The Whiteboard

psr

12:12 AM

If we want to be all functional we can have a tolerance parameter then curry it. Hard to see the downside.

Yannis Rizos

12:30 AM

> Chromium's distributed build system is called "goma"

user55340

@YannisRizos Google is trolling SE?

Yannis Rizos

12:58 AM

@MichaelT Possible. There are a couple of Google employees around, should I suspend their accounts just in case?

user55340

1:33 AM

@YannisRizos Too bad you can't merge 'em with Goma anymore...

Jimmy Hoffa

1:44 AM

@psr no, I must have misunderstood, I thought he was suggesting to try the next term, and if it didn't produce a change, add more on before calling the function

psr

2:08 AM

@JimmyHoffa - To rephrase what I said before without bringing in the database (bad habit of mine to talk about the data model, especially now that so many devs think of the DB as "the scary monster across the ORM"). I would tend to have a knowledge layer in the code that has a dimension class, and each dimension is an instance, not it's own class. This makes the C# dumbish type system pick up it's toys and go home (vs an enum or something).

But if I am writing the logic anyway I want someone to potentially be able to go into a UI and add new dimensions. I lose a place (the individual dimension classes) to hang more code, but I'd rather make the knowledge layer smarter as needed - it's easier to factor the code there. But it can be overkill if the meta layer never changes anyway.

People seem to differ on which way is simpler, but I find systems with lots of classes that might do arbitrary things harder to organize than a knowledge layer. But it tends to mean you can't get anything at all done without understanding the knowledge layer.

user41796

@WorldEngineer that comment almost deserves to be pinned.

Jimmy Hoffa

2:31 AM

@psr sounds very much like what I was coming up with. Though I was going to use instances for the rare special case to exist as a derived type where the base classes patterns would be too disrupted otherwise, or a pattern can be genericized separate when it would cover sufficiently many cases.

Wasn't figuring to have many different implementations though

Yannis Rizos

3:11 AM

@GlenH7 Pro tip: Room owners can pin messages.

user41796

11:10 AM

@YannisRizos - I should have put heavier emphasis on "almost." The comment was funny but I was wary of fostering an environment that would make this like your non-home-away-from-home (aka Politics Chat). :-)

Jimmy Hoffa

2:03 PM

3

Q: How is referential transparency enforced?

In FP languages, calling a function with the same parameters over and over again returns the same result over and over again (i.e. referential transparency). But a function like this (pseudo-code): function f(a, b) { return a + b + currentDateTime.seconds; } is not going to return the sam...

Why do I always miss the good questions

user55340

2:37 PM

@JimmyHoffa Wrong time zone. You need to constantly be heading east so you have 23 hour days and can get ahead of everyone else.

Jimmy Hoffa

Or I could just wait for the blue box to show up

user55340

And consider that the answers havne't been accepted and if you can explain it better...

Jimmy Hoffa

Nah, it's a fastest gun in the west; fairly straight forward question which whoever gets to it first that knows a bit of haskell will give a good and correct answer to

both answers are good and correct and only slightly saying different things...slightly...

There's nothing more to add from the haskell perspective, and I don't understand cleans uniqueness types well enough to add that

user55340

So you need to do the 'head east' to get there faster...

user55340

Or maybe if you head in the right direction fast enough, you can get the relativistic time shifting helping you get to the new questions even faster!

Jimmy Hoffa

2:42 PM

@MichaelT you're not looking at the whole scope of the problem, and taking the naive approach. What I really need to do is just burrow a hole at a 45 degree angle into the ground, I will then have the shortest possible path to the next earth quadrant

user55340

I'm just having a bit of "WTF" moments with the 'architecture' team releasing some code that errors out when there are two periods in the address part of an email address. Its like they never even tried testing it.

Jimmy Hoffa

@MichaelT yeah, I'm so glad to work somewhere that doesn't have an "architecture" team anymore. Those are rarely if ever beneficial to keep around..

the only useful architect is the wikiarchitect; someone who's intensely knowledgeable and always researching acting as the local walking knowledgebase to answer questions and give help only where requested. Not to drive direction or touch code.

He should have 100k+ rep on SO and no books on his shelf at work; because he already read them all.

user55340

2:59 PM

I debated putting "10k rep on P.SE" as a goal on a performance review....

user55340

@JimmyHoffa have you resolved the question of understanding the "3+ argument map" in clojure yet?

user55340

(map #(%1 %2) (cycle [f identity]) coll)

user55340

(you probably have... given some of the other bits suggested...)

Jimmy Hoffa

I said yesterday, it's a composition by all appearances

(f . g)(x) = f(g(x))
. is the composition operator

it's #(%1 %2) . (cycle [f identity])

or... nah, I guess it's actually making the first list %1 and second list %2

user55340

Here's a fun one...

Jimmy Hoffa

3:10 PM

so given multiple lists it passes them into the function as the parameter of the order it was in map's arguments. It ends when the first list ends.

user55340

user=> (map #(%1 %2) (cycle [inc identity]) [1 2 3 4 5 6 7 8 9 10])
(2 2 4 4 6 6 8 8 10 10)

Jimmy Hoffa

@MichaelT have you defined map with reduce yet? or defined map just using recursion? redefining the base functions in functional is one of the fun learning tasks I enjoyed when getting up to speed on that stuff

user55340

Haven't tackled reduce much... while I realize its an important concept I haven't had much cause for using it, and thus haven't done it much.

Jimmy Hoffa

reduce is awesome. define some things in terms of it to get a feel for it and you'll realize it can be used in a lot more places than you think

user55340

So... I understand that map thing now...

user55340

3:14 PM

with the cycle.

user55340

The cycle is returning a function, and the second collection is returning the list, that the map is then creating (inc 1) or (identity 2) and evaluating that.

Jimmy Hoffa

no, the cycle returns a list

user55340

Cycle is returning a list... but the map is creating (inc 1) which is then evaluated.

Jimmy Hoffa

map takes a function with variable number of parameters, where the number of parameters must match the number of lists given to map

user55340

(inc 1) (identity 2) (inc 3) (identity 4) (inc 5) (identity 6) ...
(2 2 4 4 6 6 ...)

Jimmy Hoffa

3:16 PM

think about it, what will

(map #(%1 (%2 %3)) (cycle [inc]) (cycle [inc]) (range 1 10)

give you?
What about:
(map #(%1 (%2 (%3 %4))) (cycle [inc]) (cycle [inc]) (cycle [inc]) (range 1 10)

map #(+ %1 %2) (range 1 10) (range 1 10)

user55340

(map #(%1 (%2 %3)) (cycle [inc]) (cycle [inc]) (range 1 10)
(inc (inc 1)) (inc (inc 2)) (inc (inc 3)) ...
(3 4 5 ...)

Jimmy Hoffa

the lists are just ordered into the parameter list the same order they're handed to map

user55340

The lightbulb was that cycle was returning the name of a function that map was then using as part of its evaluation.

Jimmy Hoffa

yeah, it was a strange use, took me a minute to see how it was working too heh

anwyatt

Anybody able to answer a (hopefully) quick Python/Django question?

user55340

3:21 PM

(map #(%1 %2 %3) (cycle [+ - *]) (range 2 100 2) (range 1 10))

user55340

@anwyatt I might be able to help slightly with python concepts, I haven't touched any of its libs in a long time.

anwyatt

@MichaelT This is really a Django question much more than Python - are you familliar with Django?

user55340

(+ 2 1) (- 4 2) (* 6 3) (+ 8 4) ...

user55340

@anwyatt Not at all.

anwyatt

@MichaelT No worries, dude. I appreciate your offer though - really.

Jimmy Hoffa

3:24 PM

@anwyatt If you want to redo it in ASP.NET I'm happy to help, otherwise; cheers heh

anwyatt

@JimmyHoffa Hahaha, I'll consider it. A complete overhaul might be beyond project scope at this point though 0.o

user55340

@JimmyHoffa just thinking of how I could do that (the cycle with operators) in perl... it would take some interesting ties to get around the way that two lists are handled in the map -- map { } @foo, @bar --- the @foo and @bar are concatenated together to form one list rather than interwoven.

Jimmy Hoffa

there's a function in haskell that I think is common (people say it's useful, I've never found a use though...) called zip where..
zip [1, 2, 3] [a, b, c] =>
[(1,a), (2,b), (3,c)]

You could create a similar function to zip in any language then map the resultant list to get the behaviour map is presenting there

user55340

I could make a tied scalar that returns something as a cycle - this is a standard cookbook thing.

user55340

Here's a simple version...

user55340

3:30 PM

package ValueRing;

# this is the constructor for scalar ties
sub TIESCALAR {
    my ($class, @values) = @_;
    bless  \@values, $class;
    return \@values;
}

# this intercepts read accesses
sub FETCH {
    my $self = shift;
    push(@$self, shift(@$self));
    return $self->[-1];
}

# this intercepts write accesses
sub STORE {
    my ($self, $value) = @_;
    unshift @$self, $value;
    return $value;
}

1;

user55340

And then using it...

user55340

use ValueRing;
tie $color, 'ValueRing', qw(red blue);
print "$color $color $color $color $color $color\n";
===> red blue red blue red blue

$color = 'green';
print "$color $color $color $color $color $color\n";
===> green red blue green red blue

Jimmy Hoffa

map (\(f,x) -> f x) (zip [(+), (-), (*)] [1, 2, 3]])

That's how I would do it in haskell, where \(f,x) creates a function matching f to the first member of the tuple and x to the second member

user55340

I could then use something like that in the map itself. Not quite as elegant as the clojure way, but doable.

user55340

On of the challenges with a language is getting familiar enough with it to think in it... be it spoken or programming.

Jimmy Hoffa

3:33 PM

yeah, I still think in C# 90% of the time. One of the reasons thinking in haskell comes to me at all is because it's so totally different from C# that when thinking about problems in that way there's absolutely no footing in C# for me to revert back to

user55340

I can think in Java or C easily... but with clojure, I still go back to the "think about it in another language" and perl is the language that most eaislly maps to functional.

Jimmy Hoffa

Yeah, I got nothing to revert back to heh

user55340

Question for @YannisRizos - when you count to 10 in your head, what language do you count in? heis, duo, treis, tessares? or one, two, three, four?

Jimmy Hoffa

Actually I have haskell to revert back to now from clojure heh

Jimmy Hoffa

4:28 PM

Huzzah for my special filtered Qs list, it found me this gem:

3

Q: What is the difference between a combinator and function chaining?

So from what I've read about combinators I can't quite tell how they're different from simply chaining function calls. I know I must be missing something but I'm not figuring out what I'm missing. I mean would g(f(x)) (where f and g are functions) effectively be a combinator if I gave it a diff...

functional-programming math

user55340

@JimmyHoffa and you got the answer in fast enough!

Jimmy Hoffa

I'm not very satisfied with my answer, but it's kind of a large concept to explain

user55340

Fastest Gun In The West suffers when the person is in the west and the question was asked in the eastern morning...

Jimmy Hoffa

@MichaelT yeah, well; that question is considerably more difficult to answer than the one about referential transparency

So I'm sure plenty of people just didn't know what to say

I should actually write part 2 of that blog post at some point.

user55340

4:57 PM

Ok, got the pi thing... though doing it will kick out a stack overflow if I try to do it too far, it looks like it would work given the smaller tolerance values...

user55340

(defn pi
  ([] (float (* 4 (pi 1 0.01 0 true))))
  ([term tol accum pn]
    (let
      [t (/ 1 term)
       a ((if pn + -) accum t)]
      (if (< (* 4 t) tol)
        a
        (pi (+ term 2) tol a (not pn))
      )
    )
  )
)

user55340

As you can see, its a recursive solution instead of a map/reduce approach.

Jimmy Hoffa

([] (float (* 4 (pi 1 0.01 0 true)))) <-- what does this mean?

user55340

Take the floating value of 4 * the calculation of pi - first term is 1, tolerance is 0.01, accumlator is 0, start out positive (rather than negative) with that term.

Jimmy Hoffa

I can evaluate that part, but the placement in the function definition I'm unfamiliar with. The way I understand it, you put it where the function's body belongs ?

actually you put it before the parameters list ?

user55340

5:04 PM

A tolerance of 0.01 returns 3.14_65678... and is on term 401... so 200 stack frames there.

Jimmy Hoffa

I'm not familiar with the way you ordered your parameters to defn

user55340

When (pi) is called, it does the first invocation. If it is called with 4 arguments, it does the second invocation.

Jimmy Hoffa

OH that's how you pattern match in clojure?

user55340

Yep.

Jimmy Hoffa

everything I saw about pattern matching was saying you needed a 3rd party lib and it wasn't built into the language, which didn't make any sense to me

now I'm a little more confused what those people were talking about..

I guess it's not technically pattern matching because you're still not showing a match on value, just a match on parameter count

user55340

5:06 PM

Its different argument lengths to different bodies.

