What is the appropriate flag to raise or action to take when I suspect question to be trolling, or submitted only to inspire flame war?

I am talking about this question, which I voted to close as opinion-based, but I am not sure if that's appropriate flag to communicate my impression.

  • 5
    Possible duplicate of Fair to VTC for trolling and the other linked questions there. SOP remains: vote to open/close based on content not on motivation, flag if you suspect foul play. – Lilienthal Feb 20 '20 at 8:06
  • 1
    another meta discussion on the same question: questions based on bad or wrong assumptions (seeing how it managed to spark two meta discussions one can indeed suspect an attempt at trolling:) – gnat Feb 20 '20 at 9:55
  • 1
    @gnat it became trolling because we engaged in meta discussion? :-) – aaaaa says reinstate Monica Feb 20 '20 at 14:39
  • of course not :) having (one) meta discussion about particular question is perfectly normal – gnat Feb 20 '20 at 14:41
  • 1
    I fail to see how this is "trolling". Are you sure you've used the right word? Calling someone a troll just because you don't like their question is pretty damned unfriendly. – Richard Feb 20 '20 at 16:15
  • 1
    @Richard i follow "posting inflammatory or off-topic messages in an online community" definition. the OP seems to be off-topic (too broad, no achievable goal) and also inflammatory (line count as productivity metric has been debunk at least since 1975 when The Mythical Man-Month came out) In fact, I support question "how to measure productivity" but it will be probably off-topic – aaaaa says reinstate Monica Feb 20 '20 at 16:24
  • 2
    @aaaaasaysreinstateMonica - As a non-programmer, I don't really see the problem with creating a productivity based metric to determine whether someone's doing a good job or not (and I'm assuming that a programmer who writes a very small number of lines is probably doing a bad job). A good answer would explain it to me, but it's not trolling to ask such a question – Richard Feb 20 '20 at 16:26
  • 3
    @Richard after some more reading i, in fact, think you are correct in that "lumberjack's productivity is measured in lumber, why can't we measure coder's productivity in lines of code" is a valid and good question. – aaaaa says reinstate Monica Feb 20 '20 at 16:27
  • 1
    @aaaaasaysreinstateMonica - Well, it's certainly a valid question. I wouldn't say "good" because OP hasn't explained a wide range of things such as code type, what the project is and what research they've already done. – Richard Feb 20 '20 at 16:55
  • 1
    @Richard yup, the context is critical here. That is why perhaps the OP is better suited for PM@SE pm.stackexchange.com as performance connected to the paradigm of project management. In some (bad) paradigm number of lines of code is a valid metric since it's the target – aaaaa says reinstate Monica Feb 20 '20 at 17:30
  • 1
    @aaaaasaysreinstateMonica As some of the linked questions state, we prefer to consider these teaching moments. It's related to the concept of the XY problem. If the OP is asking something based on a flawed analysis or poor perspective it's valid to contest that in answers. This is also the one place where the guideline "answers should address the question asked" should not be interpreted literally. – Lilienthal Feb 20 '20 at 19:37
  • 1
    There's actually a rather interesting parallel here where you considered this a troll when it's likely just a question borne from ignorance. – Lilienthal Feb 20 '20 at 19:38
  • @Lilienthal i think i rushed to attribute trolling to ignorance also because of how blatant was the title phrased. It sounds like an aim for HNQ. – aaaaa says reinstate Monica Feb 20 '20 at 20:45
  • Measuring a developer after lines of code is like measuring a lumberjack in swings per day (and firing him the week he is on duty securing his teammates and marking trees, taking no personal swings at all). If the title of the person has "software" in it, then this should be blindingly obvious. Whether someone posts something that ignorant to troll? We can only guess. – nvoigt Feb 21 '20 at 8:46
  • 1
    @Richard your example with lumberjack reveals what seems to be most troublesome about questions like that. Do we really want to have narrow technically focused, industry specific questions about performance measurements for hundreds different professions here, cooks, taxi drivers, construction workers, doctors, pilots etc etc etc. I just re-read all 10 answers to that question and none was about navigating the workplace, all were filled with details about software engineering / project management – gnat Feb 21 '20 at 8:57

"Trolling" implies intent, which is hard to judge when reading someone's words. So I agree with the generally accepted approach of taking things at face value: If someone asks an answerable question, we should focus on answering it rather than guessing intent. (Outside of obvious fringe cases, which should probably be handled with flags).

