« first day (1336 days earlier)      last day (3952 days later) » 

user55340
00:03
@gnat singer / songwriter to look up: Gregory Alan Isakov
user55340
15:27
wow, it's as quiet in here this morning as it is in my office. Normally I expect both to be making more noise by this time... perhaps everybody woke up feeling as shite as I do
Work + Headache = This day's going to suck
user15026
@JimmyHoffa Yeah, that sums up my day so far as well.
@RobertHarvey the cont monad on the other hand, natural translation Haskell->C# is considerably more...confusing: gist.github.com/JimmyHoffa/388f2f350f135addf2fc
the "operation" in the cont monad is a function that takes a function and returns an R; this way when you want to terminate the cont and execute it - you just give it the last function in the chain of continuations, and it executes the chain below it returning the R. seems strange seeing Func<Func<A, R>, R> but you can think of it as Func<data-hole,R> - eventually you'll terminate that datahole with the last function and get the R you were chaining funcs to generate
16:10
...ghost town in here...
user55340
Slow day...
user55340
and its not friday yet.
user55340
(This afternoon is a drive down to my parents to juggle them for some home repairs up here on Friday...
user55340
Being younger, I've got less problems with long drives or driving at night than my parents do... so I'm driving down there this afternoon, staying the night down there, driving back here tomorrow morning, working on the house, driving back down friday evening... then I want to go to Mitchell Conservatory in Milwaukee for some photos with the Lytor...
user55340
and then back up here either saturday evening or sunday.
user55340
16:40
@JimmyHoffa some 10k fun for you: programmers.stackexchange.com/a/111450/40980
@MichaelT 5300 rep?? WTH??
user55340
@JimmyHoffa Look at the graph of the reputation. Its from NPR early days.
user55340
16:58
@JimmyHoffa Thanks. That still looks simple, if extremely generic. Which, I guess, is the whole point. Once one understands first-class functions, it's not all that difficult to comprehend. It's the proper and practical use of them that confounds.
just from look yes, but the use is actually easy; you would do it like this:

ModelOperation<int, int>.Return(42).Bind(Increment).Bind(Decrement).Bind(Increment).Operation(Decrement), that would take 42 put it in the cont, update the cont to increment, then to decrement, then to increment, then where I actually call the operation I'm actually executing it with the last step being to decrement. That would return 42.
where Increment and Decrement are functions of type Func<int, ModelOperation<int, int>> (take a number, return a model operation of type int to int)
Looks like a Fluent Interface, with first-class functions as method parameters.
@RobertHarvey bingo. that's what bind always is
Pretty neat.
user55340
I would appreciate if anyone ( @gnat ?) could poke at Rev 11 meta.programmers.stackexchange.com/posts/6487/revisions to make sure I've got it worked right or if there is additional information there that would be useful.
user41796
17:05
I was a bit foolish and dropped a question on SO before heading out to an appointment. Fortunately, the Interwebs didn't blow up on me.
user41796
2
Q: Safe to compare C# doubles against assigned values?

GlenH7Is it safe to compare C# doubles or floats against assigned, default values? For example: public class Foo { private double defaultEfficiency = 100.0; public double efficiencyBar = defaultEfficiency; public bool IsBarAtDefaultValue() { if(efficiencyBar == defaultEfficie...

very bad! you made a typo there! :) I corrected. Other than that, addition looks good to me, and makes much sense - askers seem too often be confused about this
user55340
@gnat There were two comments I made today about that off topic reason. It might even need to be bounced around that 'favorite off site resource' be explained a bit more in the close reason.
psr
psr
17:29
@JimmyHoffa Not to me. Sure, 90% of the text is telling the compiler the types of everything, but for me (not that fluent in functional programming) it's nice to have it there. But I can see that after a while it's a lot of redundant noise.
18:00
@MichaelT Robert Harvey may like it. If memory serves, he frequently complains when our sites are used as surrogare of search engine
18:33
@MichaelT I find that whole argument about "people don't understand the close reason, so maybe we should change the verbiage" entirely without merit. People will obsess endlessly on Meta about which words should be bolded, and whether a punctuation mark should be there (as if that would make any difference).
Users whose questions get closed will always complain, regardless of what the close reason says.
@psr that's a good description of why Haskellers like inference so much, languages with all the ceremony rapidly appear to have tons of "redundant noise" that just gets in the way of the actual meaning of the code.
Increasingly, that is the value that I'm finding in these functional languages. They simply have less noise.
Of course, the folks who write large, enterprise programs in curly brace languages actually value much of that ceremony. It provides structure and signposts to the code, making it easier to read (albeit much more verbose).
@RobertHarvey unfortunately the straw man argument many use very effectively against all of that is to claim "shorter doesn't mean better"
@RobertHarvey easier to read to them. It's like phonics, it may be easier to read but after you know how to read proper english, phonics is harder to read than the alternative
Perhaps.
The problem is, the ceremony is the crutch by which we were all taught. So now we have to learn how to do it without the ceremony.
@RobertHarvey if you take all the extra cruft away and you can still understand what the code does, then would it be easier to read the code when the meaning is conveyed with less or when the same amount of meaning is conveyed with more?
@RobertHarvey which is why it's always said learning Haskell first is better (LISP is another no-ceremony language which is a good first for this reason)
18:41
Less code is always more desirable. The problem is, the less code you use during the learning process, the steeper the learning curve becomes.
@RobertHarvey that's an assumption
But a good one.
I disagree. In Haskell less code conveys concepts more clearly because it's good, it's not good because it's less
Learn in in Java first, then unlearn much of it when you get into the languages with higher expressivity. Or, try to learn Haskell first without going insane.
@RobertHarvey I think learning Haskell before you're used to thinking statement1; statement2; statement3; would be far easier
anywho... I hardly ever find a Q on P.SE anymore that has any interest to me...
18:44
Highly expressive languages really obscure the way a computer works, though. I think programmers still need to know something about that, or they will never be effective where the rubber meets the road, which is "I've got this beautiful, elegant algorithm in a highly expressive language. Now how do I get it to run in 1 second instead of 100?"
Not sure if that's me, or the site
@RobertHarvey What's a greater cause of problems these days, people not writing quality maintainable well-decomposed code, or people writing horribly inefficient code? I would say the former, and often it's people too focussed on how the code interacts with the machine that write much of it
Well, you already conceded that Haskell won't help the guy much who's just trying to make his web page work. Which, I daresay, is probably 90 percent of all the applications being written out there (if you don't count mobile). Boring business logic applications.
It's accepted that performance critical code may be ugly, hard to maintain, confusing, etc. Many people write most of their codes in not dissimilar ways because they think in terms of how the code operates instead of in the abstract
@RobertHarvey yeah but they're all written by people making $18/hr at a PHP or rails studio, churning out 2 whole new websites a month, they're a totally different group than the professional software engineers
psr
psr
public Func<AwesomeFeature<X>,Boolean> YourLanguageIsBetter(AwesomeFeature<X> reason)
{
return new Func<AwesomeFeature<X>,Boolean> (f => ImpliesBetter(reason));
}
public Bool ImpliesBetter(AwesomeFeature<X> reason)
{
return false;
}
it's a great job for many, and some of them will end up moving into real development with large problems and projects that require more than fiddling with wizards and config files
This question gives the appearance of minimal (none?) knowledge of programming, if you show us some of what you've done for this issue we might be able to answer with information at your level, as it stands though the way this reads I'm hesitant to give any suggestions as I suspect it might just be out of your experience and give you more questions than answers... — Jimmy Hoffa 21 secs ago
none of us answer Q's on SE because we've learned our way around NPM or GEM and can churn out webpages by executing package-installation-commands that construct a standard packaged website. We answer Q's on SE because we are interested in how to design around and work on large technical problems while striving for quality.
19:08
I know Thursday's not Friday 'n all that, but this is just eerily dead today. Calm before the storm?
-1, This really doesn't explain your stance that it is "still a C++ framework" - or even detail what you mean by that statement to begin with. Because it's written in C++? Because you qan use C++ to work with it? Because it was originally used in C++? Because the other language(s?) don't qount? Because C++ is preferred? Because it is the preference among C++ devs? I just don't understand from your answer why you make the statements you do, or what they fully mean. Add some more explanation and I'll gladly give +1 if it makes sense afterwards — Jimmy Hoffa 2 mins ago
psr
psr
I can't see how that question could be answered. What would a good answer look like?
Oh, your comment just showed up for me. Stupid caching.
@psr I think a good answer would have replace all the hard Cs with Qs, that's about all I've come up with for quality-metrics on the answers.
@amon really? The answer doesn't make clear how the answerer defines a "C++ Framework" - I genuinely thought the answerer meant because it was written in C++ even though not only used for C++. But further the real problem is the lacking explanation for why they claim it is a "C++ Framework" and not a "JavaScript framework" or whatever else
psr
psr
@JimmyHoffa "It will in fact be considered a C++ library, though primarily due to apathy and inertia. After 3 signs appear, the last one from the hidden moon, then will it be called a Q library."
@JimmyHoffa I really don't get what's so hard to understand about the “framework” concept. Yes, its a vague term, but it's still a standard software development concept. A part of the question translated into .NET terms: is .NET still a C# framework if it also supports VB?
In computer programming, a software framework is an abstraction in which software providing generic functionality can be selectively changed by additional user-written code, thus providing application-specific software. A software framework is a universal, reusable software platform to develop software applications, products and solutions. Software frameworks include support programs, compilers, code libraries, tool sets, and application programming interfaces (APIs) that bring together all the different components to enable development of a project or solution. Frameworks contain key dis...
In the case of Qt, this is actually a lot clearer. Qt extends the C++ languages by the signals&slots feature, which implies a very strong binding between the C++ language and the Qt framework, although many other languages like Python also have bindings.
19:27
@amon It's not the "framework" concept, it's that it's a "C++" framework - that's like saying .NET is a C# framework which makes no sense other than to refer to the fact that it's written in C#, it would be totally inaccurate to call it a C# framework for any other reason
I don't know much about Qt or C++ so I may be at a disadvantage in trying to understand the answer; perhaps I'm just missing something one with such experience would find obvious
0
Q: What specifications in software development are relevant to the clients?

Jane LeeSay your company develops software, and a client wants you handle their development. Just so happens, they have their own IT-team, but is fully tied up, so they outsource you. The project is in a quotation phase, and your client's Head of IT now asks for a requirement specification(SRS), an arch...

@JimmyHoffa Of what relevance is it to a programmer whether or not something is a "framework," by the "formal" definition? I voted to close as "primarily opinion-based."
@RobertHarvey I think everyone can concede it's a framework, the question is if it's a "C++" framework or some other kind
either way, the question is pointless
@JimmyHoffa so you interpret “$language framework” as “framework exclusive to $language”, whereas I'd think “framework which can be used by $language”? Is this the essence of our misunderstanding?
Software Framework is a language agnostic term. The only way it can be a C++ framework is if it somehow underpins the C++ language itself.
@amon I have said in as many ways as I can - I don't know how to interpret "$language framework" which is why I was trying to ask in my comment for the answerer to explicitly define why he says it's a "C++" framework rather than a "javascript" framework or "agnostic" framework
19:36
@RobertHarvey In a sense, Qt does that. To use Qt from C++, you need to use a special preprocessor which handles the “signals and slots” feature, which is a language-level implementation of the observer pattern
I can see both ways you suggest there as being his meaning, but I don't know which he means because he's not explicit about it
@JimmyHoffa ok, great, misunderstanding resolved
So maybe it's a C++ specific framework.
21
A: ADHD : Leaving the toilet seat up

XT-8147Am I the only person who thinks that it's entirely trivial for the next person to use the toilet to correct the seat position for their needs? I don't see why this is worth complaining about. Just teach your kids to make sure the seat is where they need it to be so that they can do what they ne...

haha, that's a gender equality issue... wow... talk about painting with bright colors and broad strokes...
psr
psr
OK. One rep cap away from 10k.
19:51
@psr yay, me too! (actually, I'm 8 rep ahead of you :P ). Let's hope for a hot question tomorrow.
20:07
Probably a stupid question, but what kind of programming jobs (which domain or language or company size or whatever criteria you can think of) do you think involve the least amount of configuring stuff (like, writing something to configuration files, setting things up, installing etc.)?
@iCanLearn Wrong question. The right question is "How much of my programming effort can I get out of the programming realm, and into configuration?"
What you really mean is "how can I avoid the boring stuff?"
The way you do that is by getting better at the interesting stuff, and then showing people how valuable you are doing the interesting stuff instead of the boring stuff.
20:31
@psr congrats. You can now delete posts! (assuming a certain voting spree doesn't get reverted)
psr
psr
@amon Wow, that was like being John Skeet for 10 seconds.
@Robert Harvey: You're right about what I really mean. I guess your answer is right too, there's no easy way out.
20:52
@MichaelT: Good music, but it demands more attention than I have while I'm programming. I'll have to listen to it when I have some quiet free time (the whole album is on Grooveshark).
 
1 hour later…
22:15
@iCanLearn architect.
@JimmyHoffa icanhazuml?
@RobertHarvey yep, no configuration there.
22:39
"How much of my programming effort can I get out of the programming realm, and into configuration?"
ugh
reading that makes me cringe
heh
I for one am tired of repetitive structural work in programming
I feel like we should be at the "I need a wall here, and a column here and wiring here" and not needing to manually build the walls, columns, wiring etc myself.
and the walls and columns would all be standard and made the same way.
except the reality is that the walls and columns will always be different
to roll with your analogy
@Zeroth that's a nice thought, problem is everyone elses walls and columns and wiring are shite, and if they're any good, they're only good for jazz clubs with black and red color scheming, meanwhile you need a wall for an auto garage and all the off-the-shelf ones are either triangular for some fucking reason or have dead bodies stuffed in them claimed to be insulation
in the old days there weren't walls, people spec'd out timbers, nails, and drywall in assembly
eventually some of that became common, and now you think of the wall as a contiguous standard unit
Hmm, yeah.
22:45
and once all the walls and colums are standard and made the same way
I'm just tired of writing yet another loop to iterate over a vector
@whatsisname then we'll only have one application that we all use and it only does one thing
then people are going to just want to be at the "I need a room here, and a hall here, and a kitchen here"
@Zeroth then write map, do it once, and never iterate again!
"and the rooms and kitchens would all be standard and made the same way."
22:46
with a passed in function pointer?
I wish.
the stuff I need to do in the vectors is always just different enough I couldn't make a generic function pointer
@Zeroth ? now you're just asking for something simple, and that's boring. map sure beats the wall full of dead bodies.
and after that it will move upwards in abstraction
"wall full of dead bodies" looks at code
sigh
when there's three or four code paths to almost any state...
lemond said of bicycle racing: "It never gets easier, it just gets faster"
the code becomes very hard to maintain
22:47
a similar thing will happen with programming
as our ability to manage code improves
@Zeroth mmm, NCube, yes...
splorfle
the complexity grows until it is once again at the limit to our ability to keep it under control