Jimmy Hoffa

Right, just function overloading

Ok. Now I can read the rest of it, that was throwing me all off heh

t is the reciprocal of the term?

user55340

Yep.

Jimmy Hoffa

ah and pn is your alternating toggle

I've seen in examples people using a "recur" keyword I think it was, around clojure, that might kill your stackoverflows (F# requires the rec keyword to hint it to tail-call, wondering if clojure requires that keyword the same)

LightTable disclaims that fn "Long cannot be cast to Named"

wonder if it still has my macro for / in it

user55340

I thought I had this tail recursive, but it just didn't get optimized that way.

Jimmy Hoffa

It is

clojure.org/special_forms#recur

user55340

5:21 PM

So, if this was scheme I wouldn't have this problem...

Jimmy Hoffa

> Note that recur is the only non-stack-consuming looping construct in Clojure.

@MichaelT it's probably a side effect of being JVM that it needs the hint to try and tail-call it

user55340

> There is no tail-call optimization and the use of self-calls for looping of unknown bounds is discouraged.

user55340

Yep.

Jimmy Hoffa

just like F# does because the CLR. Identifying a tail call optimizable structure is tricky I'm given to understand

Clojure is a lot less mature than people make it out to be is so far my inclination from playing with it

user55340

21

Q: What limitations does the JVM impose on tail-call optimization

Clojure does not perform tail call optimization on its own: when you have a tail recursive function and you want to have it optimized, you have to use the special form recur. Similarly, if you have two mutually recursive functions, you can optimize them only by using trampoline. The Scala compil...

scala clojure jvm tail-call

Jimmy Hoffa

5:25 PM

Just sucks that the actually mature functional languages out there get no love, the ocaml and racket and clisp and so on

user55340

And then one of the bits that I did about tail call in C...

user55340

11

A: What methods are there to avoid a stack overflow in a recursive algorithm?

Tail call optimization is present in many languages and compilers. In this situation, the compiler recognizes a function of the form: int foo(n) { ... return bar(n); } Here, the language is able to recognize that the result being returned is the result from another function and change a f...

Jimmy Hoffa

Well I'll shut up then.

low level stuff is the one thing I am severely lacking and should think about working on

the assembly you pulled out of there is just nonsense to me. I've read a little about it and get the idea but given a block of it I'd lose days to come up with some inaccurate analysis of what it did

user55340

I don't know intel assembly, though I can get the general structure of it.

user55340

The use of a j?? indicates a jump rather than call which does a function call (and adds on to the stack).

user55340

5:31 PM

So there's a nice, tight, tail call optimized loop in segment L4. The other bits are register things, there's a multiply in there, and a subtract, and a compare. Thats about it.

user55340

If you want to have fun learning assembly-ish, look at Core War.

user55340

Core War

Core War is a programming game created by D. G. Jones and A. K. Dewdney in which two or more battle programs (called "warriors") compete for control of a virtual computer. These battle programs are written in an abstract assembly language called Redcode. At the beginning of a game, each battle program is loaded into memory at a random location, after which each program executes one instruction in turn. The object of the game is to cause the processes of opposing programs to terminate (which happens if they execute an invalid instruction), leaving the victorious program in sole possession ...

user55340

2

Q: Helping understanding RedCode

I'm trying to learn redcode, because it looks fun to make a warrior. Introduction For those who don't know what redcode is, here's a short explenation. It's an ASM-like language, but far more easy an stripped. It is used to write little programs that need to shut down other programs in a virtua...

assembly

Jimmy Hoffa

yeah, the bits I've read lead me to see j is jump, cmp, is compare, and there's like b?? for conditional jumps, all the register stuff is just gibberish to me though

user55340

@JimmyHoffa intel assembly? or redcode?

Jimmy Hoffa

5:34 PM

just assembly. Let me see if I can find what I read in trying to get my head around some assembly stuff...

here it is

coranac.com/tonc/text/asm.htm

I read through that a couple times like 3 years ago when I started to realize arm was going to matter more over time and it could be fun to try and know something about how to fiddle with mobile crap

user55340

Back in college, I had two assembly (MIPS) oriented classes. There was the "write in assembly" class - pages.cs.wisc.edu/~cs354-1 and then there is the "write a compiler to this assembly" pages.cs.wisc.edu/~fischer/cs536.html

Jimmy Hoffa

All that stuck is what I just said, and that ARM is supposed to be a simpler instruction set

user55340

I think I mentioned before that VAX assembly includes things such as "Find root of polynomial"

user55340

ARM (and MIPS - (simulator pages.cs.wisc.edu/~larus/spim.html ) ) are rather small instruction sets. This makes it easier to understand the specifics of one instruction, but harder to get bigger things done.

jmalais

Where's a good place to look for mobile/web developers that will program pro bono until the app can be marketed and investors can be obtained or it can be released on sites like kickstarter/indiegogo. I don't have enough money to truly hire someone, so I'm looking towards students and people with a lot of time on their hands. Something like gamedev.net but for web/app development rather than games.

user55340

5:41 PM

@jmalais You might check with local contracting shops to see if they will write that contract... Online places to get programmers are typically more upfront or payment on delivery (which isn't what you want). Do realize that what you are asking for is for the programmer to take much of the risk and you take little. I wouldn't accept such a contract. Consider getting a loan or VC to pay from.

user55340

@JimmyHoffa in arm, you do a add and move as seperate operations. In intel you can do the add as part of the move... this makes it a bit harder for a human to program "idomatic" assembly... or understand generated assembly.

jmalais

@MichaelT VC? And by contracting shops do you mean someone who will contract the work out to other programmers? Not sure if there are any local shops here.

user55340

Venture capital

Venture capital (VC) is financial capital provided to early-stage, high-potential, high risk, growth startup companies. The venture capital fund makes money by owning equity in the companies it invests in, which usually have a novel technology or business model in high technology industries, such as biotechnology, IT, software, etc. The typical venture capital investment occurs after the seed funding round as growth funding round (also referred to as Series A round) in the interest of generating a return through an eventual realization event, such as an IPO or trade sale of the company. ...

jmalais

...and would you be more inclined to take on a project like that with a business plan and some documentation outlying plans, perhaps a draft of the UI completed or a supplemental website completed. I'd like your personal opinion.

user55340

Contracting shops as in a company that has several contractors working for it that will go do jobs for other companies. iTalent, intertech and the like.

user55340

5:48 PM

I've worked at a startup where I had a paycheck bounce. That was a very stressful situation for me. If there is any risk of not getting paid, I wouldn't consider it.

jmalais

Do you work freelance often? What sort of salary do you receive from a startup. Are we talking like $50K +?

user55340

Years ago, I was a contractor in Silicon Valley through two different contracting shops. I was being billed at about $100/hour, and that was when I had a decade less experience than I do now.

jmalais

@MichaelT I think the issue with the VC is that the investors would want to see a tangible product before backing. Or at least see that the team behind the project is capable of completing it. Someone in my position, an intern at a municipality who works on the websites, wouldn't handle. Would you agree/disagree?

user55340

Depends on the VC.

jmalais

What do you think the likely hood is of me getting respondents if I post Help Wanted at colleges in the area, again for working pro bono.

user55340

5:52 PM

When I was between jobs in Cali, I interviewed with a VC that took ideas, built the entire team, managed it, built the product... with the intent of having it get purchased by a bigger company (and then they take the lions share of the buyout). There was certainly not a tangible product (or even a business plan) at the start.

user55340

That depends entirely on the job market in your locale and the universities themselves.

jmalais

@MichaelT Did the "idea men" behind these projects walk away with a decent amount of money? I guess it is relative since they didn't do any work, but I wonder what the payout was for them.

user55340

@jmalais It depends on how much they risk up front. If they risk little, they walk away with very little.

user55340

If you were to walk up to me and say "here's an idea, implement it" I would want 90 - 95% of the value based on just the idea... maybe more.

user55340

I've got a dozen ideas sitting in my project folder that I own 100% of... you need to make it more attractive than those ideas.

Yannis Rizos

5:56 PM

@MichaelT Greek. ena, dyo, tria, tessera... What's "heis"?

user55340

@YannisRizos no idea. I once talked to a frenchman working in California for 20 years... he said it took him 10 years of living here before he started counting in english in his head.

Yannis Rizos

@JimmyHoffa mein deutsch ist schlecht

user55340

@jmalais you might also try looking into the ask startups Stack Exchange site and consider what the early business idea is worth and approaches to go from there. As it appears to me now, you aren't risking much... and ideas are cheap.

user55340

A recent question there...

user55340

5:59 PM

0

Q: What is a reasonable price for a developer?

I've been developing my own websites for a while, earning money by selling virtual items within games. Recetly I just got hired as a developer for the first time. I've making minigames for them, and I've already agreed on a $50/game flat rate with them. This is fine. However we're also looking ...

pricing developers

jmalais

@MichaelT All true.

Jimmy Hoffa

@jmalais realize that there are far more people out there like you who have ideas and want to monetize them, than there are people like us who can actually do the leg work, so as engineers we tend to have the leverage to choose who we ply our trade for, and as such by and large skilled developers will choose to do their work for people with demonstrable ability to provide for their employees

in other words, with what you want to do, you have a ton of competition fighting you for talent; so you really need to figure out how you can make yourself look like a better choice than the others, and think about that in terms of what engineers want, not what "idea men" want, because these are two totally different sets of people with completely different motivations

jmalais

@JimmyHoffa That's pretty good advice. I can't really extend out too many promises though. Would something like 100% ownership of product until development, where it goes to 60/40% or something similar work?

until after development*

Can you type @name? I've tried it once and it didn't seem to work...

Jimmy Hoffa

@jmalais unlikely. Again, think about what your competitors offer: Salaries, stock options

medical benefits

jmalais

Do few programmers like to develop on the side of their career?

user55340

6:08 PM

Consider what the engineer is risking... months of opportunity cost of not working on something that could be making money with a paid salary (and associated benefits).

Jimmy Hoffa

@jmalais few yes, and those are usually doing it for significant moonlighting money because it's rare that an engineer needs side money. It depends on what skill level you're looking for too

user55340

Many do have side things of write a small app or the like... but again, I've got a 100% ownership of the $0.99 app I submit to the app store... or improved name recognition and cred submitting to / maintaining releases of a well known open source project (that improves my marketability and billing for the next contract).

Jimmy Hoffa

The college kids you're talking about could just as well be studying to bolster their skillset in their off time putting together their own projects if you're talking about free

jmalais

Well...checking out that SE site you linked me to, This popped up:
http://answers.onstartups.com/questions/49037/i-want-to-collaborate-on-a-project

user55340

Or for that matter, going home, playing borderlands 2 to unwind and not go nuts from another group failing to test if foo@exchange.company.com is a valid address or not... not going nuts keeps me employable.

jmalais

6:11 PM

I'm not an expert clearly, so would this guy be looking into mobile development? Python isn't mobile right?

Jimmy Hoffa

@MichaelT truth

jmalais

@MichaelT I never got into Borderlands...must be a programmer thing since my friend plays it as well

user55340

Python is a "scripting" language that is most often seen server side... though there are some andorid interpeter things for Android... it isn't commonly seen as a mobile language.

user55340

631

Q: Is there any way to run Python on Android?

I like the Android platform. Actually, with some friends, we even participate to the ADC with the Spoxt project. But Java is not my favourite language at all. We are working on a S60 version and this platform has a nice Python API. Of course there is nothing official about Python on Android, but...

python android ase android-scripting

user55340

docs.python.org/2/howto/webservers.html

jmalais

6:18 PM

@MichaelT Thanks! Included it in my answer, though I think the question will get closed.

psr

6:29 PM

@MichaelT - Impressive, answering jmalais without using words like "chump" or "patsy".

Jimmy Hoffa

@psr cut him some slack, he has a lot of failures ahead of him to make his life miserable enough

psr

I'm not calling him either of those things - I just think he's looking for one. But, yes, I imagine his chances aren't good.

jmalais

@psr No, but this is the ideal time to create a startup. And while I don't have the knowledge to accomplish entirely what I want, I am motivated to find out how to accomplish it. Which is why I asked my questions.

But as for failures, I started a game development team off gamedev.net - got pretty far into before it did become a failure.

psr

@jmalais - It doesn't sound like you are offering a very good deal to who you plan to work with. If my impression is wrong, great. As for failure, most ventures fail and you are really working on a shoestring, so - it's a tough road. But you know that.

jmalais

@psr You're not wrong, but if you have any information that would actually help me find what I'm looking for that would be appreciated.

Jimmy Hoffa

6:40 PM

@jmalais software development endeavours fail constantly when run by experienced management teams with plenty of available resources on hand and experienced engineers doing the work, you can imagine what the failure rate is like when you take away any of those traits

psr

@jmalais - Well, that would be doing you a favor but I'm not sure it would be doing the other party a favor.