I think that still leaves us with a problem, with respect to the question you've linked to: Is this an answerable, on topic question for The Workplace?

I do think it passes the test of being on topic in terms of the subject matter. How you measure someone's performance is clearly a workplace problem.

To me, that is the most important factor to consider, because other factors (is it to broad, is it opinion-based, does it have a clear goal, etc) are, often, easily handled through edits. If someone asks,

Do you think X is a good measure of performance?

that is pretty blatantly opinion-based. It shouldn't matter what I think, it should only really matter what general truth I can describe or what experiences I can speak to, or what references I can provide. Or maybe the person asks something that's too broad:

How should performance be measured?

But ultimately I think those are weak reasons to close the question, because we all inherently know what the person is actually trying to get at, and I don't think it's fair to close a question purely because of the words the author chose when the actual subject they're asking about is on topic. So, perhaps we can suggest that the OP rephrase the question, or we can edit it to keep the wording on topic.

Ultimately, I think this is what you were getting at in comments when you said,

in fact, think you are correct in that "lumberjack's productivity is measured in lumber, why can't we measure coder's productivity in lines of code" is a valid and good question

Perhaps this question could be saved by editing to reflect that.

  • 2
    i thought about editing, but it feels like heavy-handed approach, and the question is kinda "spoiled" by all the existing answers already. And i am still afraid that 'how to measure performance' would be extremely broad topic. It starts with proper process, scheduling, why managers want to know performance and so on. Perhaps I should risk some of the rep to create 'canonical' question for performance metrics? – aaaaa says reinstate Monica Feb 20 '20 at 17:24
  • I don't disagree with respect to this specific question, it is a little challenging to think about how to edit it at this point. We may have lost the OP's interest by this point, but they had responded in comments earlier on, maybe we missed the opportunity to ask them some good clarifying questions and get them to edit. – dwizum Feb 20 '20 at 17:27
  • Another thought is that some "bad" questions may be appropriately responded to with a frame challenge answer, instead of voting to close. I'm happy to see that several people have already done that on this question. – dwizum Feb 20 '20 at 17:34
  • 1
    yeah, the VTC might have been premature, knee-jerk reaction. We probably should wait few days and edit the Q to reflect intent and the submitted answers – aaaaa says reinstate Monica Feb 20 '20 at 17:42

I went ahead and voted to close as opinion-based. I will try to follow what Lilienthal commented:

SOP remains: vote to open/close based on content not on motivation, flag if you suspect foul play

The issue remains: question 'how do we measure programmer's performance?' might be closed as opinion-based or too-broad

In fact, such question was posted before and closed as off-topic: How productivity or working hours are measured in software development jobs? [closed]


Sorry, I don't see how the question was an attempt at trolling. Number of lines coded is a simply but obvious measurement metric for non-developer experience management. It's not perfect, but experienced developers are expected to code more than junior developers, who may struggle at making the code even running.

My answer offers a simple mathematical modeling technique that the folks at stats.stackexchange may expect.

  • 1
    "but experienced developers are expected to code more than junior developers" no, they are not. They are supposed to solve the harder problems. They are supposed to sit in more decision making meetings. They are supposed to train the juniors. The one thing that is certainly not a target to meet is "type more". – nvoigt Feb 20 '20 at 14:22
  • This was my exact thought. It's certainly not well researched and opinion-based, but it's not "trolling" by any reasonable definition of the word – Richard Feb 20 '20 at 16:14
  • 1
    Ironically, the debate playing out right here (HelloWorld's statements versus nvoight's comment) is, to me, an indicator of why this could have been a really valuable question, if it had simply been phrased a little differently. – dwizum Feb 20 '20 at 16:36

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .