The Whiteboard - 2016-04-03

« first day (2039 days earlier) ← previous day next day → last day (2945 days later) »

12:09 AM

Hey

This will be closed on Code Review due to asking about code yet-to-be-written... I just wanted to run by you guys if it would be on-topic on Programmers?

0

Q: How specific should error testing be with cucumber/capybara?

I'm working on writing tests for a rails application using cucumber and capybara. I have a scenario for a user editing a post, and making it invalid. The scenario looks like this: Scenario: I edit a post and make the data invalid. Given I have 1 post And I view my 1st post Then I click "Ed...

ruby ruby-on-rails cucumber

12:31 AM

Wait, apparently that might be actual, executable code. Wtf, cucumber

BarryTheHatchet

12:51 AM

Yeah, it's real code

Literally the entire purpose of Cucumber

Research the fundamentals of technologies being asked about before VTC!

BTW you have a comma splice in your comment on that question.

I gotcha. This is literally the 2nd Q ever asked on CR about Cucumber, but, we rectified. I have a feeling it's not going to get answered very fast...

BarryTheHatchet

^_^

It'll be one of those that gets answered out of the blue in ten years' time

1:13 AM

On an unrelated note, just got me a 6 of Sierra Nevada "Beer Camp" Tropical IPA. Pretty good.

Doesn't kick you in the hops hard, just right, and a smooth aromatic finish /beersnob

BarryTheHatchet

nice

10 hours later…

10:51 AM

Seems like our friend @Duga has been dead for the past four days or so. She should now be up and running again.

2 hours later…

1:03 PM

> Python has [function literals] too, but unfortunately they can only be one line, so Python folks prefer to pretend anonymous functions aren't very important. This is one of the 10 or so big problems caused by Python's whitespace policy. Don't ever let 'em tell you it doesn't cause problems. It does. Maybe it's worth the trade-off; that's a personal style preference, but they should at least admit the tradeoff exists.

@AaronHall can you confirm or deny this?

It's technically true that Python's lambdas can only contain an expression, rather than a statement. The workaround is simple: def a named function in advance and use that, rather than using a lambda. However, this does rule out using higher order functions as syntactically simple abstractions, as is commonly done in Ruby or other languages.

your named-function workaround seems to confirm the "so Python folks prefer to pretend anonymous functions aren't very important" sub-assertion

For me, the more annoying part about nested functions in Python is that you can't easily assign to an outside variable (you need to declare it as nonlocal first).

you mean they aren't automagically closures like they are in Javascript?

The point is, anonymous functions (lambdas) don't give us any expressive power, but they are convenient syntax.

Python has nested functions that are true closures just as in JS. However, Python has no variable declarations and therefore a simple scoping approach: assigning to a variable refers to a variable within the current function. If we want to write to an outside variable, we have to declare it as nonlocal first. Reading from outer variables works just as in JS or other languages.

1:13 PM

while I'm not huge on anon functions myself, that's not a very useful argument since every programming language that isn't machine language is "just convenient syntax"

@amon ooooh, that's how Python scoping works...I think now I finally know why I prefer having to explicitly declare all my variables

@Ixrec yes. But if I had designed Python with the same design goals, I'd have arrived at the same solution. For example, anonymous functions are a serious impediment to debuggability. It's false that you can't have indentation-based syntax and lambdas with multiple statements in the same language (see Haskell), but it does make the syntax a lot more complicated: statements inside expressions where indentation is otherwise irrelevant is confusing.

I've never had the anon functions harming debuggability thing personally, maybe I haven't had enough anon functions nested in each other or something

but "somefile.js: line 23 <unnamed function>" is still pretty clear

@Ixrec have you written highly callback-based JS? Have you read a stack trace where 70% of functions is anonymous and just gives you a line number? That's far less pleasant than requiring a name for each function. Then again, I doubt Guido thought about that when designing this part of the syntax.

