The Whiteboard

Jimmy Hoffa

12:41 AM

data.stackexchange.com/programmers/query/102122/…

deletions start to normalize that spike down

finally

but it's still there

but it's gone down, when deletions catch up past it I think we'll see it went down. Was totally a semester start thing

Jimmy Hoffa

1:04 AM

I really don't think style is a relevant metric for a candidate to be tested on, by definition it's subjective, and a good developer while having his own sense of style, will be able to adapt to write code uniformly in the style already there such that his code is indiscernible from the rest of the code. — Jimmy Hoffa 34 secs ago

I found this interesting, my new job was the first time I was ever asked style questions, and I was a little confused by it because I've never thought it had any relevance at all

Do any of you ever ask questions like that or know what people would be looking for in them?

Bob

2:12 AM

Is it that the question is just meant to make you talk about your philosophy about style and how you would or wouldn't be able to work with standards?

user55340

2:05 PM

@JimmyHoffa I wouldn't ask "two spaces or tabs for indent" on an interview unless it was clear I was trying to be funny and reduce the tension. On the other hand, at a whiteboard I would hope to see reasonable code conventions - short methods, meaningful names -- these are things that indicate to me that someone knows how to write maintainable code. I don't care how quickly they can write crap.

Jimmy Hoffa

@MichaelT yeah, I can agree there, but I wouldn't call those things style... for instance, interviewing here I was asked if I put opening curly braces on the same line as the clause or on their own line

I wanted to just say "Uhh, what? Why? What?" but whatever, it was clear the engineers I was talking with knew their stuff so I wasn't particularly concerned regarding it. People just ask weird things in interviews sometimes I guess.

Even good engineers

user55340

I follow the code style of the code I am in no matter what the style is. I don't care if they use Whitesmiths style - I'll code that way if the code is written in it.

Jimmy Hoffa

Of course, as I said above, any good engineer makes their code indiscernible from the code around it, which is why I find style questions in an interview odd.

user55340

In one department here, its Allman style. In another, its K&R... tabs appear to be fairly consistent because the spaces advocates can't agree on 2, 3 or 4 spaces so the tab was the least objectionable to them (you can set the tabs to whatever you like in your local eclipse for 2, 3, or 4...)

user55340

Btw, look at this - nickgravgaard.com/elastictabstops

user55340

2:16 PM

Elastic tab stops. They move based on what is needed. It confuses the tab advocates... but makes them happy its tabs.

Jimmy Hoffa

lol

That's neat

So long as "tab" always means "collection of spaces" I'm happy. :)

actual \t should never be used because of cross-platform issues, as soon as your code goes in source control it may be encoded to that source controls form, and then can come out different, or if you pull it from source control on a different platform or just use a different editor on the file at one point then another, too many ways for tabs to be printed all kinds of strange

...but if the style of the code I was in used tabs, I'd use tabs...

then there's privateMember _PrivateMember m_privateMember m_PrivateMember m_PublicMember PublicMember publicMember localValue localvalue _localParam _LocalParam localParam and even more odd indicative markings I don't even know about

user55340

3:17 PM

Oh, I passed a reopen audit review... I went to click on the question itself and look at it, its edit histories and the answers it got... and it was open. Reopen audits suck.

user55340

programmers.stackexchange.com/review/reopen/28690

user55340

I was initially tempted to say "off topic" on it and let it remain closed.

Jimmy Hoffa

yeah. if a question raises any questions at all I immediately open it, and when I see any discrepancy between what I'm being shown in the review queue vs the actual Q I realize it's am audit

user41796

4:38 PM

@JimmyHoffa - I'll ask about style depending upon what I saw in their code sample. Where I'm at, we generally follow an Allman style (brace on next line). If I see K&R, Hungarian, or worse a mix of both, then I'll ask about how the candidate feels about following a team style that doesn't match their preference.

Jimmy Hoffa

Yeah, that makes sense Glen, I would tend to find it telling if somebody's own code isn't following a consistent style that they're likely pretty green, and if they claim years of experience then plausibly apathetic

user41796

For a C# shop, tabs should be the norm anymore. That allows the individual to pick their own spacing preference. Regrettably, XAML defaults to using spaces instead of tabs which causes all sorts of nuisances.

user55340

@GlenH7 Do note htat in the Java world, K&R is the style implied in the Java Style Guide. oracle.com/technetwork/java/javase/documentation/…

user41796

@JimmyHoffa - I had a candidate like that once, actually. Stylistically, all over the map. Claimed all of the sample code was his own. Couldn't code his way out of a wet paper bag. He's the guy who sparked us to always present fizz-buzz during whiteboard exercises.

Yannis Rizos

@gnat There's little point in asking answerers to expand their year+ old link only answers on closed questions, especially when those link only answers are valid answers to the question. Vote to delete the question instead (if it's answerable with just a link, and it's been closed for a while, then no reason to keep it around). Generally people will not bother fixing their old answers if the question is unsalvageable, and a couple have complained about your comments...

user41796

4:42 PM

@MichaelT - oh, I know. :-( And I totally understand the preference of technical book publishers to use a compact style. Personally, I think it's ugly and makes it too easy to lose track of closing braces.

Yannis Rizos

@gnat Rule of thumb: Moderate the question first, then (and only if there's a need) the answers.

Jimmy Hoffa

@GlenH7 yeah, it's disturbing how often people will gladly just lie like rugs in interviews...

The sad thing is, people wouldn't do it so much if it didn't get them work, so there are apparently a lot of people who get hired based on it..

user41796

@YannisRizos, @gnat - I've noticed the "you can do better than this" encouraging comments on older answers. Part of me thinks it's good because it indicates to future visitors why that's a bad answer. OTOH, I can see the value in letting a sleeping dog lay. I appreciate gnat's cleanup work though.

user41796

@JimmyHoffa - we had a sketchy candidate or two that we failed based upon our collective gut feel. We had an implicit understanding that we kind of dodged a bullet or two that way. That's what led to our putting more concrete measures in place as part of the interview process, including pre-screening. We've weeded out 3 or 4 who were iffy anyway and then got cold feet at our pre-screen request.

Jimmy Hoffa

Yeah, in reality any decent engineer will have absolutely no hesitation at the presentation of any standard hurdles.

user41796

4:48 PM

It's interesting to note that we don't use those measures as binary decision points. One of our better hires was a candidate who presented weakly at the pre-screen but simply amazed us during the actual interview. There were other indicators though that led us to green-light the interview. Without those indicators, we would have dropped the process after the pre-screen design results.

Jimmy Hoffa

and you can relax them when looking at fresh grads

Yannis Rizos

@GlenH7 Sure, no argument there. And I'm not saying comments on old answers are not valuable, for someone reading the answer today, it's age is irrelevant. But, if the question doesn't have any redeeming value, let's just kill it. You can't realistically expect people to come back and fix a year+ old answer on a question that 1) it's crap, 2) it's going away anyway.

user41796

I'd say we use proportional levels of expectation depending upon the role. An acceptable answer for a design Q will depend upon what level they're interviewing for. We had a few who could develop, but their skills didn't match their expected compensation.

user41796

@YannisRizos - agreed

user41796

@YannisRizos - can't wait for my rep to hit 20k and then I can vote to delete crap answers. :-)

user55340

4:54 PM

With those past couple days of ~200 rep each day, I'm much closer to the 10k tools than I thought I would be when I hit previous milestones.

user41796

@MichaelT - still milking that one hot question or whatever from a few days back?

user55340

I only got 50 from it yesterday, and 60 the day before... but I also got some with the "Recursion BrainTeaser with C++ and Java?" and "Handling compiler bugs in Continuous Integration" and "Custom Alphabetic Sorting of Array in Java"

user41796

Those sound like some curious questions.... Work has been fairly busy as of late, so it's cut into my time for answering Q's.

Jimmy Hoffa

...I think FP.SE is going to be pretty dead...

user41796

@JimmyHoffa - take some time for clojure then....

user55340

5:01 PM

The recursion brain teaser one is really a "I'm banned on SO, so asked it here" and I went into the difference in the compiled code, sequence points, and misconceptions of precedence.

Jimmy Hoffa

I did Euler #1 in it the other day

user55340

@JimmyHoffa which approach? The O(n) or the O(1) method?

user41796

@MichaelT - I have answered a few in hopes they would be migrated to SO and I'd get the rep gain there from the migration. :-)

Jimmy Hoffa

O(n) I'm sure. I didn't memorize your mathy technique

user55340

@GlenH7 I know this one won't get there at the start (I looked, saw the several closed, negative) and so answered it more in the concepts than the specifics of how to fix it.

user41796

5:04 PM

just GIMME TEH CODEZ!

user55340

We don't do that here... and I think he's better off for it now.

user55340

The alphabetic sorting one is kind of neat... let me pull out another lesser known Java Map - the EnumMap.

user55340

You declare an enum - so you've got type safety there... and order.

user55340

public enum Headers {
    SSN,
    NAME;
}

user55340

And then you can use the EnumMap...

user55340

5:07 PM

SortedMap<String, String> other = new TreeMap<String, String>();
EnumMap<Headers, String> headers = new EnumMap<Headers, String>(Headers.class);
try {
    headers.put(Headers.valueOf(key), value);
} catch (IllegalArgumentException  e) {
    other.put(key, value);
}

user55340

Where the keys are restricted to the enum. The actual backing implementation is a fixed sized array indexed by the enum.

user41796

Just read the recursion question - that's actually a pretty reasonable question. Makes me kind of wonder why we closed it as off-topic.

user55340

There's also an EnumSet which is really a bitfield.

user55340

It was attempted to be migrated to SO.

user41796

I think the title is misleading

user41796

5:10 PM

title implies "I have homework" or "I'm doing a coding challenge"

Yannis Rizos

> Everyone knows that you don't "STOP! Look and Listen!", but that you "STOP! Collaborate and Listen!" because Ice is back with his brand new invention. - Thomas Owens

user55340

I'm fairly sure that could be edited to be a good P.SE question that would remain open...

user41796

but the body of the question is really wrapped around how the decrement operations are handled. Sidestepping around that long-running, well-known flame war within the realm of C++, it seems appropriate to here

user55340

And I do think we gave the better answer than SO could have given...

user55340

Thank You so Much for your detailed answer! I now understand it much better! — superspacemarines yesterday

user41796

5:12 PM

@YannisRizos - good question by Thomas. Horrific reference to Vanilla Ice though.

Yannis Rizos

@GlenH7 Better than my occasional references to Bieber though.

user55340

@YannisRizos Gah... did you read his guest book entries at historical places?

user41796

@YannisRizos - Bieber's put out a ton more decent material than Vanilla Ice ever did. But that speaks more to how much I dislike Vanilla Ice than the quality of Bieber's pop stuff.

Yannis Rizos

@MichaelT I read the Anne Frank one, and was thoroughly disgusted.

user41796

@MichaelT - I don't think he's got it through his head that he's not allowed the typical indiscretions that youth is normally afforded. Too many eyes watching him.

user55340

5:16 PM

I didn't actively dislike him before... I do now.

user55340

thefw.com/bieber-historical-guest-book-messages

Jimmy Hoffa

@MichaelT you bastard. All kinds of points for a question that was asking a technical detail about how the -- operator works in C++ :P

user55340

(though I will admit I'm not 100% that's accurate or humor... but the Anne Frank one is enough)

user55340

Does the --' operator work any differently on primitave ints in C vs C++? (I've avoided C++ for a very long time - as a web guy, it doesn't have much of a place in the code I've written for the past decade and a half)

user55340

The real misconception he had wasn't the '--' operator though, it was the lack of a sequence point in the expression (which is the same in C++ and C) and what associativity and precedence means for an operator.

Jimmy Hoffa

5:23 PM

@MichaelT right, but that is a technical detail of how -- works in C/C++, sequence points are a technical detail to those languages

Either way, I'd bitch more but your answer is comprehensive and good

Yannis Rizos

26

Q: How to Avoid Users on Stack Overflow

(only half serious about this, but it was a fun exercise) I've got an issue on Stack Overflow. I'd really like to answer some C# questions to hone my skills, but certain users (cough) are way, way better than me, and always beat me to it. Is there any way I can know when these users are most a...

I'm generally asleep between midnight and 6am UK time, if that's any help. — Jon Skeet 4 hours ago

Jimmy Hoffa

lol

user55340

There is no sequence point in the expression on "foo = f(x) + g(x)" - so if f() or g() modify x, you can't guarantee the order they are invoked - its left up to the compiler. That's really what it boils down to.

Jimmy Hoffa

@JonSkeet All I heard was "I have and post the correct answer to all C# questions the world has. And I do it on 6 hours of sleep" — Jimmy Hoffa 13 secs ago

Jerk. I got a call at 6am this morning because I'm on support this week, it took me 20 minutes to figure out how to connect to the wifi, and I went to bed before midnight

Yannis Rizos

I think that's a great question and shows how stack exchange fails to do what users want. — micahhoover 5 hours ago

I'm so glad we're failing that test.

Jimmy Hoffa

5:34 PM

@micahhoover yes, SE does fail to do a lot of things users want. The truth is that's by design though; SE only wants to do one very specific thing, and has been tailor made to do that one particular thing very very well. Realize SE wants to be like an encyclopedia: You open it up and see a bunch of topics with comprehensive accurate and correct details, no topic in an encyclopedia has a listing of possibilities. The creators of SE would love to help their users with questions like this, but they didn't make SE to do that, rather they hope to help these users in other ways. — Jimmy Hoffa 40 secs ago

A while back I realized SE is trying to be an encyclopedia, everything about the place suddenly made more sense. It came with the realization that SE doesn't actually exist for the participants, the gamification was created to bring those, but they really only wanted participants so they had people to create their encyclopedia. Jeff and Joel just came up with a different take on how to make wikipedia, that's what SE is

Wikipedia doesn't exist for the contributors, they made Wikipedia for all the passive readers, and so it is for SE. As an active participant it's easy to forget SE isn't for me.

Yannis Rizos

The goal for any SE site is to eventually become a canonical resource on the subject matter. Not really an encyclopedia, but close enough.

Jimmy Hoffa

I think the encyclopedia analogy is apt though, because you can be a canonical resource on subjective gossip, but an encyclopedia doesn't deal in that form of information

Yannis Rizos

Yes.

user55340

@JimmyHoffa a P.SE meta to help you explain it...

user55340

1

Q: Can there be an area for interesting discussion?

A lot of times, development requires discussion. As programmers we know that there is not clear right or wrong answer, but a myriad of solutions, each with their own benefits and draw back. Example: Which is Better To Use and Why: Mysql, PostgreSQL or MS SQL? What are the advantages of Redis ove...

discussion feature-request recommendations

user55340

5:47 PM

Doesn't have too many answers.. mine is the only upvoted one. Lets see what else I can find...

user55340

Ahh, here we go go...

user55340

0

Q: Is there a market for "which language to learn, which book to read" questions?

With all the discussions about how to handle "which language should I learn" or "which book now" questions, and the consensus being to close these questions, I wonder if there's room for a new SE website specifically for those questions. I mean, there are so many questions like that, I think the...

discussion closed-questions off-topic

user41796

apropos of nothing, where's our "resident teen" been lately? I haven't seen Dynamic in chat for quite a bit.

Jimmy Hoffa

4

Q: Is it possible to write a code without class methods, globals, and class variables?

I'm designing my own programming language for fun, and i'm thinking of making it fully Object-oriented (no statics, no globals, no class variables, no class methods), so i need to find a way to completely eliminate the needs of class methods, but, first, i need to make sure that i know all the pr...

java python ruby language-design

Do any of you understand him? It looks like a great question, but his definition of OO makes no sense to me such that I can tell...

classes without variables or methods? what?

user55340

@GlenH7 Last seen 19 days ago...

user55340

5:50 PM

chat.stackexchange.com/users/21969/dynamic

user55340

Still active in P.SE though (2 days ago).

user55340

> I'm actually using C, but i wrote the example in Ruby so it will be easier to read.

user41796

@MichaelT - noticed he hadn't really had any activity since beginning of April

user41796

@JimmyHoffa - I think that Q is trying to "purify" OO even further, but ... I'm not enough of a language guy to really understand the main concern.

user55340

@GlenH7 Given the teenish, I think it might be a strong emphisis on school at the end of the semester taking up time... or sports...

user55340

5:54 PM

@JimmyHoffa I think he's trying to decide if it needs static methods and variables.

Jimmy Hoffa

@MichaelT but he doesn't want global variables, which is what statics are...

user55340

Nope... and I don't think he wants statics either.

user55340

But a private static variable can be quite useful for creating a flyweight cache inside of a factory... and factories are often static themselves (unless you start making singletons)

user41796

we need a "not a real problem" close question. ;-)

user41796

just trying to be snarky; I don't really mean to say that Q should be closed or not.

Jimmy Hoffa

5:59 PM

I'm just very confused what he wants.. sounds like he's asking for a computer without a cpu

Yannis Rizos

@JimmyHoffa No, can't say I understand the question. It's probably because he's using Ruby in his examples, and my mind just shuts down.

Jimmy Hoffa

@YannisRizos Maybe it's a truly masterful troll. His question is +4 and no one has any idea what he's talking about

Yannis Rizos

I thought in ruby it's called class methods... In PHP it's static methods. But there are instance variables and methods. — random_guy 2 mins ago

There's your answer, he's trying to grok OO through PHP. That explains the confusion...

user55340

@YannisRizos did you see Bart's comment to a query about Jon's activity?

user55340

There is a bug in your query. It seems to suggest Jon Skeet has an off-time, which is obviously wrong. — Bart 5 hours ago

Yannis Rizos

6:17 PM

@casperOne I've tried that with SQL Server. It's a deep, dark place where all hope to answer questions is but a fevered dream of the madman. — Nick Vaccaro 4 hours ago

Heh.

user55340

Lets see what lurks in P.SE's oldest unaswered questions...

Yannis Rizos

17

Q: Y combinator and tail call optimizations

The definition of a Y combinator in F# is let rec y f x = f (y f) x f expects to have as a first argument some continuation for the recursive subproblems. Using the y f as a continuation, we see that f will be applied to successive calls as we can develop let y f x = f (y f) x = f (f (y f)) ...

f# recursion tail-call continuation

user55340

0

Q: printable PHP manual - 'all but the Function Reference section'

My Motivation I find it easier to learn things by reading 'offline'. I'd like to lean back and read the narrative part of a paper version of the official php manual. My Scuppered Plan My plan was to download the manual, print all but the Function Reference section and then read it. I have do...

php documentation documentation-generation

user55340

From programmers.stackexchange.com/…

user55340

1

Q: How do I use a preprocessor with multiple AS3 files?

I am trying to use the GNU cpp (C preprocessor) with multiple AS3 files, but from what I see the cpp command only accepts one source file and one output file (and using a directory's path didn't work). I would prefer some preprocessor that would work with flashdevelop (not limited to cpp).

actionscript

user55340

6:21 PM

Gah...

user55340

6:35 PM

How would you implement singleton without statics? — svick 16 mins ago

user55340

Just create one, then attach it to your main class or something, so everything is dependency injected, IMHO it's a better practise. If the arguments gets too long, then i have to find a make it shorter, in more elegant way. Yeah, it's a weird idea, no?, that's why i need to ask. — random_guy 6 mins ago

user55340

How do you attach it to the main class if the main class doesn't have statics?

user41796

@YannisRizos - does the IP addr check out ok for the Java no-methods, no-globals, etc... question? A little voice in the back of my head is wondering....

user55340

@GlenH7 I doubt it, just given the age of the account...

user41796

should I run and go flag, or ... :-)

user41796

6:46 PM

if true; would be a new tactic. Account was opened a while back and left to age for a bit.

Yannis Rizos

@GlenH7 There's nothing suspicious there. He's just young (his profile says 19), and trying to do too much with concepts he doesn't really understand.

We've all been there.

user41796

@YannisRizos - thanks. It's interesting to note how a well crafted troll question algins with a n00b question.

user41796

The responses (so far) haven't been hostile, which would be a key difference I think.

Yannis Rizos

That's the worst thing with Leslar's flavour of trolling, it's not just that he wastes everyone's time, but he also makes us suspicious of honest (but noobesque & confused) questions.

user41796

agreed. I've mentioned it before - A well crafted troll question requires a degree of intelligence. And the ramifications of it can be quite long lasting.

user55340

6:55 PM

@GlenH7 Look up Poe's Law...

user55340

Poe's law

Poe's law, named after its author Nathan Poe, is an Internet adage reflecting the idea that without a clear indication of the author's intent, it is difficult or impossible to tell the difference between an expression of sincere extremism and a parody of extremism. The law and its meaning Poe's law, in broader form, is: The core of Poe's law is that a parody of something extreme by nature becomes impossible to differentiate from sincere extremism. A corollary of Poe's law is the reverse phenomenon: sincere fundamentalist beliefs being mistaken for a parody of that belief. History The st...

user55340

Instead of extremism, noobism.

user41796

there is great truth in that law

Jimmy Hoffa

7:13 PM

@YannisRizos I think I could answer this question if i wanted, I'm pretty sure the code he posted does tail call, I could compile it and post the IL to show the tail call opcodes are there, but he would never see it to select it as the answer and no one else would see it to have any idea about it at all

Yannis Rizos

@JimmyHoffa Hm? Post an answer, people will see it.

The original asker may not, but everyone else will...

Jimmy Hoffa

...I'm getting tired, I meant to say no one else who saw it would even know what we were talking about (in large part)

Jimmy Hoffa

7:26 PM

@MichaelT My euler #1 clojure solution, is this seem idiomatic normal way of using clojure from what you've seen? paste.lisp.org/display/137007 granted it doesn't use the fancy O(1) method

user55340

Design your language, on paper. Use your language to solve real world problems, again on paper. After you've actually used your language a couple of dozen times, go and develop your language. Otherwise, you're reinventing PHP. — Gilbert Le Blanc 12 mins ago

Jimmy Hoffa

(avert your eyes those of you who've never solved euler #1)

user55340

It looks reasonable... this isn't python so there's no clojureific....

user55340

Consider the use of filter though...

Jimmy Hoffa

Oh yeah..

user55340

7:28 PM

And since most people aren't using clojure for Euler...

user55340

and the answer wouldn't help them write something else...

user55340

(defn e1 [limit]
  (reduce + (filter #(or (zero? (mod % 3))
			 (zero? (mod % 5)))
		    (take (- limit 1) (iterate inc 1)))))
(e1 1000)

user55340

IBM has a good bit on clojure using Euler problems - ibm.com/developerworks/opensource/library/os-eclipse-clojure

Jimmy Hoffa

ah zero? that's a good one, and I knew # was for lambdas somehow, but hadn't taken the time to get to them yet... though I'm confused how you're using the lambda function as a partial? do the holes in a partial just get filled in automatically even if they're not consecutive as such?

user20683

@MichaelT I'm debating trying to work through the SICP again

user20683

7:30 PM

now that I actually have time

Jimmy Hoffa

@WorldEngineer since when do you have time for anything? Aren't you working on some javascript insanity?

user20683

@JimmyHoffa that was done tuesday

user20683

and it was vomit inducingly bad code

user20683

I shall pretend I never wrote it

Jimmy Hoffa

haha

now I really want to see it

user20683

7:31 PM

@JimmyHoffa you can't since i never wrote it

user20683

;)

Jimmy Hoffa

Don't pretend you didn't write horrific code. We all did, never forget. I wrote something like 2kloc in a single method using structs instead of objects years ago not knowing the dfiference, it was pure procedural, I didn't even have members I don't think

user20683

I am debating doing the SICP in JS though

Jimmy Hoffa

that would be cool.

user20683

@JimmyHoffa I wrote 1k lines of broken Java in 6 hours for my Software Engineering project

user20683

7:33 PM

to be fair, I can write good code...probably

Jimmy Hoffa

I doubt it

No offense :)

user20683

@JimmyHoffa I can write code that functions

user20683

that much is true

user20683

and I know how it functions at the end

Yannis Rizos

@WorldEngineer Should we publish that blog post?

Jimmy Hoffa

7:35 PM

Yeah, but that's a long ways off from being able to write good code, none of us can until we've been in industry with good tutelage for a little while, or at least, the extreme majority of us can't

user20683

@YannisRizos about to look at it

user20683

give me an hour and then we'll publish it

Yannis Rizos

Cool. Didn't had a chance to look at it a second time, and right now I'm in my... beer time. It would be best if I avoided publishing anything.

user20683

@JimmyHoffa I do have the advantage of having a "me casa es tu casa" attitude toward the code I write in projects

user20683

@YannisRizos I like the ellipsis

Jimmy Hoffa

7:38 PM

@YannisRizos perfect, publish a 3 page article on tequilla

user20683

there is a homebrewing SE

user20683

go do a moonshine tour of the Greek mountains

user20683

then write about that

user20683

assuming you're still conscious at the end of it.

Jimmy Hoffa

He'll be concious, but he'll no longer have sight...

user20683

7:39 PM

@JimmyHoffa only if they use the wrong kind of still

Yannis Rizos

@WorldEngineer I do have an excellent mountain tequila story. The excellent part is that I didn't fall off the cliff.

user20683

@YannisRizos that sounds about like my grandpa as a young man. He was very much a drinking and fighting man

user20683

at that stage anyway

Yannis Rizos

7:52 PM

-6

Q: Programming, Bio-engineering, or both, or what?

I don't want a moral or optimistic response to this question. I want to be a programmer but I do not like humans, I believe in friendship but in general I look forward to a supernova or something similar, something the would wipe out humanity. So my concern is that if I am planning on destroying ...

programming-practices self-improvement

user20683

Troll

Jimmy Hoffa

@WorldEngineer reeeally?? heh

user20683

@JimmyHoffa I've seen the handle before but I can't place where

Jimmy Hoffa

I love that everyone is avoiding saying the obvious. Engineers the lot of you, efficient application of the rules on file always trumps creative application of new rules

user20683

I loathe extinctionists

Jimmy Hoffa

7:55 PM

like "No asking how to destroy the world on SE", though I'm not certain that rule would stand up under scrutiny with this crowd..

user20683

@JimmyHoffa it'd be a legit question in Physics or Biology or Philosophy .SE

user20683

kinda

Jimmy Hoffa

@WorldEngineer yeah, defeatism taken to it's extreme conclusion

user20683

@JimmyHoffa more than that to me but yeah.

user20683

I'm okay with people having fewer kids and all that and taking care of the environment but we owe it to the planet to at least try to make it a more healthy and diverse place

Jimmy Hoffa

7:57 PM

What's that latin fanciness for aggressive/recursive application of a logical rule until reaching the most basal and always ridiculous end?

I wish I had the ability to ever remember any of those logical oddities etc

Yannis Rizos

@JimmyHoffa Ad assholum

user20683

@JimmyHoffa Reducio Ad Absurdum

gnat

@YannisRizos unless I miss something, I believe that DV with comment pointing out the problem is really needed in such cases. You see, what would you expect my vote to delete to accomplish? I just checked, I've got 46 pending votes, most "ancient" one dated Jan 28 - and it doesn't look like these are going to have any effect in foreseeable future,...

user20683

lit: reduced to absurdity

gnat

...now what do we get? Look, there's a crappy question with some crappy answers, closed, and with oh! delete vote, yeah. Site visitors won't see that freaking vote, what they will see is that some piece of crap somehow got 5-10-20 upvotes. The question is closed, so what? if one gets 50-100-200 reputation for low effort shot, that's not a big deal. That is exactly what worries me. Visitors seeing low effort / useless stuff getting good score, will try to repeat it...

... They won't care if it's old, they won't care if it's to closed question, they will se it's tolerated and upvoted.

Jimmy Hoffa

7:59 PM

@WorldEngineer No, I think @YannisRizos had it

gnat

Don't get me wrong - I would really really prefer something easier than my current routine for these answers, but I honestly can't see how it can be accomplished. And, you see, your idea with voting to delete doesn't look like solution, does it?

Yannis Rizos

@gnat Then flag the question and ask for it to be deleted.

If 9/10 answers are crap, it makes more sense to just get rid of it than commenting on every crap answer...

gnat

@YannisRizos I see, that should work indeed. I actually do that sometimes already, just not too often

that sounds like a good plan, thanks

oh we discussed something like that at meta already...

Yannis Rizos

@gnat I regularly check pending delete votes as well, so in most cases your delete vote would be enough.

gnat

3

A: Are readers supposed to vote, edit, comment, flag answers on old one-liner honeypots?

If you can't find any evidence that the poster is still active on the site (recent posts/votes etc.) then there would seem to be little point in commenting as they are unlikely to see the comment and act on it. Similarly with down-votes. However, in both cases you aren't just telling the poster ...

@YannisRizos okay deal :)

user20683

8:04 PM

I generally annihilate or convert to comment the old, unused stuff

user20683

or the new crap

Yannis Rizos

@gnat Yes, a comment and a DV makes sense. However sometimes it makes more sense to just get rid of the whole thing in one swift motion.

user20683

if it's got 5 upvotes I generally leave it be

user55340

@JimmyHoffa I think I my solution was similar to the IBM one...

user55340

(defn divisible-by-3-or-5? [num] (or (== (mod num 3) 0)(== (mod num 5) 0)))

(println (reduce + (filter divisible-by-3-or-5? (range 1000))))

user55340

8:09 PM

Destroying humanity is not exclusive to the realm of software development, therefore your question is off topic. — maple_shaft 30 mins ago

Jimmy Hoffa

@MichaelT yeah, I didn't think to use filter; need to think aloud to myself how I'd do things in haskell before doing them in clojure..

user55340

> Weinberg's Second Law: If builders built buildings the way programmers wrote programs, then the first woodpecker that came along would destroy civilization.

user55340

We've already got destroying civilization down.

user20683

@MichaelT developing software is the same as the universe operating under physics. It's a complete mess held together by entropy, inertia, and gravity.

user55340

@JimmyHoffa My perlistic nature was showing when I wrote that - grep the list, and then sum it... oh, suming a list is a reduce thing in the language...

Jimmy Hoffa

8:12 PM

If builders were working under the adept managerial and organizing hands of software organizations, construction workers would die on the job by the thousands

user55340

@WorldEngineer SF book to read (its a good one)...

user55340

amazon.com/Implied-Spaces-Walter-Jon-Williams/dp/1597801259

Jimmy Hoffa

@MichaelT yeah, I only knew reduce because in haskell you fold stuff constantly, and I had to remember the etymology study I did once; fold in haskell == aggregate in .NET == reduce in lisp == something else in java

and yet something else in scala

user20683

@MichaelT

user20683

amazon.com/Clockwork-Rocket-Orthogonal-Greg-Egan/dp/1597802921/…

user20683

8:14 PM

this one is on my list

user55340

There is some in refrences in there to other SF books too that add a bit of help... Strossian singularity, and the Basilisk...

user55340

A web published version of a bit of the basilisk world... ansible.co.uk/writing/c-b-faq.html

Jimmy Hoffa

I strongly encourage this one

amazon.com/Jerks-Training-space-cadets-Stine/dp/0590447459

user55340

There's also a Greg Bear short story about where doing certain equations changes the nature of reality...

user20683

I've long considered writing something along the lines of a combo of "Snowcrash" and "House of Leaves"

user20683

8:16 PM

@MichaelT the one where wizards get introduced to the calculator?

BMitch

Seriously, you guys are calling in the mods about that?

user55340

No... its where people are 'hacking math' with huge computers to change the nature of reality...

BMitch

can't everyone play nice in the sandbox?

user55340

Wait, no, it might be Benford, not Bear.

Jimmy Hoffa

@MichaelT I heard an interesting bit on the radio the other day reviewing a book where a physicist is positing that he decided laws of physics are all subject to change over time, that time effects them all and they are not in fact actually laws

user20683

8:18 PM

@JimmyHoffa

user20683

amazon.com/Time-Reborn-Crisis-Physics-Universe/dp/0547511728/…

user55340

Try trolling Christanity.SE and Physics.SE with this book...

Jimmy Hoffa

'hacking math' just made me think of that. Gives an interesting idea; hundreds of millions of years ago the laws of physics may actually have not been precisely what they are now. That sensation that time is going faster as you age? Maybe it actually is. Reality itself is shifting over time, maybe that sense of time dilation is our physical reaction to that

user55340

amazon.com/Physics-Immortality-Modern-Cosmology-Resurrection/dp/…

user20683

This guy I think?

Jimmy Hoffa

8:18 PM

Yes

user20683

@JimmyHoffa I prefer the "Logarithmic Experience" theory on age and time perception

Jimmy Hoffa

Never heard it

The obvious postulate from it sounds logical though heh

user20683

as a practicing Christian and a practicingish scientist (in so far as CS is a science), I largely tend to not care about proving things like that

user20683

what are we ants to comprehend the nature of things? Abstractions, Analogies, and Approximations one and all and to the last.

Jimmy Hoffa

We live in the land of the halting problem, by definition being a computer scientist means you accept the unprovability of things, perhaps that's one of the major differences between us and physicists..

user55340

8:22 PM

@WorldEngineer The idea of the Tipler book is that at the end of the universe, there will be such time and energy to be able to calculate the entire pre history of the universe and simulate everything... and such... and he's a bit crazy.

user20683

@MichaelT sounds like he found a Boltzmann Brain and toked it.

user55340

feature request for universe 2.0

user55340

> Unfortunately, many of the Request For Enhancements mentioned above are partially implemented in the current version. The main reason was the rush to get it out the door - seven days is not enough time to go from rough specifications and a beta version to a full working version.

user55340

> Different requirements by the various marketing devisions have also confused the issue. Some wanted a soul, some wanted reincarnation, and others refused to promise the end user anything at all. With such wildly differing promises to the consumer, things got a bit hectic. When push came to shrove at which point one of the VPs said 'hey, it compiles! ship it!' and it went out the door with a big bang and much fanfare.

user55340

> Initial sales of universe v1.0 were good, and the sales market sold them faster than we could make them. Physicists now account for this period as the 'inflationary' period. After things cooled down a bit and sales leveled off customers kept sending in bug reports. As a quick patch the developer team wrote up a black hole hack and released it. This allows the user to shove anything that is causing a problem into the void and not worry about it again.

user55340

8:25 PM

> At no time should the event horizon be removed from this patch - naked singularities are prohibited by the Universal Decency Act. We are aware that in several trillion years, there is the chance that these may be exposed again, however we believe that most forms of government have difficulty over such lengthy periods of time and hope to have it repealed by then.

user55340

> Thank you for your interest in Universe version 1.0 and please be certain to update the service pack regarding dark matter and dark energy. Things seem to be picking up again and we would hate for you to miss out on these latest features.

