« first day (2021 days earlier)      last day (2961 days later) » 

7:00 PM
oh
 
our team doesn't have a standard folder structure though, we mostly make that up on the spot and change it whenever it gets annoying
 
sure
we've lost Thomas again
 
his cpu is probably maxed out from rebuilding a nasty STL template tree
 
ew why would you use the STL
 
the STL is fine for most stuff
 
7:02 PM
this isn't 1996!
 
@whatsisname Barry's pointing out that the C++ standard library is technically not "the STL", I assume
 
when's the last time it even had an update
 
@BarryTheHatchet what's the difference between having auto complete and being able to google?
Sorry, reaching really far back
 
@Ampt absolutely nothing as far as I can tell
which is why I cba with an IDE
truly marginal benefits
 
so it's really a preference thing, in which case it doesn't really matter, no?
 
7:03 PM
meanwhile you're lumbered with their way of doing things
@Ampt indeed
 
@Ampt: speed
 
minus the fact that every time I want to look up a function I don't need to spend 10 seconds googling
lumbered?
how so?
 
you do if you want to find out what the function does and what the preconditions are
 
the vast majority of IDEs support it and I can always go back to googling
 
and you do want to find that out
 
7:04 PM
part of the reason I don't get the adoration of autocomplete is that any time I don't remember the method name I want, I probably want to read its documentation before I use it, and that means searching the docs in another window
 
@BarryTheHatchet I get the doc string as a part of the auto-complete
 
I think autocomplete is glorious because I can remember most of stuff but not 100% of it
 
it gives me all the documentation you would get from googling, automatically
 
@Ampt I mean, once you're using an IDE, you're trapped with any constraints imposed by that IDE. Their build system. Their way of doing precompiled headers. Their this, their that. Don't even get me started on how badly IDE project files work with version control diffs, omg
 
and autocomplete bridges that gap for me
 
7:05 PM
So if there's no notable benefit in autocomplete over simple Google or using your brain, why not stick with the flexible approach
 
@BarryTheHatchet THIS
 
@BarryTheHatchet Agreed - Eclipse and Netbeans are awful at this, but IntelliJ is highly configurable that way
 
great then
 
also, quit committing your local dev environment cruft into the version control :P
 
@Ampt well, I don't
 
7:06 PM
that's the solution to bad version control. that's a local thing, not a team thing
 
my makefiles are in version control and I don't actually need anything else because I'm not using an IDE
 
I use intellij, my teammate uses eclipse, neither of us commit out IDE specific files
 
there is notable benefit in autocomplete, it is considerably faster
 
You just make it sound like you're vehemently anti-IDE
 
how so?
just because I don't use one doesn't mean I want to flay anyone who does
 
7:07 PM
and you don't have to move the mouse or alt tab to another window and wait a moment for google to return something, then make a decision on what link to click, etc
 
also, btw:
21 hours ago, by amon
Far more entertaining than a meaningless flame war pro/con conditionals is watching @Barry spectacularly failing to ignite said flame war.
 
unfortunately our infrastructure team effectively forces us to use their proprietary IDE for Javascript, so I don't have any choice in the matter
 
eat it.
 
34 mins ago, by BarryTheHatchet
You'll be completely hosed when it comes to job interviews, y'know
 
and our C++ is sufficiently special that the typical IDE benefits simply won't apply to it
 
7:07 PM
@BarryTheHatchet this ain't even a flame war sweetie.
 
that's not about using IDEs
 
this is us understanding eachother viewpoints
 
that's about Thomas allegedly being entirely reliant on auto-complete for producing code
good luck with that when they ask you to sketch out a C++ program on the whiteboard
 
of course, whiteboards don't compile
 
so what if you're reliant on it
 
7:08 PM
well the same could be said when you are asked to do it without the internet?
like you admit you google it - what's the difference?
 
yeah if you're a scrub, sure
 
if you're at the whiteboard and you can't remember the order of arguments for something then just say so
 
we're probably all reliant on documentation at least
 
and anyone who has a problem with that is a loser and you don't want to work with them
 
@Ampt the implication was that he needed auto-complete on his own code, even
I hadn't really quite figured out what his thing was yet, at that point
admittedly
 
7:09 PM
@BarryTheHatchet I don't think that was the implication
most of us work with fairly large codebases
 
still haven't because he's gone
 
I think thomas is moreso
 
knowledge in the mind vs knowledge in the world
 
my codebase has 900,000 LOC
don't talk to me about large codebases :P
 
I really need to go count what ours is at
 
7:09 PM
I dunno, maybe it's just cos I'm savant-like :D
 
autocomplete lets you keep less in your head, more in the world, and so you can think about more pressing things
 
