The Whiteboard - 2016-03-08 (page 4 of 6)

« first day (2013 days earlier) ← previous day next day → last day (2991 days later) »

user41796

8:00 PM

@enderland Part of the challenge is that I think there is an expectation of a silver bullet to resolve the flow of crap questions.

@GlenH7 I more just want to not feel bad downvoting/closing them and people who click through a "yes, this is not an implementation question" make it a lot easier to not worry about... I think @MichaelT said that 20% of recent closed questions are those

it's definitely a lot

those are also by far the most objective and unfixable closures, the "too broad" and POB ones could sometimes be argued both ways

right, I'd rather fix 15% (if say this would stop 75% of the 20%) and then move on to bigger things :)

total speculation about 15%... ha

@GlenH7 you didn't know we have a silver bullet? Oh dear, how long you've been here and you never noticed @gnat; how strange...

user41796

I do kind of wonder where the semi-recent influx of implementation questions has come from. I don't think that the prevailing theory of q-banned on SO holds anymore.

user55340

8:04 PM

user image

user41796

@JimmyHoffa Hey now. I like our first line response AI bot!

the prevailing theory is that SO triage kicked in

user55340

30 day stats. 18% of all closed. None reopened.

user41796

misread that, sorry

user55340

2 day is 23%.

8:06 PM

that's probably non-deleted questions too?

@GlenH7 I've been wondering this for months now; the q-banned on SO theory seems a stretch given the volume, but the timeline appears to match

user55340

have to dig up MSE post. I believe all.

isn't the volume miniscule compared to what SO triage nukes?

though I admit the only reason I believe the theory is that there are no competing theories

user41796

@Ixrec Sure, the volume we see is tiny compared to almost anything over on SO

user41796

But ... we have comparatively fewer resources for handling those issues

8:10 PM

We have gnat.

but that's about it

user41796

1/2 the close votes, for example. And definitely fewer active reviewers. Not sure about proportions there though. Apathy on SO may actually work in our favor from a proportional perspective

yeah apathy on SO is high

user41796

I have had a lot of work induced apathy here.

user41796

Haven't really touched the review queues as of late

my apathy has been pretty high lately

me neither, though that has more to do with irl obligations

8:12 PM

ok. let's play name that method... currently it's "wait_for_container_to_stop_and_get_out" which is so horrible it's not even funny

VTC as needs more context

@Ixrec I have one. The internet is a bountiful place; and the tree of bytes saw us wriggling and writhing on the floor of her cosmos, and said "Let there be moar!" and upon such pronouncement unicorns on wings of lightest rainbow did set forth with strident fervor to strike upon us with questions of debugging; and whoa be unto our plight said the tree of bytes, for we did set this upon ourselves in riposte with inertia and quality that she sensed our tests needed increasing

@Ixrec that's why it's so long :( haha

user41796

Thumb_twiddling

You wanted a competing theory. I feel my theory vindicates itself; don't you? We are in agreement then. All hail The Tree of Bytes.

8:14 PM

leave_after_containment?

user41796

@JimmyHoffa -1 for failing to mention the superiority of emacs.

I'm wanting a method to get stdout/stderr on a docker container, but waiting for it to stop first

get_container_output_after_stopping

user41796

break_free

some containers automatically stop and I want to include a looping check for that

waitUntilStopped()
getOutput()

cheating?

8:15 PM

hrmm. maybe, that's not a bad way to do it actually :)

BarryTheHatchet

otherwise a blocking getOutput() sounds fine to me

@GlenH7 you tempt The Tree of Bytes' wrath with such candor, brother.

as we all know, the SRP says "if your method name contains the word 'and', break it into two methods"

@BarryTheHatchet well the problem is that sometimes it might be blocking and I need a way to have the non-blocking too

BarryTheHatchet

@enderland but this function won't be handling that, will it?

8:17 PM

hence "word", not "string" =)

BarryTheHatchet

@enderland If it is, then fine, it just returns immediately! (or throws an exception if "output" is meaningless for non-blocking calls)

the SRP spec uses \b's in its regex

BarryTheHatchet

i.e. take the select model.

@BarryTheHatchet I might want to do getOutput on a container that is not stopped, though

BarryTheHatchet

@enderland right, so it blocks until the container is stopped and output is available

not seeing the problem

8:18 PM

except if I want to get output on a container that is not and will not stopped

BarryTheHatchet

dude what

25 secs ago, by enderland

except if I want to get output on a container that is not and will not stopped

;-)

BarryTheHatchet

28 secs ago, by BarryTheHatchet

dude what

-.-

hah

function waitUntilStopped() {
    if(nonBlockingContainer) {
        return container.promisifiedStop();
    } else {
        container.stop();
        return Promise.resolve(container);
    }
}

is what I'm hearing

BarryTheHatchet

8:19 PM

yes

you don't use the term promisified in your code. I know it. The Tree of Bytes wouldn't stand for it.