Jimmy Hoffa

...and you plan to take away all of them

I have a good friend who tries the same from time to time, my best advice to you: Learn to live on ramen.

psr

6:54 PM

@JimmyHoffa - You don't have to live on ramen if you keep your day job and have other people do all the work.

The trick is recruiting. I'd suggest becoming a cult leader and using FizzBuzz to help screen who is destined for salvation.

Jimmy Hoffa

haha

fizz buzz as entry requirements.. wow, I am so one of the chosen people...

user55340

@JimmyHoffa The priesthood will do it recursively... or map reduce.

Jimmy Hoffa

Just totally imagining saint peter now opening his conversation with "Solve FizzBuzz verbally off the top of your head in the language of your choosing"

"You don't know what I'm talking about? hmm.. a pity..."

@psr the folks I've met who are into that sort of thing are so convinced of their ability to succeed they don't bother with the day job; it just gets in the way anyway. That's not necessarily everyone who's trying that though..

I suppose there's the best possible advice right there "Don't quit your day job"

@MichaelT map and reduce are both recursive operations

user20683

@JimmyHoffa Saint Peter would be more like "Gut this halibut with this sewing kit"

Jimmy Hoffa

Whoa, your saint peter is more like saint judge dredd

user20683

7:03 PM

@JimmyHoffa Jesus is the one who forgives. Saint Peter was called "The Rock of the Church" for a reason.

Jimmy Hoffa

@WorldEngineer because of his awesome 80's hair?

user55340

@JimmyHoffa they're not stack frame increasing recursives though.

user55340

@psr As I said, ideas are a dime a dozen... and at the current value of ideas, my hard drive is worth $0.10 or so (more if you count the DLC for borderlands 2 and such...). The idea types think that ideas are valuable, without realizing how little work has been put into them, how many ideas there are, and how much more work is necessary.

Jimmy Hoffa

Quick, I need to downvote something for a perfect half of moderator tools, I think that will mean I get to see the first half of every deleted Q

user55340

7:26 PM

An observation - with the active (and rapid) downvoters on P.SE we appear to fairly quickly lock down the accounts that ask bad questions. On SO, there are people who ask a dozen or so questions that languish at 0 or -1 before the question ban gets enforced because the bad questions are diluted in the pool of other (bad) questions.

user55340

Heh - informationweek.com/global-cio/interviews/…

user55340

> We have no shortage of ideas. Here are a few of mine: I wish my kitchen cupboards washed the dishes so I could just put them away dirty and take them out clean. I'd like my smartphone to unfold into a laptop, then fold back up again and fit into my pocket. I wish our CRM software tracked our reps' locations, figured out which customers they met with, listened to their conversations, and completed the call reports automatically. Ideas aren't that hard to come by.

user55340

@JimmyHoffa go make me that cupboard, and we'll split the profits 60/40. Go!

user55340

And from C2.com - c2.com/cgi/wiki?CreativeEconomy c2.com/cgi/wiki?BrainsAsaCheapCommodity c2.com/cgi/wiki?IdeasAreWorthless

user55340

> Build it, and we'll see. Ideas, dreams and visions are a dime a dozen. Only implementation counts.

Jimmy Hoffa

7:34 PM

@MichaelT you haven't already made one of those yourself? I thought that was just one of those kit projects everyone does like when you make your own box car

psr

8:09 PM

@MichaelT - Hide those ideas until I find some engineers!!!

user55340

@JimmyHoffa whats your take on 'cond' in clojure?

user55340

(map #(cond
        (zero? (mod % 15)) 'fizzbuzz
        (zero? (mod % 5)) 'buzz
        (zero? (mod % 3)) 'fizz
        :else %
        )
  (range 1 100))

user55340

(Years ago, when I learned lisp, I started out everything with a parenthesis... no, wait... an emacs guy I was in my class and looking over my code said of it "you write beautiful C in lisp")

Jimmy Hoffa

@MichaelT eh, I found condp useful when I was poking around, cond looks like guard statements where condp is a case which is more up my alley since case is more haskelly

user55340

@psr I'm saved!

Jimmy Hoffa

8:18 PM

in haskell you're more likely to apply a function then case on the possible outputs, guard statements tend to be less preferred

but it's all the exact same

doesn't make any bit of different

@MichaelT find me the composition operator in clojure.. I was failing in my search for it earlier

user41796

@psr - props for calling out how others would most likely perceive that project request. @MichaelT - props for recognizing the difference in thought processes

user55340

16

Q: Function composition in Clojure?

Can Clojure implement (g ∘ f) constructions like Haskell's g . f? I'm currently using workarounds like (fn [n] (not (zero? n))), which isn't nearly as nice :)

clojure

user55340

8:41 PM

uncle-bob -- good idea? bad idea?

user41796

@MichaelT - bad idea. Just not enough material there to justify a tag. None of the questions would be harmed by removing that tag. I'd vote to remove the tag from those questions and let it expire automatically

user41796

wouldn't hurt to post the question in meta though (if you haven't already). More people stop by meta than they do chat

user55340

I'm still letting my thoughts settle on it. I'm on the fence, though realize that its because I haven't paid too much attention to tagging other than the active question I'm fixing up (and then only if its way off). The tag showing up was more of a "oh, 5 questions at the top of the activity with the same person acting on it... whats going on... hmm..."

Jimmy Hoffa

9:17 PM

@MichaelT I'm not certain, but I sincerely like the jimmyhoffa tag

psr

9:28 PM

We may have to amend the FAQ to make questions about the end-zone of Giant's stadium specifically off topic.

Yannis Rizos

10:10 PM

3

Q: Why is there no 'finally' construct in C++?

Exception handling in C++ is limited to try/throw/catch. Unlike Object Pascal, Java, C# and Python, even in C++ 11, the finally construct has not been implemented. I have seen an awful lot of C++ literature discussing "exception safe code". Yet it seems to me that many of the problems encountere...

c++ exceptions

8 comment flags - 17 deleted comments. WTF? Locking...

user55340

@YannisRizos I saw some of those cleanups... its... yea.

user55340

Can that lock extend to the answers too?

Yannis Rizos

@MichaelT I think I'll just purge all the comments, then post one comment saying "Even PHP has finally, muahahahahaha" and lock it forever.

user55340

"even php has XYZ" is not a sufficient argument for its inclusion in any other language... because php has everything short of an instruction for launching emacs.

Yannis Rizos

@MichaelT Yes. Locked the two top answers for an hour.

@MichaelT exec() can launch emacs

user55340

10:19 PM

But there isn't one call for launchEmacs(). I expect to see it in a year or two given the rate of additional call inclusion in php.

user55340

And see, then emacs can have m-x php invoke, and you can get it going until you've Eventually Malloced All Computer Storage.

user55340

Exabyte Memory And Constantly Swapping --- (back when I originally saw this expansion, it was 'eight megs', though that's nothing anymore)

Jimmy Hoffa

10:38 PM

@YannisRizos finally...

I get the distinct idea that those C++ developers don't really understand the point of finally, it's really not about resource release; that's just a side effect... Whatever, C++ developers brains I imagine to be something akin to an encyclopedia of the weird, that is to say I try not to imagine their mindset much.

psr

@JimmyHoffa Mindset of Haskell developers

Feb 21 at 0:58, by psr

I can imagine this not ending well:
Police: So tell us again why you killed your wife.
Programmer: You don't understand. I had to. She was impure. She had side effects. I had to put her in the monad.
Police: And is that garbage bag the "monad"?
Programmer: Well, sure, of course, obviously. Once it's *declared* a garbage bag it's always exactly a garbage bag for all time, always. It's *defined* to be. The monad is just a construction that embeds the underlying type system while adding it's own features.

Jimmy Hoffa

@psr yeah, I've seen that. The funny thing is every haskeller thinks that's comical. A similarly dark joke involving the mechanics of C# would involve another "What terrible immoral immature antisocial sexist people developers are" message abounding the web

psr

Every Haskeller? I'm Haskell famous?

Jimmy Hoffa

10:54 PM

Eh, I must be recalling wrong. Thought I'd seen that linked from #haskell

Probably just saw it here

psr

They would have to link to this chat or else someone re-posted it, because I wrote it.

Jimmy Hoffa

Just my memory playing tricks on me likely. I need to lay off the haskell (or maybe I need more?? hmm...)

user20683

11:29 PM

Grades are in and I am now officially done.

2

user20683

Operating Systems: B
Automata: B+
Computer Graphics Imaging: A
Semester GPA: 3.43
Computer Science Program GPA: 3.11
Religious Studies Program GPA: 3.08

Transcript for