@BarryTheHatchet Thomas works on government regulated aerospace stuff
I think he probably has you beat
 
I know he does
Great good for him
 
you're savant like? lol
 
Still doesn't really explain why he can't figure out how to write a play C++ app without autocomplete
it's all moot unless he decides to come back and finish the conversation he started
I do like how Notepad++ has a feature to convert silly space indentation into awesome tab indentation
makes importing code written by idiots into a less arduous affair :)
 
7:12 PM
He just wanted to know how a typical C++ dev has their workspace setup
 
I suspect the answer is they have it set up by their nearest C++ workspace developer
 
and I didn't understand the question
or why you're repeating the conversation blow by blow
 
there are advantages and disadvantages to using autocomplete and not
acting like one is superior to the other in an absolute sense is... inane
 
oh god
 
Because I think you're being rather rude about the whole thing.
 
7:13 PM
there are advantages and disadvantages to using COBOL
 
so whatever they put on my hair today smells really nice
got a free head massage too :D
 
@Ixrec Advantage: Job security and $$$,$$$
Distadvantage: Wanting to be embraced by a face hugger every day.
 
thanks, now I'm having Alien: Isolation flashbacks
 
I should probably learn me some COBOL one of these days
 
also one thing I experienced when doing C++ is that configuring a debugging environment is considerably worse than something like Java
 
7:16 PM
I could make some good cashmoney and it would be only marginaly more awful than what I do now
 
@enderland gdb myProgram ... run
so hard
 
Yeah I've found GDB to be better than Java's alternatives myself
 
I just assumed Java works well, you mean with all the infrastructure pieces they have there isn't a good debugger? :o
I've not written much java and hope to keep it that way
 
one of the downsides of having gigabyte-sized executables is that gdb loses most of its practicality
 
@enderland nah, they're ok, GDB was just always really easy to use IMO
 
7:18 PM
whatever debugger eclipse and intellij are hooked up to for java are fine
 
0
A: nm get undefined symbols for whole archive, not separate object files

BarryTheHatchetThe archival tool just creates ... well, an archive. It's not a linker. References are resolved by your linker, and this process will find foo.o's reference to bar to be resolved in bar.o. The archiver can't do that, and nm won't do it either - it's just running itself over each object inside the...

fact check please
 
what the hell is he trying to do? Reference bar() from foo and missing() from bar()?
that's not how you include methods from other files
 
I think he's trying to add to his workflow "check that every function referenced in but not defined inside this archive is something I know about because I deliberately define it in some other archive"
 
unless I'm missing something shouldn't he just have a bar.h with the function prototype that he then imports into foo?
 
well, yeah, ideally
that wouldn't change anything about the question though
 
7:23 PM
well it's not finding bar because he's not doing that though, right?
 
I suspect that element is deliberately abstracted away
no, this is about definitions, not declarations
it's a link thing
 
like how does his linker know the extern into bar() is defined in bar.c?
 
because it's a linker's job to know
that's what a linker does
the problem he's got is that he's not invoking a linker so he's not getting the informational output from nm that he was hoping for
 
psr
If you were a linker you would understand
 
ok, what if he had a foobar.c that also had a bar() defined?
 
7:24 PM
then that's multiple definitions
and you can't build your program
 
that would stop a build? even though nothing is explicitly linked?
 
what
"explicitly linked"?
what's that
 
via headers and function prototypes. Why cant you re-use the same method name inside different c files?
that's not allowed?
 
headers have nothing to do with the linker whatsoever
declaring by including a file that contains int foo(); and just writing int foo(); directly are exactly the same thing
 
having two definitions for the same name (in the same namespace) has always been a fatal error...
 
7:26 PM
@Ampt of course not? you can't have two functions with the same name..... how would that work?
 
because they're in different files?
@Ixrec plain old C has namespaces? I thought that was a C++ thing?
 
global is a namespace
 
@Ampt namespace are the C++ solution for this problem
 
we're hitting the limits of my C knowledge here
 
iirc the C convention is libraryname_actualfunctionname
 
7:27 PM
@Ampt yeah you seem to be confusing includes for imports
#include is simple text replace, nothing more
there's no "explicit"/"non-explicit" linking
your compiler needs to be able to see a declaration, and your linker needs to be able to see one definition
that's it
we have a ton of conventions to make things not blow up in practice (e.g. header files), but theoretically there are a gazillion ways to achieve a successful outcome
 
shouldn't GCC be doing the linking? or do you need a flag for that maybe?
 
GCC does the linking, yes
GCC includes a compiler (gcc) and a linker (ld). When using the -o flag the linker is automatically run after compilation, but it doesn't have to be
This guy is looking at the output of compiling, only
 
man serious perk - free beefsticks and cheesesticks
 
He's compiled two objects (call them "modules" if you like), and bundled them into an archive (or "static library") using ar .. which is really just an archiver, hence the name
Now he wants to know why nm thinks that, within the scope of those two "modules", bar is not defined... when it is
 
probably best not to call them modules, they have nothing in common with real modules
 
7:31 PM
The answer, I think, is that nm is just telling him what functions are undefined in each component module
And I think that because he hasn't done any linking yet so how could it know
Also otherwise he wouldn't get that result
@Ixrec Agreed - trying to bridge the language gap a bit. Hence quotes!
They're strictly "objects" but that's ambiguous too, so yikes
 
I'm trying to figure out why Ampt thinks the one definition rule implies that no linking is going on
or whatever it was he was implying
 
I gave up trying to figure out why Ampt thinks anything ;)
I mean, he's a sock ... what do you expect?
 
@Ixrec It was my understanding that you had to explicitly include a function from one file to another, but I see why that's wrong.
 
"translation unit" is probably the closest thing we do have in current C++
 
@Ampt It's just one of those things that's a really, really, really good idea
 
7:33 PM
@Ampt ah, yeah in C and C++ unfortunately you can only include files, not functions/objects/classes/methods/etc
 
But in order to simplify the MCVE, I'm glad the OP decided to keep the file count low here and just give us the function declarations inline
 
@Ixrec yes, and those files contain function prototypes that you want to include
 
Dunno why he wrote extern, though.
 
again, I am of the opinion that the lack of proper modules is modern C++'s biggest weakness
 
I think its biggest weakness is the utter mess that is TMP
it's a huge abstraction leak
 
7:34 PM
@BarryTheHatchet usually it's for embedded stuff where the function is defined in the compiler or whatever
 
you have to be a brain surgeon to do basic things with C++ templates. it's utterly horrible
 
at least that's how I saw it used
 
in second place is the semantic mess that is auto (make up your mind what you want to be, C++!) and std::move (which doesn't move anything)
 
that situation improved greatly with the recent C++ standards
 
@Ampt No, it has no effect whatsoever.
@Ampt extern is useful for "declaring" variables in the way you'd declare functions (i.e. without defining them at the same time)
 
7:35 PM
and personally I've never needed to do anything more complicated with templates than instantiate "STL" containers
 
But for functions it's completely redundant
 
yeah you're right
 
@Ixrec I think it's getting worse not better
 
what was the friggin keyword for the compiler included library then...
 
C++03 was passable
@Ampt There isn't one
 
7:36 PM
variadic templates? decltype?
 
The C and C++ standard libraries are in "user space" - they exist in the same form your own code does
 
features like that completely obsolete some of the more common TMP madness
 
@BarryTheHatchet Not in the base C++ standard, no
 
I'm talking about standard C and standard C++ of course
 
I was using a proprietary compiler and standard though
 
7:37 PM
though I've done a bit of embedded work and haven't seen anything that diverges as much as you're describing, I can imagine it
 
there was a way to say "Put this function for this chip here kthx"
 
and now I have no clue what ti was
it wasn't a macro, I know that
 
are you sure it wasn't just a predefined macro
lol k
 
haha
yeah it eludes me now but it was something goofy
those compilers were even worse though
 
7:38 PM
sounds like a thunk
@Ixrec yeah but they do it in such an obnoxious way
it's not expressive at all
upvoted for "try to" — BarryTheHatchet Mar 11 at 12:25
lol wtf
 
7:53 PM
wooo free lunch tomorrow and friday
 
user41796
> It's a trap!
 
@GlenH7 nah, it's a celebration day type of thing tomorrow (somehow meaningful to the company, not sure what the significance of teh day is, but hey free food) and every two weeks we get a team lunch
 
yep, sounds like a trap to me
hopefully it's not a tarp though
nothing worse than rubbery food
 
user41796
Um, I feel completely out of it. I just saw the notice that MichaelT has had his account disassociated. :-(
 
@BarryTheHatchet our cafeteria has ridiculously good food
I still bring my lunch because I'm trying to lose weight and eating here... not the plan for that hah
@GlenH7 yeah.. :'(
 
user41796
8:00 PM
Skimming the earlier conversation about it
 
yeah MichaelT was recently diagnosed with Disassociative Identity Disorder
> And, remember, every $10 is an entry so the more you donate, the better your chances are to win. So if you want 5 entries, just donate $50, if you want a billion entries, donate 10 billion dollars, etc.
lol
0
Q: How to configure list of containers / apps for Docker to run?

Milan IlievWhat is Docker's native mechanism for specifying to Docker Engine or Docker Compose the list of applications (and therefore, via compose.yml, the containers) to run on a given host? An illustrative scenario I would like a given host to run all required containers for web applications A and B. L...

for you, @enderland
 
user41796
So skimming the earlier conversation about Michael, and having briefly read that meta post ... it makes sense to me.
 
@BarryTheHatchet hahaha that's our exact use case
 
user41796
It's the exact same reason why he left SO. Except this time he bailed before he let his professionalism drop.
 
why did he bail?
 
8:08 PM
@GlenH7 I was reminded of a long break I took after writing this meta post after seeing some of the context on that meta post
 
i missed it
 
user41796
@whatsisname enderland's own summary is fairly apt for MichaelT as well
 
gotcha
 
it's all speculation basically but I suspect this comment is the tl;dr version:
I agree, but am more concerned that no one prompted the asker for the information needed to make the question better (either for our site or for migration somewhere else). It is key to lead by example. programmers.stackexchange.com/q/312743 programmers.stackexchange.com/q/312761 programmers.stackexchange.com/q/312721 . I'm tired and done. Its someone else's turn now. Go through the recently closed list, check to see where there is confusion and better guidance for users. Thats what the 10k mod tools are for. — user40980 yesterday
 
user41796
Realizing that he was focused on building a high quality community, and combine that with a lot of tone deaf responses from CMs ... I'm not surprised.
 
user41796
8:10 PM
At some point you have to say "I'm doing carrying this load, it's someone else's turn."
 
I still don't understand the link between wanting to not spend time on moderation and lobbying any more ... and deleting every single trace of yourself from the entire network
looks like a ragequit to me tbqh
 
user41796
tbh, there's been a dramatic shift in (lack of) community moderation after the 3VTC experiment wound down.
 
@enderland Hey! My answer is right below yours on that list :)
 
user41796
I think there was disappointment from many of the regulars, and then several of us also ran into pressing real world issues
 
yeah ok but then just don't log in for a while or something
 
user41796
8:11 PM
@BarryTheHatchet Not a ragequit
 
user41796
@BarryTheHatchet He had already done that for a bit
 
people who ragequit don't just... quit, they rage quit
 
then explain to me the link between stepping back from moderation, and entirely removing yourself from the entire network
how does account disassociation aid with the stated goal
 
user41796
At some point you realize that your actions aren't having the effect you want them to have. It's perfectly logical to say "FTS, I'm done." at that point if it doesn't appear that there's any hope of things changing.
 
one thing I could buy is that it will help you not slip back into the patterns of work you're trying to escape
 
user41796
8:13 PM
@BarryTheHatchet that's exactly why he disassociated his SO account
 
user41796
so that he couldn't be tempted
 
alright
well maybe it's that then
 
user41796
It is.
 
could have said goodbye though!
@GlenH7 how do you know? you didn't even know he'd left until five minutes ago!
 
it's no different than a breakup, tbh, you want the same thing - a clean break
 
user41796
8:13 PM
@BarryTheHatchet Go read the meatball wiki on how people leave a community. He played this one by the book on how to leave properly.
2
 
dunno still seems like a massive overreaction to me but there we go
 
user41796
@BarryTheHatchet How do I know? I've talked to the guy for years in this room.
 
@GlenH7 a well-educated guess then but you're stating it as if it's a fact you've heard direct from the horse's mouth
and it's not
to be clear, I don't disbelieve you
 
user41796
I'll grant that point. But you also have to grant that he's consistent in his behaviours.
 
8:15 PM
that's certainly so
 
for many of us MichaelT is someone we would consider a close friend
 
do you have a way to contact him off-network?
 
sometimes, it is easy to predict/understand what and why a close friend would do what they do
 
if you do, perhaps now would be a good time
 
user41796
@BarryTheHatchet pretty certain, yes
 
8:16 PM
@enderland I agree
 
@BarryTheHatchet why?
 
@enderland because you consider him a close friend and he may need your support right now?
more importantly, don't forget to vote on my name
 
user41796
@BarryTheHatchet I don't think he's upset. I think it was a perfectly rationalized decision on his part.
 
haven't got all month y'know
@GlenH7 alright then
 
user41796
And I'd be willing to bet he's been mulling this over for months
 
8:18 PM
I'd have expected him to write a goodbye message to all of you lot were that the case
where "him" = "a person"
not him specifically
you know what I mean
anyway I'm sure he'll be back
 
what @GlenH7 is saying, hah
 
bbiab in a min from pub on MOBILE CHAT UI BETA
cos I'm leaving for the airport at 4am
so I'm going to the pub now at 8.20pm
what could possibly go wrong....?
 
user41796
@BarryTheHatchet I'm not so sure about that. Looks like he had all of his SE accounts disassociated.
 
@GlenH7 yep he did
but it's trivial to set new ones up
give it six months
I wonder what name I'll be on by then... hmm
 
@GlenH7 I think if I ever feel SE is less than beneficial in my life, I will do the same thing
 
user41796
8:29 PM
The sites are a nice little diversion, but that's all. I haven't seen much to demonstrate that it would have a significant impact on real world events.
 
what do you mean? you are better at separating your non-work life from SE than some of us... ;-)
 
user114359
@BarryTheHatchet no, because he was clearly annoyed for some time and would not want to be perceived as whiny or ragequitting. He likely correctly assumed someone would see one of his prolific posts was no longer attributed and word would spread. And it is not like we have no idea why, like I said, he was clearly annoyed by a few things.
 
user114359
23 hours ago, by gnat
^^^ MichaelT has left Stack Exchange
 
user41796
@enderland The benefits of site participation are all intangible.
 
user114359
that's all he needs. Besides, some of us do know how to reach him outside of SE. I might, but I would give it a week or two first.
 
user41796
8:34 PM
I'd agree
 
perhaps he will focus on writing a blog where he controls it and doesn't have to deal with the weeding and janitorial aspects
 
user41796
And he was planning on getting a road bike. Spring is in the air. And it sounds like he's got an excellent excuse to get out and ride now.
 
it needs to rain a bit more though
 
user41796
It's been a dry winter / spring here as well
 
well the rain would be more to remove the salt from the roads
 
8:39 PM
11 mins ago, by enderland
@GlenH7 I think if I ever feel SE is less than beneficial in my life, I will do the same thing
don't tempt me too much ;-) haha
 
user41796
@enderland Walking away from the keyboard can be a good thing at times
 
user41796
I think your break from the Workplace benefited you in that regard
 
yeah absolutely
 
user114359
@GlenH7 I have, at times, stepped away from a bit (e.g. last December). But never deleted my account. I doubt I would do that.
 
user114359
Right now I am spending more time hanging drywall in my basement than I am reviewing questions on this site.
 
8:43 PM
conversely I have intentionally avoided ever getting too close
 
I have also walked away from online communities that I really cared too much about
 
I rarely participate in the meta discussions and whatnot
 
user41796
@Snowman Keep in mind that he ended up saying some less than civil things during heated discussions on MSO. And he really wasn't happy about losing his cool like that. So... sometimes it's better to make sure that sort of a mistake can't happen again
 
ultimately SE is a for-profit enterprise that I am not a shareholder of
 
user41796
I had asked for proof as I didn't believe that was in his character. I saw some of the comments and had to agree that they were very out-of-character for him
 
user114359
8:44 PM
@whatsisname sort of like a for-profit wikipedia, with just as much drama
 
right
my willingness to engage in the drama is very limited because the work/payoff ratio is not good
 
user114359
at this point in my life I mostly just laugh off the drama in anything I do. I don't have time for it, and I graduated high school 20 years ago. That kind of crap is in the past.
 
the second is a bit.. PII heavy (though I guess not too hard to find if you know the github?)
 
Well, it's not like I'm posting his email address or cell phone number.
 
@RobertHarvey well a lot of those have real name associated with them
 
8:57 PM
hello from mobile chat ui beta
@GlenH7 lol that's so meta... when you quit your job on Workplace :D
 
I was hoping this question would get updated with details... I like algorithm questions
but I'm not about to go learn Othello to answer a question here :p
 

« first day (2021 days earlier)      last day (2961 days later) »