very trustworthy in the past with claims like this.

@GlenH7 that's right, I forgot about the N. Korean War of '86, personally I wish the unicorns had won.

user41796

Shhhh, I think they did

ahhh you know, that makes sense, it would totally explain the water slides, everyone knows unicorns love water slides...

6:15 PM

The "Let's Trim Our Hair In Accordance With The Socialist Lifestyle Unicorn Fun Play Wet Park" is totally worth a visit.

@psr Have you run into the term Combinator yet and/or gotten a definition of it in haskell?

6:37 PM

I've looked it up before, but I don't think it's been mentioned in a haskell specific source. I've also read a couple of books about DSLs that had some detail on combinator parsers.

7:00 PM

@psr yeah people usually only hear the term related to parsers because you can create great parsers using combinators, but the concept is more generic than that and something you'll hear people talk about a lot regarding haskell. I wrote a little blog on them last night here but that's meant to be generic info for any dev, since you're learning haskell I'd suggest you go learn combinators here:

SKI combinator calculus

SKI combinator calculus is a computational system that may be perceived as a reduced version of untyped lambda calculus. It can be thought of as a computer programming language, though it is not useful for writing software. Instead, it is important in the mathematical theory of algorithms because it is an extremely simple Turing complete language. All operations in lambda calculus are expressed in SKI as binary trees whose leaves are one of the three symbols S, K, and I (called combinators). In fact, the symbol I is added only for convenience, and just the other two suffice for all of th...

SKI is just one system of combinators, there's a variety of others but the important point is to understand the behaviour as in haskell it's good practice to break things up into generic combinators, $ for instance is the B combinator from BCKW combinator system

7:21 PM

@JimmyHoffa When you turn this into a "useful" programming language, do you wind up with Lisp or ML?

user20683

@RobertHarvey when you turn it into Lambda Calculus then yes

Gotta put that in my bookmarks. I have a thousand or so of those now; one of these days I'll get around to actually reading all of them. :)

user20683

@RobertHarvey there's a good, cheap book by Dover on the subject of Lambda Calculus

user20683

amazon.com/…

user20683

it's not very rigorous but other than that, it's pretty solid

7:25 PM

Thanks.

@RobertHarvey and if you leave it as an unuseful language, you end up with..

Unlambda is a minimal, "nearly pure" functional programming language invented by David Madore. It is based on combinatory logic, a version of the lambda calculus that omits the lambda operator. It relies mainly on two built-in functions (s and k) and an "apply" operator (written `, the backquote character). These alone make it Turing-complete, but there are also some I/O functions to make it possible to interact with the user, some shortcut functions and a function for lazy evaluation. There are no variables in the language. Basic principles As an esoteric programming language, Unlambda is...

But as a useful language it's most related to LISP, after all LISP was originally developed to be the computer based variant of church's lambda calculus, which combinatory calculus is barely different from

ML would be more recognizable in the eyes of...

Lambda-mu calculus

In mathematical logic and computer science, the lambda-mu calculus is an extension of the lambda calculus, and was introduced by M. Parigot. It introduces two new operators: the mu operator (which is completely different both from the mu operator found in computability theory and from the μ operator of modal μ-calculus) and the bracket operator. Proof-theoretically, it provides a well-behaved formulation of Classical natural deduction. One of the main goals of this extended calculus is to be able to describe expressions corresponding to theorems in classical logic. According to the Curry&...

user55340

Ghads... any more talk of this, and @YannisRizos would feel right at home with all the greek. ;-)

Lambda-mu being an extension of lambda calculus which implements types, and ML is all about types

user20683

Lazy K is the most horrible version

user20683

esolangs.org/wiki/Lazy_K

7:34 PM

> Lazy K was designed to address a shortcoming of Unlambda; namely, its lack of functional purity

lol... what about it's other shortcoming, being totally useless

user20683

@JimmyHoffa Lazy K is even more useless

Yeah, but I just love that somebody felt they should fix one of the shortcomings in a turing tar pit, that's like somebody deciding "Huh, I don't like the case sensitivity in lolcode, I'm going to go fix that..."