@Ixrec the problem is that container.stop() doesn't immediately make .resolve(container) work, I need to wait until I can also confirm it is stopped as stop() is executed asyncronously

@JimmyHoffa I wanted to make it more obvious for the people here who do not use Javascript promises on a daily basis. Though Bluebird does have a "promisify" method.

@enderland what do you think promisifiedStop() does?

@enderland have you tried harder lately?

that main site question looks eminently answerable

why do we have a philosophy tag!?

8:22 PM

@Ixrec ah yes, this would be one of the reasons Bluebird has shown to shy from the fiber of The Tree of Bytes; and has gone to the dark side

@JimmyHoffa you're supposed to blame Node.js for that

user41796

@JimmyHoffa we promise we would never promissify anything in our codez

(promisify() is for converting Node-style callbacks to promises)

@Ixrec Node.JS uses a consistent CPS style; The Tree of Bytes lends great support to such methods

Node.JS' CPS is easily made into a monad

@Ixrec I think I'm going to make this a bit easier and make my "stop" method wait until it verifies it is stopped... and then solve this entire problem :)

8:24 PM

container.noReallyIMeanItJustStopPlease();

@JimmyHoffa great, here we go again.

@amon :D

@Ixrec lol. python makes that look way less cool with underscore naming convention ;)

user41796

@amon Just click on his name, hover over the "mute" section, and then be done with it. Things are much quieter that way.

blame @Ixrec for straying from the fiber optics of The Tree of Bytes. His philistine ways are disturbing indeed.

8:26 PM

@JimmyHoffa You haven't even seen our Fortran.

How do I C++?

You increment the C.

Btw, the xkcd-1305 seems to be kaputt :(

specifically post-increment, so that you think it's still C at first but then it turns out not to be

that "why are software projects never done", I know there is the essence of a good question in there but I am not sure what edits to make to bring it out

8:29 PM

the good question is a dupe anyway

@aml90 so? How old is your cell phone? 10 years old? It's disposable. Presumably you have "upgraded" your cell phone many times over the years. A dam is not disposable and so it receives physical updates. Likewise, most software is not disposable and thus receives more updates. — enderland 2 mins ago

that's the answer, basically

I've definitely seen questions about that in the >3yo pile

@amon wat happen?

@enderland: I don't think its a matter of being disposable or not

@MichaelT this is your chance to get @whatsisname to write a blog post

8:30 PM

@whatsisname the whole question is based on so many false premises

@JimmyHoffa there's no XKCD when I do the clickity-click, because the info tab is nowhere to be seen

it's more that one is information and the other is "stuff"

Additionally, phone manufacturers do not throw away their old designs and start from scratch - they refine existing designs. So in some sense they are doing exactly what software does. — enderland 39 secs ago

Also, software can be copied. So unlike all other kinds of engineering, there's often no point writing a piece of software unless no one else has ever written it before. Dams, bridges, cars, etc. have been built many times before. — Ixrec 6 mins ago

that also applies to it

8:31 PM

yeah, your second quote and ixrecs is more onto it

@amon I know... but.. why? It were there..now poof. The Tree of Bytes will not be pleased.

But software projects are done.

When they run out of time and money.

BarryTheHatchet

Arguably that's cancelled not done

It could be a planned thing.

Here's enough money for 5 people over 2 years and a set of requirements. Go.

In 2 years, that project will be done.

BarryTheHatchet

Will all issues have been resolved? Unlikely.

That's what's being asked about.

8:32 PM

"in the sense that the developer has stopped committing code altogether" - Java 6.

It is end-of-life. It receives no more patches. No more fixes. It's final. Done. Complete.

Our flat has had "unresolved issues" the entire time we've been in it. Not unique to software.

The question doesn't mention "unresolved issues".

BarryTheHatchet

@ThomasOwens So, cancelled.

I'm half-quoting ~~Barry~~ Supergirl

@BarryTheHatchet No. It was planned.

8:33 PM

@Ixrec hehe that's part of my answer (well, homes)

the question definitely is based upon bogus premises but they are common misunderstandings

BarryTheHatchet

@ThomasOwens It doesn't need to use the exact same words in the same order for us to determine what the OP means by "done".

user55340

@whatsisname ... And you want to write a bit about that off site?

next HNQ: What was this question about software being done actually trying to ask?

BarryTheHatchet

@Ixrec cue 40 downvotes

8:34 PM

not really

as tempting as it is

anyone remember if downvotes increase or decrease hotness?

BarryTheHatchet

dunno

user41796

decrease

BarryTheHatchet

I can tell you it's neutral for lightness though

user41796

kinda

user41796

8:35 PM

the HNQ looks at net Q score, num answers, and net score on answers

user41796

do down voting does decrease the HNQ score

user41796

Only the top voted 10 answers count to the question though

@JimmyHoffa Hmm, I can see the edit history but can't reanimate the tag. We need some question that legitimately uses this tag. “Hai I write social network which language should I use to make bestest comunity plz?”

user41796

2 or 3 voters can safely knock a Q out of HNQ with (in)judicious voting

let's clone gnat a few times, ban his copies from writing comments, and then we'll be golden

BarryTheHatchet

8:37 PM

where's he been today anyway

all those meta posts about the site scope and gnat a peep

I never see him in chat anymore

user41796

@Ixrec he's been in less frequently

user55340

@Ixrec the chat community is changing.

I feel like that started around the time people started pointing out that gnat is uniquely negative among the "cabal" here

what cabal

user41796

8:38 PM

He also started becoming more involved in other SE sites as well

@amon hmm alright. What never-going-to-be-deleted Q do we have that merits being the butt of a joke, @MichaelT @GlenH7 @ThomasOwens ? Ideas?

the scare quotes cabal

BarryTheHatchet

i.e. around the time he became uniquely negative

Peter Tòmas Scott

I don't think changing the scope s good; it is just in the list for the sake of completeness. One thing I would like to discuss that I am unsure of is objective vs subjective. SO is clearly for objective questions (debugging, how do I...) but Programmers.SE is less clear cut; questions about architecture and high-level design are much more opinionated in general. So... is Programmers.SE for objective or subjective questions? – Peter Tòmas Scott 1 min ago edit

BarryTheHatchet

@PeterTòmasScott All SE sites are for objective questions. It's a Q&A network.

8:39 PM

that's the wrong question to ask, it's the sort of mindset that encourages "this isnt' good enough for SO, migrate to programmers!"

user55340

@PeterTòmasScott objective / subjective is the wrong way to look at it

BarryTheHatchet

You have a point though that design things are generally pretty subjective in nature. So wtf.

it's better to compare them as implementation issues versus design issues

BarryTheHatchet

I think that's part of the problem.

user41796

Good Subjective, Bad Subjective

Robert Cartaino on September 29, 2010

Stack Exchange is about questions with objective, factual answers. We've been crystal clear about this for as long as I can remember, even back to the earliest, pre-beta days of Stack Overflow. It's right there in the standard Stack Exchange FAQ:

BarryTheHatchet

8:40 PM

Things get closed as "subjective" or "opinion-based" when all software design is precisely that.

questions here need to have enough detail that answers can be evaluated, judged as better or worse than other answers, which means at least some degree of objectivity

while design issues tend to be more subjective than implementation issues, that is a side effect of the difference between us and SO, not the actual difference

user55340

When this is the reception that one gets, it is a way to drive someone out of s community b

user55340

0

A: Great Expectations... or, at least, user expectations

The problem isn't that "people disagree with the site's enforcement; i.e. thinking that valid questions are being downvoted," although there is definitely an element of truth to that. The problem is that it's being done in a hostile manner. The author of the linked post was completely accurate ...

Peter Tòmas Scott

I keep seeing the statement that questions here must be objective, but I think there needs to be a recognition that we are not talking about objectivity; we're talking about utility.

@amon Resolved. Care to push the edit history back to current?

8:43 PM

I agree that talking about the words "subjective" and "objective" is a waste of time

good subjective, bad subjective is really hard to grok

user55340

@PeterTòmasScott answerable without having to go back and forth a dozen times to refine it.

this is one of the reasons I keep using this word "answerable"

user55340

It is just as difficult to get a good question as it is to get a good requirement from the business.

@enderland it's better to work on the cross-section of the two; I call this, bgoaod sobjactive

8:44 PM

is this what functional programming does to you?? :o

@enderland The Tree of Bytes will see you punished for your insulin!

at least twice I have suggested a question in chat that someone else went and posted then earned tons of rep for; I am okay with this because in both cases they wrote it up far better than I would've been able to

asking a good question is hard

@JimmyHoffa I prefer the Tree of Nibbles

easier to prune

@Ixrec ah, but it has grown much since then.. the earlier version is just a silly memory to Her Most Factorialness The Tree of Bytes.

wait, how can you have a tree of bytes anyway? a byte isn't big enough to store a pointer on any computer I know of, much less two pointers

@Ixrec you'd do well not to question The Tree of Bytes in this domain; the internet grew from her and she'll quash this site with a mere snap of her bits if she so decides...

8:48 PM

I'll assume it's Huffman encoded

BarryTheHatchet

@Ixrec sure it is

Peter Tòmas Scott

Viziionary is now expressing glee that someone else is gunning for gnat.

BarryTheHatchet

for example, the null pointer can be stored in a single bit!!

I am now legitimately curious if the C++ standard allows the null pointer to be represented as one bit

user55340

@PeterTòmasScott they've tag teamed him before.

8:49 PM

that seems like exactly the nonsense it would technically permit

BarryTheHatchet

(assuming a 2-byte memory space)

@PeterTòmasScott where?

BarryTheHatchet

@Ixrec heh

I CANT TURN AWAY

BarryTheHatchet

@Ixrec no, a byte is the smallest unit of storage

Peter Tòmas Scott

8:50 PM

1

A: Great Expectations... or, at least, user expectations

The problem isn't that "people disagree with the site's enforcement; i.e. thinking that valid questions are being downvoted," although there is definitely an element of truth to that. The problem is that it's being done in a hostile manner. The author of the linked post was completely accurate ...

Here

@BarryTheHatchet a word is the smallest unit of speakage. A baby is the smallest unit of seepage.

BarryTheHatchet

@PeterTòmasScott lol nice

@JimmyHoffa what are you hoffing on about now

@BarryTheHatchet if a bit isn't a unit of storage, then I don't want to be persistent

BarryTheHatchet

@JimmyHoffa cya

Feb 2 at 17:51, by Jimmy Hoffa

Jan 29 at 21:20, by Ampt

Jan 22 at 15:20, by Kit Z. Fox

<-- helping

8:54 PM

WHY DID YOU CLOSE THAT?

GAH. I wrote an epic answer.

I need to put it somewhere.

BarryTheHatchet

heh

I'm going to turn it into a blog post.

@ThomasOwens link?

user55340

@ThomasOwens blog!

link

user55340

8:55 PM

I happen to know of one.

gist.github.com/ThomasOwens/39f2563a9e27a760b55a

Can you guys see that?

BarryTheHatchet

what question?

user55340

And if you edit the JavaScript you can post to closed questions if started before its closed.

-3

Q: Why are software projects never done?

Unlike other products that are engineered, for instance, a hydro electric dam or a bicycle, why are software products never fully done, at any point? I've never encountered a software project that was supposedly "complete" - in the sense that the developer has stopped committing code altogether a...

engineering philosophy

BarryTheHatchet

closing people's questions so you can write answers instead on your cabal blog is tantamount to fraud, you realise

8:55 PM

@ThomasOwens put it on TWB blog

@JimmyHoffa TWB?

github.com/the-whiteboard/the-whiteboard.github.io

I'm going to clean it up and post it to my own blog first.

@BarryTheHatchet it is? you can already write answers on the blog and the main site

user55340

@ThomasOwens the-whiteboard.github.io

BarryTheHatchet

8:56 PM

@Ixrec but now they're seizing the opportunity to get new blog content, instead of re-opening what has obviously transpired to be an answerable question, and answering it there

see I knew it was all a conspiracy!

user55340

Too broad. Read the chaos reports.

Thomas can always edit the question

I'd appreciate comments on my answer.

If you guys can see it. Otherwise, I can make it public.

that's why I didn't even bother writing an answer

I suspected it would get closed before I'd finish it

user55340

projectsmart.co.uk/white-papers/chaos-report.pdf

user55340

8:58 PM

If you modify the Dom after its closed to renenable this button, it posts fine.

for the record, serious voice now, does anyone other than Supergirl actually disagree with that question being closed as too broad? (no judgement on whether edits can get it reopened)

BarryTheHatchet

I don't even know what the question is

did someone link to it?

@Ixrec The projects end question? It's not too broad. But it may be primarily opinion based.

user41796

@ThomasOwens I'd support you rolling the close status on that "software done" question in order to post your answer

@BarryTheHatchet twice I think

BarryTheHatchet

8:59 PM

oh, yes, Thomas did

Had no idea you were referring to the "done" question

user41796

ATM, I don't see anything redeeming in the post and will be voting to delete. If your answer is that good, then I won't VTD

BarryTheHatchet

meh I'm not desperate for that to be re-opened

If someone casts a reopen vote, I'll reopen. I feel bad for mod reopening to post my own answer.

BarryTheHatchet

I wouldn't have VTC but to claim that I disagree with it being closed is a misrepresentation

« first day (2013 days earlier) ← previous day next day → last day (2991 days later) »

Transcript for

Mar '168

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