The Whiteboard - 2016-03-06

« first day (2011 days earlier) ← previous day next day → last day (2978 days later) »

1:23 AM

I don't wish to be rude, but this site is for "professional and enthusiast programmers", it's not - by design - a site to offer tutorials to people that don't know what they're doing. Before asking here please: try to learn some JavaScript. Break the problem down into small steps and use a search engine of your choice to find out how to implement each individual step. Once you've written code to solve your problem then come back and ask for help, explaining in what way your code "doesn't work". — David Thomas 19 secs ago

user55340

2:00 AM

@Ixrec on that "math.se" thing...

user55340

-1

Q: Error message in fortran code

I am attempting to solve the heat equation in two dimensions numerically using Fortran. I obtain an error message when I try to compile the code, I cannot figure out how to fix it. The message is " Named Constant 't' in variable definition context (assignment) at (1)". In the code below, I r...

algorithms math error-messages fortran numerical-equations

user55340

2:49 AM

@GlenH7 and @Ampt - for some engineering and woodworking : youtube.com/channel/UCcXhhVwCT6_WqjkEniejRJQ

user55340

3:16 AM

btw, hello @AnissaAlexander - you need 20 rep to chat in here. I've written a bit about this in:

user55340

7

Q: How do you get 20 rep to get into chat?

People keep telling me that there is this wonderful area in Stack Exchange where you can ask questions called "chat" or "The Whiteboard", but it keeps saying that I need 20 reputation. I have such trouble asking questions here, I can't even get above 1 rep. How do I get to that much rep when even...

support reputation new-users chat

3:39 AM

@PreferenceBean The timer should reset every time you try to change it.

"another 30 days as PB"

7 hours later…

BarryTheHatchet

11:08 AM

@AaronHall -.-

5 hours later…

3:46 PM

I think this would better be a programmers.SE question. — Tersosauros 18 secs ago

user55340

@Duga Thats a question that was asked in '12, and closed as a duplicate in '14 that is a entire one sentence question that is "have you read the standards?" Gah... the things people try suggesting asking here.

user55340

20

Q: What's the difference between the implements & extends keywords in Java

What's the difference between the following keywords in Java: implements, extends?

java inheritance extends implements

not to mention it just got OP-deleted

OK now I think we should make you wait a year for a name change.

it should be exponential: 1 day the first time, then 1 week, then 1 month, then 1 year

4:00 PM

In IT, information comes before technology. That's our architecture thought for today.

4:13 PM

Android APIs are generally terrible. I bet you the android dev suicide rate is through the roof.

user image

That's the state machine for the MediaPlayer class.

I fail to understand why the dev needs to know about the "Prepare" set of states.

I fail to see why "idle", "playing", "paused" and "error" is insufficient

in fact I'm not even sure "idle" has a purpose

BarryTheHatchet

4:57 PM

happy mother's day from mobile chat beta ui

5:08 PM

No. I am your father.

BarryTheHatchet

let's just barry the hatchet

Who is Barry, and why/how did they hatch? Is this a Programmers-related pun? If not, your name is off topic for the main site. Acceptable for chat though.

My name is even more off-topic then.

BarryTheHatchet

5:46 PM

@amon How can a name be "off-topic" now???

@BarryTheHatchet Isn't everything off topic on Programmers? /s

it was probably a joke

6:09 PM

What is current average downvote count on main first page? 5?

-4 gets it off the main page, so that's impossible (unless you made the all questions page your "main")

yeah it looks pretty normal to me

@JohanLarsson I've been on tour for the last 4 days, and when I came back, first time I looked at the site, 12 out of the first 13 Qs on the "newest" page were either already downvoted, closed, both, or about to be. (One could also say the first 9, or 13/15.) That's pretty much standard.

Currently, it's something like 22/24.

Ah, one just got deleted.