Hello! I'm looking for some guidance with implementing user permissions that does not follow the regular "user may read/write/edit/delete" flagging. I've got multiple objects storing booleans for specific UI functions which is better off implementing a true/false "matrix" instead of bit assignments. My problem is storing this, I'm thinking of storing state blobs, but is this even a good idea? Anyone with a good question link here on programmers opr stackoverflow

user55340

@Daniel Is this matrix forever a fixed size?

@MichaelT Properties may be added, but not removed

Defaults set in objects

Defaults to* object properties

user55340

7:47 PM

I tend to think you want to store this in a more db friendly format - so you can search and update... and then... if you want that boolean matrix, write something either in the database or after you pull the data back to create it.

user55340

Storing blobs is one of those 'icky' things when I see it in a database... especially if there is actually structured data within it (rather than just "here's a file" or "here is a serialized object that I really can't do anything with."

user55340

@Daniel comments on that? Thoughts?

Yeah, I was afraid of that. Though I'd like a good way to implement these permissions that can be quickly re-instatiated becuase no real modifying will be done unless the objects are live.

I guess I'm looking for some sort of article about actually implementing these kind of things.

And how to store them efficiently

I see the problem, unless the originating objects are available the data is fixed. Which is not a good thing.

I'm researching the area and the current system is implementing a column-based matrix which the system is outgrowing

user55340

8:42 PM

-2

Q: Should I include poker on my resume?

In 2009 I graduated with a degree in compsci, 3.0 gpa from an average school. I worked for a few professors as a research assistant doing fairly simple programming in C/C++. After graduating I found a better paying job at a different school. This job is hard to describe as my duties were all over...

interview job-market resume jobs career-transition

user55340

If this question was de-resumed, might it be one that could get moved to the workplace?

user41796

@MichaelT what do you mean by de-resumed? The question is inherently about how to explain that period of time on his resume. In any case, I thought "how do I explain ..." Q's were ok for Workplace. Have you flagged the Q for review?

user55340

I was under the impression that workplace tried to avoid resume questions.

user41796

could be; I don't frequent there often enough to say. Thought I had seen some along those lines though

user41796

did you flag? Otherwise I'm happy to do so and I'll put a comment on the P.SE Q

user55340

8:54 PM

workplace.stackexchange.com/…

user55340

"Please review my resume/CV"
Questions need to apply to more than just you. Since this site is here to help everyone, and not review to a specific resume, these are not "questions" to us as they don't have definite answers.

user41796

I went ahead and flagged on that one for migration consideration. Not sure if it's on-topic or not for Workplace, but the mods can figure that out more quickly than you or I can. :-)

user55340

Yep... mods are human exception handlers. Still not sure if they want to catch that Blackberry exception though.

user20683

@MichaelT /dons biohazard suit and gets to handling with a mop and bucket

9:39 PM

Alright. My monadic state machine works in C#... Not yet sure how this will go over with my new team, either they're going to start raising crucifi at me or shrug and just ask if it works... stepping through it debugging is really damn confusing even to me..

@WorldEngineer Don't do it, just wait for RIM to go out of business then claim it unsupported

yeck

10:35 PM

@WorldEngineer I just looked at that "Old Man's War" article on wikipedia closer and found it describes the FTL concept, yeah you're right that's basically how the IO monad works. That's pretty funny actually.

Another self-deletion.

programmers.stackexchange.com/q/191636/1204

\o/

« first day (931 days earlier) ← previous day next day → last day (4078 days later) »

Transcript for

Mar '1322

The Whiteboard

General Discussion for softwareengineering.stackexchange.com P...

coffee-day horror-stories monads my-code-is-compiling scotch the-other-monitor

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile