« first day (1180 days earlier)      last day (1310 days later) » 

11:04 PM
11:20 PM
Q: Improving the display of Picture collections from a user's picture Library

Tolani Jaiye-TikoloCurrently, I have been digging up my old codes. One of the applications I worked on was an Image editing phone app that displays a collection of random pictures from a user's picture, which I limited to less than 80 pictures because of performance. These random pictures are chosen from different ...

@syb0rg Cmake couldn't find sphinxbase for me, here's the error message:
`CMake Error at C:/Program Files (x86)/CMake/share/cmake-3.2/Modules`/ExternalProject.cmake:1696 (message):
error: could not find git for clone of sphinxbase`
though that's not really a dealbreaker, usually I download this stuff separately anyway
11:38 PM
@Pimgd I'm not sure that the Cows and Bulls question is a good example to discuss. The author openly admits that the code does not behave as intended.
It's a really bad example but...
It's what happens when you post an answer
it ... ends abruptly with "the code is broken" and you can't review anything from there
because from contradiction, you can prove anything.
(okay that's a bit of a leap but still, large swaths of commentary that seems reasonable can be wiped out by OP stating "oh yeah that, it was a bug, nvm, fixed it" and there goes (for example) your optimization based on all values being positive values)
Well, ideally we hope that such questions don't get posted at all here. When the question openly asks for help to fix a bug, we shouldn't answer.
So, I want to talk about another example.
Q: Testing if array is heap

kidinme12This method tests an array of integers satisfies to see if its a max binary heap. This means each node is greater than or equal to it's children each node has at most 2 children the leaf nodes are filled from left to right This is for heaps that do not use element 0. See here. /*returns true...

from today =)
Yes, I was thinking you might ask.
Code doesn't compile, obviously.
at the time, it was VTCd broken code via the code, no comment, obvious in hindsight but it slipped my grasp; had I compiled the code via Ideone (a 15 second effort) I would have known
11:44 PM
It would have been nicer of me to comment there.
Which is what I'd want the community guideline to be
you vote to close like that, you add a comment.
Tunaki did so and I went "Yep"
That's fair.
It... does have the side effect of telling the OP what is broken
I dunno how important that is.
Keep in mind that the explanation of the close reason also acts as a comment, of sorts
@Phrancis Agreed.
11:46 PM
> Questions containing broken code or asking for advice about code not yet written are off-topic, as the code is not ready for review. After the question has been edited to contain working code, we will consider reopening it.
It's not as "personable" of course but if the OP reads it, all the information is there
I am not looking for a comment that explains to the OP "heeeey your code is broken you should fix it or we're not gonna answer your question"
I probably should have just written "Code won't compile." as a comment.
I am looking for a comment that explains why there's votes to close for broken code on a question where the OP states that everything works fine
@200_success that would have sufficed for me
"Code won't compile" would have struck a good balance between transparency and giving too much information, in this case.
11:49 PM
@Pimgd That's getting into pretty "grey" territory, as there is no way for any of us to test the code on the OP's machine
@Phrancis situation as follows: OP says code is fine. Someone else votes to close the question because the code is not fine. We're left guessing in what way the code is not fine.
@Pimgd Oh. yeah. In cases like that (where it's not obvious or evident), a comment would be nice
… and that puts users who process the Reopen queue in a bad place.
which is why I'm proposing that we make it a guideline,
similarily to how you're not supposed to hit "No Action Needed" in first posts queue just like that
@Pimgd That's all well and good, but enforcing it would be completely voluntary, and more to the point, the users least likely to follow guidelines are the users most likely to do exactly what the guidelines warn against
11:53 PM
Which is when you point them to meta, or a help center article...
and they'll not do it next time?
Devil's advocate argument: if it's not obvious why it's broken, then it's actually on-topic.
It has to be voluntary.
@200_success Stargreed
@200_success this scores points, but I think it's not strong enough to win the argument
Commenting definitely has to be voluntary, due to the intricacies that I pointed out.
11:55 PM
@Pimgd That is a fair point I think, although, perhaps someone's question (on-topic or otherwise) is not the best place to educate another, unrelated user
@Phrancis Perhaps this is a different category, but... we do this all the time?
I suppose so...
With Code review comments on SO questions
Fair enough
11:57 PM
But the proposed voluntary guideline is not that different from the status quo, is it?
Hmmh, well, I was kinda hoping to see a "yeaaah we should totally do that" response to my meta question
and then I'd opt for "minor edit to help center page" so that recommended course of action after voting to close for obvious bugs = leave a comment
It's more like "Yeah, it would be a good idea, but people may have valid reasons for not commenting."
@200_success which is why it cannot be mandatory
(How would the system even know between you voting to close because OP says "fix mah bugs plz" and you spotting an obvious bug in the code?)
which is not that different from how most people normally behave.
Well, I'm curious how @Pimgd would word that guideline, if it were to be something that we would refer to for time to come, it really sounds more like some unwritten rule that most people just kind of figure out after a while (by the time they reach VTC privileges, most likely)

« first day (1180 days earlier)      last day (1310 days later) »