Jimmy Hoffa

@MichaelT It's crap without the Higgs expansion pack though

user55340

I wrote that back in '00... Higgs was not as prominent then... and its harder to pun on.

user55340

"and it went out the door with a big bang and much fanfare."

Jimmy Hoffa

Aye, funny stuff, well done

user20683

8:30 PM

Programmers Stack Exchange Blog: Why CS teachers should stop teaching Java applets

Andrew Thompson on May 03, 2013

As a veteran of applet development (see credentials below), I think it is high time to make a call that teachers should stop teaching applets in general, & especially AWT based applets. Applets are an advanced and specialized type of app. that has little place in the real world around us (most things that are done by applets can instead be better achieved using Javascript and HTML 5).

It saddens me to see new students ‘tossed in at the deep end’ to learn how to make an applet.

The reasons are multiple, but first, I’ll look a little at the history of applets and why applets were ever considered a good idea for teaching small, simple projects. …

user20683

Cue the firestorm!

user20683

MWHWHAHAHAHAHAHHA

gnat

@YannisRizos for the sake of precision, I can't recall questions with 9/10 answers crap - guess these were deleted before. :) Typical "dose" I've seen in one-liner honeypots is about 1/3 - 1/2 crap

5

Q: Are readers supposed to vote, edit, comment, flag answers on old one-liner honeypots?

Some old closed questions have many answers that probably were considered OK back then but are not quite up to current quality standards. Are readers supposed to vote, edit, comment, flag answers in these questions? If it's better to leave these alone, then I would like to also understand how r...

support answer-quality

I did some stats when preparing above question

user55340

Another fun one I wrote... I like the idea of the "modern pantheon" of gods... and someone wrote about Casual Gods (the album)... so I wrote about a set of deities... everything2.com/user/m_turner/writeups/Casual+Gods

user55340

> The casual gods have extended their influence to create a temple of their own known as "TGIF", named after the prayers to the casual gods just before the weekend.

user55340

8:33 PM

Then I've got the Traffic Light Gods...

user55340

> The modern day god of the traffic light is a distant relative of the Roman god Janus (with two faces). The god of the traffic light (known by the abbreviated title of GOTTL) has three faces (though some statues of him have more). Each face shines in a different color - green, yellow, and red.

thorsten müller

Wouldn't the traffic light god be a relative of Legba, master of roads and pathways?

user20683

@thorstenmüller probably an adopted child

Jimmy Hoffa

@thorstenmüller no, his only living descendants are the God of Pot Holes, and Garmin

user55340

> GOTTL has the responsibility of making certain that all the cars move about on the great beltway in an orderly fashion - for he the child of order and chaos. When all is well, things move orderly, but at times his face takes to blinking red and yellow and all around there is chaos to behold.

user55340

8:39 PM

Roundabouts are created by heretics.

user20683

@MichaelT Two words: Spagetti Junction

Jimmy Hoffa

@MichaelT in clojure is it your understanding that [ bla bla bla ] (vectors?) are just clojure's version of tuples?

That was confusing me at first until I came up with the conclusion to just use them wherever I would use tuples in haskell

user55340

I thought it was closer to Lisp's quoted lists...

Jimmy Hoffa

...I don't know what those are...

is it easy to add and remove things from a vector?

that is grow and shrink it

user55340

26

A: When to use 'quote in Lisp

It says "don't evaluate me". For example, if you wanted to use a list as data, and not as code, you'd put a quote in front of it. For example, (print '(+ 3 4)) prints "(+ 3 4)", whereas (print (+ 3 4)) prints "7"

Jimmy Hoffa

8:44 PM

That is totally not the understanding I got of vectors...

user55340

Let me get the current version of light table so that I can fire up some clojure...

Jimmy Hoffa

getting light table to run in windows is a huge pain

The enormity of the executable seems to just chug, I'm guessing mccafee is just choking to death on it

There's probably clojure mode for emacs

ah not by default..

user55340

Mac and Linux are not bad...

Jimmy Hoffa

yeah, I was running it in linux at home

really cool IDE for sure

user55340

The LT I had on this machine was an early version... it just had a rev recently.

user55340

8:50 PM

... and its really changed from what I remember...

Programmers Blog

Programmers Stack Exchange Blog: Why CS teachers should stop teaching Java applets

Andrew Thompson on May 03, 2013

As a veteran of applet development (see credentials below), I think it is high time to make a call that teachers should stop teaching applets in general, & especially AWT based applets. Applets are an advanced and specialized type of app. that has little place in the real world around us (most things that are done by applets can instead be better achieved using Javascript and HTML 5).

It saddens me to see new students ‘tossed in at the deep end’ to learn how to make an applet.

The reasons are multiple, but first, I’ll look a little at the history of applets and why applets were ever considered a good idea for teaching small, simple projects. …

user55340

Hey - its your favorite topic... "Zippers - Functional Tree Editing"

user55340

Ok... playing with it... I kind of remember when I was playing with it before..

Jimmy Hoffa

@MichaelT What are you looking into zippers for? Other than because they're awesome :)

the first chapter of okasaki's is what made me understand them, they're confusing as crap at first

(for me)

user55340

clojure.org/data_structures

user55340

8:57 PM

Related functions
Create a vector: vector vec vector-of
Examine a vector: get nth peek rseq vector?
'change' a vector: assoc pop subvec replace
See also zippers

Jimmy Hoffa

Ah yes, they aren't tuples, tuples are O(1) access but O(n) grow/shrink

user55340

> Vectors support access to items by index in log32N hops

user55340

So this is a B* (if I remeber that right) tree.

Jimmy Hoffa

I am just confused, it seems in the example code I've been seeing, they're used all over, as much as if not more than lists, which seemed strange to me

user20683

I feel like I learn so much more sitting in here reading you twos banter than I did in most classes I took

user55340

9:01 PM

Lists are LinkedLists... and only have limited access to it. Vectors are much more array like.

user55340

List - adding a new item is at the front... Vector, adding a new item is at the end.

Jimmy Hoffa

@MichaelT yeah, and that's exactly why I want to treat them like tuples

tuples are O(n) grow/shrink but O(1) index access (and the count is a part of the type); they model determinism where lists model nondeterminism

but they don't have the grow/shrink characteristics of tuples I'm thinking, so they're just lists that have been tweaked for performant index access (by being a tree)

user55340

If the vector is less than 33 entries, it is O(1)

Jimmy Hoffa

ah

I'm going to pretend they're tuples until I know better. It will at least give me a sense of when and why to use them

user55340

That's the index log32(N) hops... so log32(32) = 1... so you can access something 32 entries or smaller in one hop, which is the same as just saying "its there"

Jimmy Hoffa

9:09 PM

aye, you access tuples as:
first (a,b,c) = a
second (a,b,c) = b
third (a,b,c) = c

But they run into trouble because the size is a part of the type information so if you want to do something across all members of a tuple that may have variable size, you'd have to implement that function separate times for each length tuple you support.

tuples are invariable size, vectors I'm guessing are variable size...

user55340

> conj puts the item at the end of the vector.

user55340

So yes.

user55340

Though...

user55340

> conj[oin]. Returns a new collection with the xs 'added'. (conj nil item) returns (item).

user55340

so it isn't modifying the vector, but rather returning a new one.

user55340

9:19 PM

@JimmyHoffa So... vectors are not variable in size. Once it is instantiated, thats it. There is a transient vector, though that looks like you can't actually get at it.

user55340

Nope, that's a different thing....

Jimmy Hoffa

@MichaelT can you map and fold and do all the list things to a vector?

Is there a way in light table to get the sig of a function? Best part of haskell's repl:

Prelude> :t map
map :: (a -> b) -> [a] -> [b]

Yeah, they're just...arrays... you can always use a list where you use a vector, and vice versa, but which one you use I guess is just a switch you make here and there to tweak performance characteristics for particular use cases...

Jimmy Hoffa

10:09 PM

ack, clojure lacks good pattern matching.. for a language that expects you to rely on recursion, that's near unforgivable because pattern matching is the best way to implement your exit cases in recursive functions

user55340

@JimmyHoffa clojure.github.io/clojure/clojure.core-api.html#clojure.core/… - there's the map.

user55340

I also find clojure.org/cheatsheet useful to see the "what do I want to look for" useful.

Jimmy Hoffa

Yeah, the documentation is good for sure

user55340

@JimmyHoffa It's not an Apache incubator project... (ohh... burn...)

Jimmy Hoffa

map was giving me crap until I realized I needed to '(my list) as you mentioned earlier

I'm trying to make a macro for a more haskell like lambda

user55340

10:21 PM

(map inc '(1 2 3)) --> (2 3 4)

(map inc [1 2 3]) --> (2 3 4)

user55340

The '() is a List, while [] is a Vector. Both collections.

Jimmy Hoffa

Right, I think it calls them ISeq

user55340

ISeq is an interface that extends the IPersistentCollection and adds first, next, more and cons methods.

user55340

While IPersistentCollection just maintains a count, cons, and empty.

user55340

github.com/clojure/clojure/blob/master/src/jvm/clojure/lang/…

user55340

10:26 PM

github.com/clojure/clojure/blob/master/src/jvm/clojure/lang/…

Jimmy Hoffa

@MichaelT postfix notation is the only notation isn't it? lisp as a rule doesn't allow infix functions does it?

user55340

@JimmyHoffa Eh? Lisp and Clojure use prefix notation only.

Jimmy Hoffa

postfix prefix... whatever it's called

but no infix

user55340

Postfix (think forth):
5 4 3 + * ---> 35

Infix (think C)
5 + 4 = 9

Prefix (think lisp):
(+ 4 (* 3 2)) ---> 10

Jimmy Hoffa

(defmacro / [arg & body]
  `(fn [~arg] ~@body))

(reduce (/ x (/ y (+ x y))) [1 2 3])

no currying fail..

Jimmy Hoffa

11:01 PM

Maybe this would fit me more...

Qi (programming language)

Qi is a functional programming language developed by Mark Tarver and introduced in April 2005. A new version was reimplemented and issued as Qi II in November 2008. The first version was free software, licensed under GPL. But, as GPL was perceived as unfriendly to commercial use, Qi II is available via two proprietary licenses: one for personal and educational use, and another for producing closed source software. Qi is written in Lisp. It includes most of the features common to modern functional programming languages such as pattern-matching, currying, partial applications, guards and (...

a lisp with automatic partialling/currying, built in pattern matches and guards

programmingkungfuqi.blogspot.com

apparently it's trying to just take all the haskell stuff and make it available to CLOS programmers

has type classes, which is awesome.

Transcript for