1) we never have that many callbacks in one place because promises are awesome
2) we never have that many functions inlined in one function because, duh, we break stuff out into named private methods or named utility functions at some point (where the name is chosen for readability regardless of stack traces)

myIfThenElse(
  cond=condition,
  ifTrue=lambda:
    statement
    statement
  , # where do I put the comma?
  ifFalse=lambda:
    statement
    statement
)

1:26 PM

I'm trying to imagine a situation where I'd have a function call with arguments like that but it's not coming to mind

except when it would look something like myIfThenElse(cond, myIfThenElse(cond, literal, literal), myIfThenElse(cond, literal, literal))

@Ixrec you already mentioned promises.

oh, we don't do the then(func, func) thing, it's always then(func).catch(func)

the latter is generally preferable anyway since then the error func also catches errors from the first func

Yes, regardless of language: In nearly all cases, a well-designed function would only take a single callback.

func(error, result) and all that

@Ixrec There's not much of a tradeoff- a lambda expression returns an object that is basically the same function type as a function. Most use-cases for functions would benefit from the full function definition's naming and docstring - and clarity for allowing logic to resolve in more than one expression. lambdas can be very useful in limited circumstances (allowing brevity where verbosity would be redundant and wasteful.)

1:32 PM

I would have to disagree with the idea that "most use-cases for functions benefit from naming"

see promises and map/fold/reduce/etc

Good names can limit the need for docs and comments

Extreme programming

I love long explicit names on the functions that actually need names, no convincing required there

In Python, map and filter (with their lambdas) are made redundant by comprehensions and generator expressions

but when the only correct name to give a function is theCallbackThatMethodFooPassesToMap(), that's a complete waste of the next guy's time when he reads the code

it's already obvious because he just read "map(function"

@AaronHall aaah, that is an interesting point I hadn't thought of

I definitely lack the Python experience needed to agree or disagree there but it sounds plausible

note to self: next time I write some Python, learn how to docstring

@Ixrec youtu.be/iGfggZqXmB0?t=3m47s

1:39 PM

<obligatory complaint about how long my list of things to read/watch is>

but that one I will watch in a minute

:)

It's not the best talk ever, but it's probably my best talk.

how wrong would it be to compare LINQ to list comprehensions?

? I dunno. Oh yeah, closures aren't necessarily immutable - nonlocal (Python 3) can modify them

I need a small checklist of errata for that talk.

what does it even mean for a closure to be mutable or immutable?

oh I got to the part you're talking about...that scoping rule is a real mind-bender

my gut is insisting that changing return x; to x = 1; should not change what variable x refers to

Well in Python 2, you can't do this, you'd have to stick the number as an attribute on the inner function or in a mutable data structure:

>>> def outer(n):
...     def inner():
...         nonlocal n
...         n += 1
...         return n
...     return inner
...
>>> counter = outer(0)
>>> counter()
1
>>> counter()
2

Python 2:

>>> def outer(n):
...     def inner():
...         inner.n += 1
...         return inner.n
...     inner.n = n
...     return inner
...
>>> counter = outer(0)
>>> counter()
1
>>> counter()
2

1:58 PM

...that hurts my head

Python is crazy dynamic. It might look reasonable, but I'll stick with Perl for now for all my scripting needs.

I think it makes sense when you consider that outer generates instances of inner - which needs a way to store and mutate data.

that's not dynamic, that's something else, but I can't put my finger on what

Object oriented? :)

is there any reason to actually design code like that instead of a more sensible approach?

2:02 PM

the part that weirds me out about it is definitely not an object-oriented thing

The dynamicism I take offense at in this case is the ability to externally and dynamically add new member fields to an object (here, the func object inner)

how about "explicitly specfying scoping rules because default Python scoping isn't good enough"?

I think that's what it is

@amon that's the part I dislike about Python the most, there is nothing stopping you from adding methods/data to objects/methods/whatever dynamically

@amon don't all "dynamic languages" allow that? or is inner.n = 1 in Python doing something different from what it would do in JS?

and does Python not have an equivalent of Object.freeze()?

well, like I said the other option is to store it in a mutable data structure

2:05 PM

Monkey patching is not inherently evil, and less so when done on a class level, rather than an instance level. In JS, I'd have less of a problem with this since it doesn't have a meaningful class–instance distinction.

>>> def outer(n):
...     n_container = [n]
...     def inner():
...         n_container[0] += 1
...         return n_container[0]
...     return inner
...
>>> counter = outer(0)
>>> counter()
1
>>> counter()
2

now I'm struggling to understand how this qualifies as monkey-patching

you're all making me feel stupid, stop doing that

So here inner closes immutably over the mutable list.

@enderland it can be totally sensible, e.g. if you were implementing an iterable class in a dynamic language your getIterator() function would look extremely similar to this

It's the inner function that is called the closure, I believe.

2:08 PM

@Ixrec monkey patching is modifying an existing class, e.g. adding new methods from outside the class definition. The inner.n is modifying an existing instance and “declaring” a new field outside of its class definition. For me, this seems analogous to monkey patching.

I thought inner was a function, not a class

class Ref(object):
def __init__(self, x):
self.get = x

I actually think I prefer to stick n on inner as an attribute, because it makes it easier to check the value from outside.

@Ixrec inner is a function, and as such an object, and as such has a class that defines which methods and properties are available.

hm, okay

to me "monkey-patching" would be adding something directly to the class such that all existing instances also get it, as if it was an actual part of the language, which I thought was only really possible in JS because of prototypal inheritance plus mutable prototypes

2:12 PM

>>> class Counter(object):
...     def __init__(self, n):
...         self.n = n
...     def __call__(self):
...         self.n += 1
...         return self.n
...
>>> counter = Counter(0)
>>> counter()
1
>>> counter()
2

that one actually feels like normal-ish code to me

Eats bacon jam on toast for an encore

class c():
    def outer(self):
        return self.innerfunc()

    def innerfunc(self):
        return 1

def foobar(self):
    return 4

def foobaz(self):
    return 10

counter = c()

print counter.outer()   # 1

c.innerfunc = foobar
c.foobaz = foobaz


print counter.outer()   # 4
print counter.foobaz()  # 10

that sort of stuff is weird to me about python. maybe it's just the dynamic language thing, but...

Well the class definition is much more extensible - and maybe, as the designer, I don't want people extending it with methods. nonlocal is probably the most restrictive way of writing it.

@enderland that example would be "the dynamic language thing", you can do that in JS just as easily (unless of course you Object.freeze() the thing first)

2:20 PM

@Ixrec it's still weird :P

it's definitely weird to actually do that sort of thing in production code

the problem I run into is that there is no "safety" against accidentally doing that and it's convienent at times

but since there's no distinction between objects and hashmaps in these languages, allowing it is somewhat inevitable

"it's convenient at times" sounds like something that would not pass code review =P

well I don't mean the method overriding but more just assigning variables to a class

also note that in JS, normally you'd define all the methods on the prototype anyway

ah right

2:22 PM

I'm sure there are usages for method overriding and I'm sure all of them would probably make people cry

I can't actually think of any tbh

pretty sure those all end up being signs of a design problem

I'm imagining some crazy convoluted "factory" thing where you assemble the object by assigning its methods :P

doing it at initialization time is totally reasonable, we're talking changing the methods after init

though even in a factory like that you'd probably just select one prototype that has the method blob you want

right

that seems... far more resaonable

in our JS code our "serious" classes (the ones that aren't merely POJSOs or "helpers" or whatever) generally use a bunch of Object.defineProperty() and Object.freeze() and the like so that most of this madness is explicitly disallowed

2:26 PM

actually that factory thing... are there any things you would have to do with OO code you couldn't do with dynamically overriding a a base object?

I assume the class implementation our proprietary framework provides does something similar since it also doesn't allow adding new properties post-instantiation

OK, after careful consideration of everyone's viewpoint (mostly mine) I have decided that BTH's new name shall be "Comma Splice"

2

shouldn't that be "Comma, Splice"?

2:46 PM

> If there's one thing I've learned over the years about type systems, it's that you have your slob type systems and your neat-freak type systems, and it comes down to personal preference. The neat freaks (Java, C#, C++, Pascal) know damn well that the slobs (Perl, Python, Ruby, JavaScript) are just as productive. Maybe even more so. Nobody really knows for sure.

> One thing is clear, though: everyone's getting stuff done, regardless of their language choice. The whole debate isn't actually about productivity at all, even though most people still think it is. It's about whether you can stand to live in a house where the bed isn't made.

interesting

BarryTheHatchet

2:57 PM

@AaronHall I don't think I can take your personal viewpoint into consideration, comma splices are evil and I'd never associate myself with them, however thank you for your suggestion, by the way good morning

3:56 PM

hmm, I'm getting more votes than I deserve on a really short answer again...possible HNQ alert?

BarryTheHatchet

4:22 PM

I notice that non-terrible answers on P.SE get upvoted a lot and quickly

Wonder whether HNQ has some kind of per-site weighting

That is a pretty good answer though. The key was to obliterate the OP's false premise that languages "do" things.

It's also excellent in comparison to the other answers, which are waffly and far more technical/scientific than is needed here

oh well, can't complain about my delete vote quota getting incremented

BarryTheHatchet

hm?

> you get five votes to delete per day, plus 1 vote per 1000 reputation points above 10k (up to a maximum of 30 votes per day).

nvm I'm nowhere near the next multiple of 1000, not sure what my brain was doing there

BarryTheHatchet

scrub

votes to delete "Tips for noobs"

BarryTheHatchet

4:31 PM

"trivial answer converted to comment" FSCK OFF

that's literally the only answer

that feature is so dumb

length and answerness are orthogonal

?

BarryTheHatchet

even if the former can be an indicator of lack of the latter

0

A: Repository UUID equivalent

No. There is no unique identification for a Git repository. Read http://stackoverflow.com/q/34874343/560648.

This. I originally made the second sentence a link to the SO Q, which looked/read much better. But it gets auto-converted to a comment by silly SE.

that seems like a case where you could (and should) summarize the SO question with a few sentences of your own

BarryTheHatchet

it's a yes/no question

though I agree the question badly needs an explanation of what the OP is looking for

BarryTheHatchet

4:34 PM

the bulk of the SO question relates to the original OP's goal

we don't know what that is in this case

so there's nothing else to quote :(

yes/no questions are not bad, the inability to justify the yes/no is bad and usually indicates a question in need of closing/downvoting rather than answering

BarryTheHatchet

I can justify the yes/no

well, aside from that it's impossible to prove a negative

but it's just a simple fact that the answer's no

though this question is sufficiently uninteresting I can't be bothered to say any more on the subject and you've already said everything that can be said until the OP clarifies what he's trying to do

BarryTheHatchet

anyway, I think yes/no questions are somewhat underrated because we focus too much on some specific person's specific goal ... nothing wrong with adding facts to the repository with which future visitors may build their knowledge and add it to the stack of facts they use to solve their own personal problem. there's too much "has someone done precisely what I need to do before? can I copy/paste their solution" and not enough building block construction

however for me what they usually betray is a distinct lack of research

agreed, a literal yes/no question (or as we'd say here, an "is it possible..." question) almost always is answerable via a quick google

BarryTheHatchet

4:38 PM

yeah I mean perfect example really as I didn't know the answer until I saw the question

answered it within a minute of seeing it ;p

DeliriousSyntax

I feel as though I spend a little too much time on this site and I'm debating on whether or not I should follow the same path as MichaelT

BarryTheHatchet

@DeliriousSyntax If you carry on the way you were yesterday, you may not have much of a say in the matter.

@AshleyNunn GTA V Deer Cam, presumably?

permanently nuking your account from orbit seems like an extreme overreaction to "I spend a little too much time on this site"

I'm pretty sure MichaelT had additional reasons

BarryTheHatchet

That's what I said two weeks ago. Nobody agreed with me. :(

bah

?

BarryTheHatchet

4:50 PM

good question

DeliriousSyntax

@Ixrec It's just an account on a little site it's really no big deal. I could live without it.

I meant like I've lurked this site for 3 years before actually creating an account. And then I've been in trouble twice on it which was also no big deal to me. I don't answer or ask questions any more. Now I really just use it as a place to hang out and blow time.

I'm not even a helpful user like MichaelT, so when/if I leave it will be as if nothing really happened except a head ache has left

Perhaps he felt disrespected.

did he delete his account?

I believe he asked to be completely removed from the network

these days he can only be found out in the untamed wilderness beyond the control of SE

5:05 PM

ouch, proper rage quit

he could have just stopped coming imo

BarryTheHatchet

1

Q: Answers in comments, revisited

I've raised this topic before more generally, with a terrible lack of success. But with this specific example I'm going to try this again as it's really starting to annoy me! Observe: The entire answer to the question has been given as a comment. Why? Answerer didn't want rep? Couldn't be bot...

@JohanLarsson that's what I said

the assumption seems to be that he was growing increasingly frustrated with the status quo and wanted to sever ties permanently rather than risk coming back on a whim and growing even more frustrated and possibly doing something stupid

BarryTheHatchet

so just don't log in

simple

that is probably what I would do

BarryTheHatchet

when you feel the need to obliterate the entire record that you were ever here, something else is involved

be it rage, or a feared spectacular lack of self-control

he also didn't say goodbye to anyone, so factor that in

5:09 PM

@BarryTheHatchet what is the question?

BarryTheHatchet

@JohanLarsson deliberately censored to protect the perp

1) one hour max per day per account on main
2) close all meta sites

@BarryTheHatchet I mean your question

answer in comment is convenient

BarryTheHatchet

sigh

where convenient = lazy, sure

yeah

BarryTheHatchet

do you not see how it completely violates the entire purpose of Stack Exchange?

to replace Experts Exchange / Yahoo Answers / forum style sites?

the system is designed to treat posts in the answer section as answers

5:11 PM

not to side with Barry necessarily, but how is a comment answer any more convenient than an answer answer?

BarryTheHatchet

all "this is an answer" metrics work on that basis

downvotes, review queues, bounties, flags, .... all work on the ANSWER SECTION

@BarryTheHatchet not really, the question can still be answered propertly no?

BarryTheHatchet

so when you post an answer in the comment section instead just because it's "convenient", you're breaking everything. and for what :(

@JohanLarsson What's the point in having an answer and then the same answer in the comment section? Just don't post the comment? Doing nothing is even more convenient!

u mad?

BarryTheHatchet

yes!! nobody seems to agree with me on this and I can't figure out why. seems like an extremely simple logical thing to me :(

pet hate of the year for sure

5:13 PM

I agree with your logic

BarryTheHatchet

and what if the answer was wrong. you can't downvote it. you can't edit it. you can't do anything you would normally do on it.

it's just simply the wrong place for things that attempt to answer the question

don't understand why it's so popular

@BarryTheHatchet you can provide a correct answer

DeliriousSyntax

Well, In my case I have no choice, but to answer in a comment because "mod" took away that privilege

BarryTheHatchet

starting to think it's a kind of FGITW & "humble brag" thing - "I know the answer and I will help you and I don't even want rep! How amazing am I"

I'd expect a proper answer to be more visible than a comment

BarryTheHatchet

5:14 PM

@JohanLarsson yes but incorrect answers must be downvoted to protect visitors

@JohanLarsson comments are higher up the page and thus seen first

@DeliriousSyntax that's called ban evasion. again, do not do it. You DO have a choice: don't write anything.

"I wrote an answer in comments because I am banned from doing it properly" is not acceptable

@BarryTheHatchet do you have a user study to back that up with?

@BarryTheHatchet just to check, you're talking about comment answers on SO rather than PSE, right? since the comments I've seen on PSE that could be called comment answers don't fit the profile implied by a lot of these arguments

BarryTheHatchet

@JohanLarsson I don't need a user study to tell you that the comments are higher on the page because it's a physical fact

writing good answers takes a lot of effort. It's easier to dump your knowledge in a comment and let someone else write the full answer if they're inclined so. Of course, this breaks the Q&A system.

BarryTheHatchet

and people tend to scroll downwards. not jump to end then scroll upwards

5:15 PM

:)

BarryTheHatchet

@Ixrec yeah I decided to post that on MSO to focus on SO. I'd argue it's a network-wide thing but I'm not able to justify/defend that so yeah just SO atm

the voting system is pretty broken

and @amon pretty much just said the only thing I have to say on this

BarryTheHatchet

@amon I'd rather you just skip the dump-knowledge-in-comment part and let someone with inclination to write a proper answer do so. it could actually be harmful to write a roughshod comment with no finesse, no caveats, explanations..... and, again, those cannot be downvoted/edited, so it's bad. the approach you describe only works if one assumes that the commenter is 100% correct. we cannot make that assumption, ever. SE's entire premise is quality control by voting. voting on ANSWERS!

not that the idea with voting is bad but the result is not very impressive

5:17 PM

@JohanLarsson there's a big assertion out of the blue...could you elaborate a bit?

BarryTheHatchet

@JohanLarsson looks fine to me

@Ixrec 1) Speed is everything, fast answer gives 5-10 votes. If a better answer is added to a question that is one month old I would not expect it to get many votes.

Also, not uncommon to find bad advice / code with a high number of votes.

ah that problem

the speed problem?

I can definitely agree that the current voting system is terrible at dealing with late answers or subtly-incorrect answers

5:20 PM

also there is probably a skeet effect, I'd expect superstars like skeet to get a bunch of votes just because of their fame

I have never seen skeet write a bad answer though.

@BarryTheHatchet The SE network exists solely for my personal enjoyment. Whether I comment, vote, answer or ignore a question, it's because I find it interesting (or not). I'm not here to create and curate content for our SE overlords.

10

A: How is hotness score calculated at larger sites?

Rough pseudocode: IF siteId in (over-represented sites) OR voters include (gnat) THEN Score = Score*0.2 May not be entirely accurate, but close enough to give you a rough idea of how primitive this is. There's also an additional penalty applied to questions if a previous question from the sam...

BarryTheHatchet

@JohanLarsson definitely

@amon Yes, you are.

You're in denial!

0

A: Answers in comments, revisited

I agree with you, sometimes the direct answer pops out in a comment, but there's nothing to be done about it except to write up the answer and flesh it out. (When I comment "Answers go in the answers" the response is usually agreement, but they didn't have the time to risk putting out content tha...

Skeet probably was big on Experts Exchange and probably wrote a lot of documentation before that too.

BarryTheHatchet

5:37 PM

I find Skeetmania boring

I think a lot of big Experts Exchange users came here.

BarryTheHatchet

he's just a guy picking off a lot of c# low-hanging fruit

does he know his stuff? yes

@BarryTheHatchet did you do Experts Exchange?

BarryTheHatchet

is he the best person ever? no..

@AaronHall I have had an account since January 2000

@AaronHall I couldn't really claim to ever have "done" it

I didn't do it.

I wasn't in a position to.

BarryTheHatchet

5:40 PM

Why?

I only created my SO account 4 years ago, and that's when I started learning how to answer questions on a site like this.

hey, I gotta go.

cheers!

BarryTheHatchet

bai

BarryTheHatchet

6:02 PM

gah

I just can't comprehend why people cling on so much to answers in comments

The "but I'm helping people" has started already

Seriously, am I missing something really fundamental? Or are people really dumb?

:( :(

Q

&

A

I would hate to get a look at an ORM written by anyone who writes answers in the comment sections. Must be one heck of a mess..

messy orm is a pleoansm

I'm too busy teaching jimmy calculus to go read that MSO thread about comment answers unfortunately

BarryTheHatchet

speaking of jimmy

srsly wtf is he

I think he's spending all his chat time out in the aforementioned wilderness where MichaelT is

not sure why

BarryTheHatchet

disappearing without a trace twice in one lifetime is impressive

@Ixrec especially odd since he has been on P.SE (ostensibly) as recently as 16h ago. Although not posting anything.

6:07 PM

where is this wilderness?

BarryTheHatchet

well exactly

hmm I haven't won a pittance on the weekly lottery for a while :(

@JohanLarsson I remain unsure as to whether I'm supposed to mention its existence or not

BarryTheHatchet

I should take this opportunity to reduce my play rate, as my subconscious may now accept that action

ok np, was just curious

BarryTheHatchet

@Ixrec The Lounge?

awww what they took out the Top Cat game :(

6:12 PM

the consensus seems to be that I cannot be any less cryptic than that, sorry

BarryTheHatchet

really don't care

"I have a question, which will most likely be closed in a few minutes/hours". Then ask it in Programmers rather than SO. — Jérôme 38 secs ago

@Duga what the f

BarryTheHatchet

ahahaha

I don't think we've ever had such a blatant confirmation of SO still having the "toilet bowl" mentality towards us, as opposed to something slightly more understandable like "subjective/open-ended questions"

BarryTheHatchet

6:23 PM

I thought he was joking to start with

Because that's spot on, frankly

But, in reality, he didn't actually mean "shit questions go to P.SE" and I don't know why you'd ever think that anyone would

More likely to be a mistaken assumption of dichotomy. Not on-topic here? Must be on-topic there, then.

the "toilet bowl" NPR mentality was more or less an assumption of dichotomy (plus strong quality connotations)

but w/e

BarryTheHatchet

lol, see, someone downvoted this

1

A: Repository UUID equivalent

No, there is no unique identification for a Git repository.

it's literally THE answer to the question posted. how is that not "useful". do me a favour.

tools

BarryTheHatchet

7:12 PM

at pub now yay

apparently the FB mobile app is at version 70

7:31 PM

after spending way too much time on that MSO answer, I currently think the most interesting part of that thread is how the detractors are (probably not intentionally) only posting comment answers there instead of real answers

Hey, guys, answers go in the Answers. — Aaron Hall 2 hours ago

BarryTheHatchet

well, Meta's a bit different

though it is ironic. as the discussions have evolved most of them are now ready to be converted to answers (which can be voted on) but I'm not holding my breath

indeed, even my answer is 50% responding to the comments

BarryTheHatchet

if I were to be cynical, I'd wonder whether there were a sort of "hiding from downvotes" situation going on

but I don't really suspect that atm

it would be nice if SE still enforced the "quality help over any help" aspect since it seems to have been forgotten

I don't think anyone's intentionally doing that, but I think that may be the primary consequence of the attitude that "comments don't require effort, but answers do"

BarryTheHatchet

contributes in large part to the site's decline as far as I'm concerned. not far off just another YA, EE...

mebe

7:34 PM

which is a bit ironic since here on Programmers.SE, I usually spend far more effort on my comments than on my answers

BarryTheHatchet

and your close vote reasons? ;p

those are a significant portion of my comments, yes

BarryTheHatchet

knew it!

though most of my close votes are for the standard reasons (as it should be)

BarryTheHatchet

does anyone else pronounce P.SE as "pissy"?

j/k

BarryTheHatchet

8:11 PM

anyway

we didn't declare process mangement/methodology questions as off-topic at some point, did we? programmers.stackexchange.com/questions/314482/…

BarryTheHatchet

9:01 PM

probably

oh oh oh

i've got it

probably.stackexchange.com

yeeeahhh

2 hours later…

10:53 PM

I'm voting to close this question as off-topic because it is a polling type question (Please make me a list of...) rather than a question relating to an actual problem with programming or programmers tools as defined in the help center guidelines. — Ken White 25 secs ago

@KenWhite: On the other hand, it's a fun question, and I like it. If anything, it should perhaps be taken to Programmers.SE, but I'm fine with it. — Dolda2000 32 secs ago

1 hour later…

BarryTheHatchet

11:58 PM

oh god and now someone's claiming that the "foo* a, b;" edge case is sufficient justification to confuse the frak out of newbies

sigh.

no wonder there are so many awful new programmers out there when these are the folks teaching them :(

« first day (2039 days earlier) ← previous day next day → last day (2945 days later) »

Transcript for

Apr '163

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