I had an answer migrated, guess it will be downvoted now :)

Not important

which one? (if it was on a legitimately ok question it's probably fine, though a lot of what gets migrated here is also crap SO didn't want)

6:25 PM

2

Q: Visibility of abstract factory's implementation's product

I'm implementing a factory like this: public interface IMyProduct { void DoSomething(); } public interface IMyFactory { IMyProduct CreateMyProduct( string aParameter ); } internal MyFactory : IMyFactory { public MyFactory( ISomeService someService ) { _someService = som...

was from code review

ah yeah, that one looks fine

judging by the comment by op I did not understand the q

it is frequently the case on this site that a question has what we believe to be insufficient detail/explanation, people attempt to answer it and call us jerks for close-voting it, then the OP says they wanted something completely different and we just facepalm

:)

my personal belief is that most novice programmers drastically underestimate how much "the correct solution" depends on your specific use case, and seem to think there's a simple list of good ways to do X they need to pick from which all the good coders have memorized

6:30 PM

yeah but there is no substitute for the 10k hours

nope

a less common but equally strange pattern is the answer that only answers a small piece of the question, or explains why the question is impossible to answer...and the OP accepts it

can perhaps be explained by politeness

user image

yes I stopped in just to post this. You got a problem? Wanna fight about it?

no

user55340

yesterday, by MichaelT

user image

user55340

6:44 PM

Not at all.

user55340

(stars on the sideboard need thumbnails for one boxed images)

2

@MichaelT You beat me to it

not often there is code that can be copy pasted from so ime.

surprisingly often answers with high vote count have poor code

7:04 PM

Questions that ask "where do I start" are typically too broad and are not a good fit for this site. People have their own method for approaching the problem and because of this there cannot be a correct answer. Give a good read over Where to Start, then address your post. — Jan Greve 41 secs ago

what did duga pick up there^?

user55340

@Duga I love it when I see that post show up... just wish the OP would have linked the 'share' version so that we could get all those publisher badges.

user55340

44

A: Where to start?

The "where do I start?" question has several issues with it that make it a poor question to ask. It is unclear what the level of knowledge of the asker is, has too many possible answers, and ultimately doesn't have a clear problem defined. Unclear Let's head over to Seasoned Advice (Cooking St...

user55340

A link to meta.programmers.se

7:17 PM

@BarryTheHatchet where did this March name come from then?

@MichaelT ... now you have me wondering about a script for precisely that....

BarryTheHatchet

7:43 PM

@JimmyHoffa answer is in transcript. do teh search

Barry are you lrio?

8:04 PM

@JohanLarsson correct

I may have a question for programmers:

I'm looking for something to use as baseline in benchmarks

for example:

for (int i = 0; i < 1000000; i++)
{
    sum += i;
}

user55340

Awkward as sometimes it gets optimized by different things.

user55340

Do that in C, and cpp will make it a constant.

the reason would be to remove the effects of hardware

The problem with such a baseline is that it's extremely dependent on optimization. In some languages you can prevent optimizations via volatile variables or dynamic linking, but it's awkward.

8:15 PM

@MichaelT yeah, that would be my question

I would tag it c#

user55340

There's the story of Haskell optimizing an entire program to a no-op because it didn't output anything.

:)

user55340

With Java (and I presume C#) you get into issues of the JIT kicking in.

Would be nice to have a way to measure how an algorithm performs in a hardware agnostic way

Yeah a warmup would be required but benchmarkdotnet handles that nicely I think

can still be tricky due to cache sizes etc

Benchmarks are always relative. E.g. Solution A has 102±12 iterations per second, whereas solution B has 94±3. Benchmarks are just statistics.

Linux uses BogoMips for a hardware-dependent timing

“the number of million times per second a processor can do absolutely nothing”

user55340

8:19 PM

I suspect the answer is "compile it to byte code and analyze that"

not much of an answer for notrivial stuff?

I'm thinking putting benchmarks in version control and track them over time

that is when hardware agnostic would be nice

user55340

(related for the question of "what is the fastest way to flip a boolean in Java" - javaspecialists.eu/archive/Issue042.html )

user55340

It gets into some of the problems of benchmarking

user55340

time for flag ^= true: 12397ms
time for flag = !flag: 11356ms
time for flag = flag?false:true: 11326ms
time for flag ^= true: 5697ms
time for flag = !flag: 11326ms
time for flag = flag?false:true: 11326ms

user55340

> We can learn two lessons from this:

flag ^= true is faster than flag = !flag
Never trust Java performance statistics.
Don't forget that an intelligent compiler could've recognised what you were doing and done it on a bit level. There are many factors that affect Java performance: architecture, compiler, hotspot compiler, hardware, etc. and these all play a role when it comes to determining performance.

8:22 PM

@JohanLarsson Hardware-agnostic performance is what Big-O notation/algorithmic complexity is about. Remember that O(f(n)) can be derived from a term denoting the run time by eliminating hardware-dependent constants until only the scaling behaviour of an algorithm is represented by the expression. This takes a hardware model into account, but not the actual hardware performance.

@MichaelT wonder what happens if he changes the order in the first benchmark

user55340

@JohanLarsson I skipped to the end - that was in an earlier part of the blog post.

@amon big-o is mostly relevant for processing collections right?

user55340

> What happened? I have to assume that some part of the hotspot kicked in after some iterations and that the second example was only faster because it was second, so I ran the examples longer.

@JohanLarsson big-O applies to any algorithm for which "input size" can be defined, it's not in any way tied to "collections"

like factoring a prime number, multiplying matrices, etc

8:28 PM

I put matrices in the collections box

you could define numbers as collections of bits too (that's how "size" is determined for things like prime factorization), but that's getting pretty far away from the normal use of the term "collections"

yeah, I don't put them primes in the box :)

@JohanLarsson Big O ist mostly relevant when discussion how algorithms scale with a problem size, and collections are one example where we have many things, and need an intuition how much longer an operation will take when we double the number of elements. However, Big-O is just a simplification of running time. Instead of “foo(x) will take O(n²)”, I can make a more precise statement and say “foo(x) will take T(n) = a·n² +b·n + c” where the various parameters could be fitted via test data.

Big-O is just relevant for algorithms period. It has two dimensions; time and space. Big-O will bend time and space to boldly go where no man has gone before. Wait, I think I'm getting something mixed up here..

primes is a nice counter example

8:32 PM

@Ixrec prime factorization is about defining the space of the problem domain by constraints; that's how you determine the input size for such things. It is a collection- just not an instantiated one, but rather a collection of possibilities you determine based on known problem space constraints.

it's true that big-O is often about collections because that's where most of our common problems with a variable "input size" show up

@JimmyHoffa like I said, you could interpret it that way, but that's a completely different use of the word "collection" from the more common one that means arrays, vectors, hashtables, trees, maps, etc

the n in a big-O calculation can be either a count of actual items, or a count of possibilities. The n is fungible in Big-O.

it's like how you could interpret "agile" as meaning "let programmers get things done faster by not wasting time on specs", but everyone knows that's not how you're supposed to use the word

@Ixrec everyone is a strong (read: inaccurate by many folds) term to use in referencing the size of the population who see's it as such.

I'm trying to cut off an uninteresting tangent!

8:35 PM

@Ixrec I'm trying to Sunday!

not gonna ask the baseline q

yeah, you already got the answer for that: there is no such thing because performance is too complicated for that

another possible counterexample is graph algorithms; I wouldn't call them "collections" personally since you'd never use them just to store things, only to represent relationships between things

“what is this annoying reality, and why does it always get in the way?”

"How dare the language implementers do all my optimization for me! What am I getting paid for!?"

“I want a Hippopotamous for Christmas” - amonymous

8:39 PM

“Unicode is awesome” – ~~anonymous~~ – amon

well, it is

user55340

@JohanLarsson GCD has various big Os depending on the algorithm used... some use bits, some use digits.

@JimmyHoffa first, you're thinking about Rhinos, and second, it was an Unicorn.

Gayla Peevey I want a hippopotamus for Christmas

user55340

(btw, edits / comments: github.com/the-whiteboard/the-whiteboard.github.io/blob/master/… )

8:43 PM

^-- actual video of @amon. True Facts.