N^3 parsing

Jan 2 at 20:03, 3 minutes total – 15 messages, 3 users, 0 stars

Bookmarked Apr 24 at 16:08 by Jimmy Hoffa

I see some of your files have exactly 2439 lines of code in them...
to be fair
we do have some of that abstraction already
in gui toolkits
button here, scroll list here, header here
and this spot needs to be rainbow only on contradictory conditions
22:50
Then just use C, ffs. — Robert Harvey 26 secs ago
let me be very clear about this now, punchcards are not an abstraction. — Jimmy Hoffa 8 secs ago
Says you, hater of punch cards.
Hahahahaha
@RobertHarvey whoa whoa, I'm not hating, just saying...you don't get much more concrete than code you can hold in your hands
want to delete some code? Just feed the punch card to the shredder. Abstract? no...
[still getting used to Jimmy's Gravatar being a twisted version of Zeroth's]
22:53
excuse me? zeroths gravatar is a twisted version of mine.
in fact, with my 2.66 years on P.SE I believe I have the right to make a takedown request of Zeroth's gravatar. I insist that you replace your gravatar with a picture of a kitten
Okay.
How about a pet snake or pet gecko? I have one of each.
@Zeroth both.
aw, no!
just changed my gravatar, what do I do now? leave and come back?
reminds me of this sweet pic I saw last week of a lizard being eaten by a snake tail first
ah I tried to refresh but it's cached. I must insist that you just go somewhere else and not come back until you've stopped infringing on my intellectual property. Get off my lawn
22:58
NO!
You continue to use them because they are not the vacuous black holes that discussion forums are. — Robert Harvey 2 mins ago
@whatsisname iduno, I just see pure stoicism. Full-on "This doesn't bother me, you can't hurt me, I'm fine; 'tis but a scratch" written all over that lizards face
psr
psr
@RobertHarvey On the bright side, you evoked a "hurp hurp derp".
Yeah, I enjoyed that.
23:04
what an ass
stick
asked the wrong question and gets pissed off about it
He thinks the main site is a hangout. This is the hangout.
guy wants to build a 50TB raid 0 array
lolz
lol
23:13
And he can't afford $30 per hour. He's probably trying to build another CraigsList.
reminds me of someone that posted "I want someone to build me a clone of google. Serious offers only!"
good ole craigslist gigs section
torrentor
"About computer problems: has to deal with audios, with being able to stream from NAS on smart tv and with other misc. issues."
Sounds like the vapors.
so
it turns out
that even in 2014
people write new code that is hardcoded to read files from C:\
and write to those files
for very expensive software, and in situations where the developers should know better
makes me sad
As opposed to doing what?
Oh, you mean the root of C:\
Yeah, that's a really bad idea.
I don't think Microsoft ever bothered to solve that problem.
23:29
yeah, the root
23:45
All it takes to solve the problem is to add a folder to the root. But even that is not ideal. Life used to be much simpler; now you have to use "$app_folder" and "$app_data_folder" using Microsoft's byzantine folder structure, and hope you don't run into any security issues.
or
a more sensible solution, is to have the caller pass you a filename, then you don't even have to worry about it
but that somehow slipped past these guys
Ack. They're not using File/Save dialogs?
0
Q: Classes shouldn't call other classes?

Brian SnowI've just finished reading The Art of Unit Testing, by Roy Osherove. It was an interesting book, but I'm having trouble with something he mentions near the end: Identifying "roles" in the application and abstracting them under interfaces is an important part of the design process. An abstr...

Anyone understand this guy's question?
they're not using anything
its software used with ECG analysis
and it reads a specific filename from C:\
then writes it back out to that filename
that question was pretty confusing to me, I thought about answering it but was too lazy
It's not clear enough to be answerable. That's what happens when you quote someone out of context; happens all the time in the media.

« first day (1336 days earlier)      last day (3952 days later) »