@JimmyHoffa This video is not available in Germany, because […] GEMA (German RIAA equivalent) – sigh, time for the unblocker.

@MichaelT gcd?

@JohanLarsson greatest common denominator

<-- read the intro to TAOCP before dying of dryness

ty

user55340

Euclidean algorithm

In mathematics, the Euclidean algorithm, or Euclid's algorithm, is an efficient method for computing the greatest common divisor (GCD) of two numbers, the largest number that divides both of them without leaving a remainder. It is named after the ancient Greek mathematician Euclid, who first described it in Euclid's Elements (c. 300 BC). It is an example of an algorithm, a step-by-step procedure for performing a calculation according to well-defined rules, and is one of the oldest algorithms in common use. It can be used to reduce fractions to their simplest form, and is a part of many other number...

8:53 PM

@MichaelT I prefer the Eleusinian Algorithm; it's where you gather a group to ceremonially drink wine and dance naked until the denominators denominate themselves.

user55340

@JimmyHoffa I must remind you of that .gif from the other day...

@MichaelT @amon's crazy, no doubt about it. At least it's not a rhino. There's just no running from something with such distance coverage.

Nancy Reagan has died.

9:15 PM

Jimmy, I got some bad news

user55340

@Ampt was talking with my father about that last night.

@Ampt shutup shutup shutup! lalalalaicanthearyoulalalala

Luckily there's no shortage of good Irish coming up. I'm not so picky as to demand Scotch frankly; it is a happy place, but a bottle of Bushmills single malt or Jameson black makes me quite happy as well.

user55340

@Ampt this problem has been brewing for some time.

Peter Tòmas Scott

9:31 PM

@Ampt Well I live in Scotland and generally drink 15-18 year-old malts; of which there are plenty, so it doesn't affect me. I expect that it is much harder to acquire good aged drams abroad.

@MichaelT Hate to be pedantic but... brewhaus.com/…

oh dear, it's one of those meta questions

user55340

@PeterTòmasScott can't you just accept a bad pun.

Peter Tòmas Scott

I could, but pedantic baiting has its merits.

Gah, you beat me to it

Wasn't there a guy selling whiskey treated to taste like aged stuff, and then he converted to tequila or something?

user55340

Could you identify which of those 12 of 15 questions were on topic for the site as described in the tour and help center and elaborated in the FAQ? — MichaelT 14 secs ago

Peter Tòmas Scott

9:34 PM

Improper grammar has been my downfall.

What's the time-limit for deleting posts on this? I thought it was 5 minutes, but evidently not.

9:52 PM

#36 — gnat 1 min ago

Peter Tòmas Scott

@MichaelT @Ixrec 'm not exactly a regular round here, or on SO for that matter, so please feel free to correct me. meta.programmers.stackexchange.com/questions/7906/…

user55340

@PeterTòmasScott Its not wrong. And I'll pull up one of the starred posts...

user55340

3 hours ago, by Ixrec

my personal belief is that most novice programmers drastically underestimate how much "the correct solution" depends on your specific use case, and seem to think there's a simple list of good ways to do X they need to pick from which all the good coders have memorized

@PeterTòmasScott while I disagree with you, opinions on this subject vary so wildly I think it would be presumptuous of me to claim that you're in any way wrong; I can only claim that my down/close/delete votes are never based on the OP's lack of experience, at most their lack of effort

user55340

Its that it takes experience to understand why some things are not answerable.

user55340

9:54 PM

And it also takes experience to be able to fully ask a question with sufficient information.

user55340

And then there's the "copy and paste some code because you can't ask it on SO for some reason" that we get quite often.

which is a comparatively new thing, even I remember a time we weren't getting too many of those

user55340

@Ixrec tighter controls on SO leads to less exposure leads to more people trying to find other sites to do it... and we're one of those.

oh yes, I've heard that argument many times, and it probably is correct

user55340

There are dozens of ways to answer a "how do I do" question. That spelling question.. there are numerous ways of doing it. Just try something first.

9:57 PM

but the fundamental problem that we get tons of off-topic/unanswerable questions has been the case long before that

user55340

And then there's that odd bit from Math that keeps showing up not infrequently.

Peter Tòmas Scott

A beginner can usually gain enough knowledge on their own to ask a reasonably focused question, perhaps 'lazy beginner' would be a better term. Googling 'how do I do X?' will typically give you a few starting points to try. And looking at the homepage, this does seem to be a theme see programmers.stackexchange.com/questions/311935/…, programmers.stackexchange.com/questions/311931/…

However, I'm not sure if those people good gain enough experience googling abut their issue to then ask suitable questions for this site.

user55340

-4

Q: How to get number of network interfaces in Java

How can I get the count of network interfaces? I don't need to list them, I just need the count, so if the program detects there are more than 2 network interfaces it can do something different.

java interfaces networking

user55340

A google search for listing network interface java brings up Listing Network Interface Addresses as one of the Java Tutorials from Oracle and How to enumerate IP addresses of all enabled NIC cards from Java? on Stack Overflow. — MichaelT yesterday

user55340

They need only type in the title of their question.

Peter Tòmas Scott

10:01 PM

Yes, quite.

it's more that if you bothered to use google at all you'd never need to ask any variation of the question, on any site

there's no real point contemplating how a question like might be "saved"

Peter Tòmas Scott

Yes, and I can say from personal experience I have started to write questions and found that I've solved them before posting through research and thinking about how to ask my question more often than I have posted questions (this applies to both Programmers SE and SO)

it's possible some of these beginners need to hear "no, nobody has all that stuff memorized, that's why documentation exists"

user55340

@Ixrec want to write about that?

nah, I have translations to do and Doctor Who to watch

user55340

10:04 PM

@Ixrec Too bad the site is off line... gallifreyan-9000.rhcloud.com

that's fine, I'm doing all my complaining over here

(I happen to be deliberately watching a few of the series' low points tonight, so it actually is complaining this time)

user55340

(thats a site that gives translations in gallifreyan of English)

was that the fanmade language?

user55340

user image

oddly enough I can remember seeing "Old High Gallifreyan" a few times on screen, but no modern Gallifreyan until New Who

</nerd>

user55340

10:09 PM

(you said you were translating Dr. Who... right?)

lol

user55340

11:08 PM

0

A: Is there something wrong with this site's rules for posting questions?

Ok, I've done this before... lets look at the most recent closed and down voted questions and look to examine if they are being closed because of overly strict rules, aggressive enforcement or if they just aren't properly answerable. Running .NET desktop application on Linux (-2, 1 close vote) ...

user55340

sigh

Peter Tòmas Scott

"Despite all the objective evidence of the failure of this community"

Wow... they're not pulling punches

they also forgot the objective evidence that NPR failed and the current Programmers.SE is a massive improvement over NPR

user55340

@PeterTòmasScott Look inward.

most of us are just sick of having this debate with people only interested in yelling at us for being harsh on terrible questions, even though we're no harsher than any other site on the network and the real problem is that we get way more terrible questions (as a percentage of total questions) than most of the others

even the example that person selected for one we should be less harsh on is blatantly off-topic and poorly written, so their argument is "objectively" wrong

which is why I'm venting here instead of attempting to engage over there; it's just a big waste of everyone's time

Peter Tòmas Scott

11:15 PM

I dislike hyperbole like this. I'm not established in this community but even I'm not interested in hearing someone out if they are so willing to make untrue exaggerations.

exactly

user55340

I want to know if people think its 3%, 10%, or 50% that we're off on.

Peter Tòmas Scott

It is not a "failed community" because it has numerous active users that do provide useful answers to people.

I've seen failed internet communities and they fail because they lack any engagement from a core group of users; that is not the case here.

we are definitely an anomaly, but pretty much everyone who's been active on this site for a while agrees that it's incoming question qualiy, and if there is a problem with our reaction it's that we aren't nuking the bad questions quickly enough

Peter Tòmas Scott

As to whether more questions should be accepted and answered instead of closed, that comes down to the site's stated aims. I think it is better to phrase the discussion as a 'what is the objective of Programmers.SE?' rather than 'how can we accept more questions?'

11:18 PM

it's only people who are very new, or experienced with other SE sites then come here and look at stats without actually participating that conclude it must be our fault

at least so far

</more useless venting>

@PeterTòmasScott agreed, and this leads back to why we don't have a site name which reflects whatever that objective is (we can all agree it's not "answering questions about programmers")

user55340

@Ixrec get to 20k rep... they you can join in the speedy deletes and get all those questions off the front page so no one can see them.

Peter Tòmas Scott

Saying "This site has more close votes than normal, therefore this site is causing those close votes" is the same as saying "I observe a correlation between X and Y therefore X caused Y".

on the plus side, once in a while someone posts a meta Q about a specific question they think shouldn't have been closed, and often that actually does get it reopened

user55340

btw, I'm amused that the example question in the comments there, was closed by @Ixrec and a mod (software recommendation).

@PeterTòmasScott "This site closes 70% of its questions when we reduce the close threshhold from five votes to three votes, therefore we can't trust the user base with a reduced threshhold."

@MichaelT lol, nice, I didn't even check

user55340

11:22 PM

@Ixrec We close 74% of them based on the most recent 50... even without the 3 votes.

inorite

yet more reasons that whole thing made no sense to me

user55340

Closing with 3 rather than 5 is a matter of how fast we can provide feedback to the OP that the question is off topic.

user55340

(I'm out of close votes because I tried working the first post queue for a bit)

oddly enough, I think we've improved at quickly closing/downvoting the obviously bad questions, at least compared to when I started paying attention to such things

I've been too busy/apathetic to spend my votes lately... >_>

user55340

@WorldEngineer thank you for all those close votes I'm seeing.

11:25 PM

/hides

user55340

Granted, its only two so far... but that's still half a dozen fewer community votes that need to be spent.

those last few questions might be good

user55340

@Ixrec that they do.

Therefore we do not hate everything. Therefore Lightness/Bean/Barry/Supergirl is incorrect.

I should get back to doing something useful

user55340

(FWIW, the OP of that meta question has 1.6k rep on SO - just hides all network accounts)

user55340

11:29 PM

1.7k 1 9 33

polymer polymer-1.0 javascript paper-elements firebase google-apps-script data-binding

user55340

Amusingly, the most recent question asked on SO has this comment:

user55340

This question is either too broad, opinion based or requires discussion and so is off-topic for Stack Overflow. If you have a specific, answerable, programming issue, please provide full details. — Paulie_D 12 hours ago

must resist urge to pursue ad hominem attacks

we do not need to resort to logical fallacies to justify the status quo

we shall defeat the Hello Kitty with honor!

user55340

@Mowzer the questions that we don't close are a good fit. However, many people seem to ask questions that are off topic here for some reason or another. I have no idea why we keep getting so many debug questions when stack overflow is right over there. There are numerous questions that aren't closed - that are fit only here (Stack Overflow doesn't want them because they don't have code). It isn't easy to ask those questions - but they fit well here. If you wish to contend that this site shouldn't be, you should try asking on MSO to make these questions within the scope there. — MichaelT 12 mins ago

ok seriously back to something useful now

user55340

11:32 PM

I've yet to see anyone take me up on trying to get MSO to agree that we shouldn't be.

yup

Peter Tòmas Scott

0

A: Is there something wrong with this site's rules for posting questions?

Redirecting the Question The sites rules for posting questions support and direct the site towards its stated aims I think don't there can be disagreement that the current posting rules accurately direct questions towards this site's intent; so the question ceases to be 'are the posting rules i...

@PeterTòmasScott +1

Peter Tòmas Scott

Ta

user55340

btw, @PeterTòmasScott for all those off topic bits of opinion and advice that don't fit here... you've read the blog I keep trying to push people to write on? the-whiteboard.github.io

Peter Tòmas Scott

11:37 PM

@MichaelT I have not; but I'll give it a look later this evening, thanks.

« first day (2011 days earlier) ← previous day next day → last day (2978 days later) »

Transcript for

Mar '166

The Whiteboard

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

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

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile