Please check the question. Software developer work day productivity: Is there an ideal code editor usage percentage? After initially getting a good response, the question got targeted and closed as off topic by 5 users/moderators. If you check the answers and comments on the question, there are many experienced stack exchange users who didn't mentioned or pointed anything wrong with the question. It has got fairly good answers and caught interest of many people. I believe there's a system to vote for re opening it. Please do so if you believe it is closed unfairly. Thanks

  • 4
    Popularity is irrelevant if it's not on topic. On the face of it, that's a question asking about whether a particular metrics for software development is useful and what a "good" range is. An interesting debate but likely unanswerable and just not really a topic that Workplace tends to handle. We also don't have questions on what the ideal brick-to-mortar ratio is when building a house or how many customers a hairdresser should be able to handle in a day. Software Engineering might be better for this. Check help center as well. – Lilienthal Apr 14 '20 at 15:59
  • @Lilienthal I believe ultimately its about productivity which would fit better under workplace than under software engineering. That's just my opinion. And if it was asked at a wrong site, I would've appreciated if it was pointed out in the early stage than closing it abruptly when a large number of users had already engaged with it. Its still surprising to me why none of the large nunber of users who upvoted, answered and commented on the question didn't pointed out that its the wrong site for this question. – Vijendra Parashar Apr 14 '20 at 16:15
  • 2
    Because the vast majority of them will have come to your question from the Hot Network Questions. This influx of off-site users will upvote the question and answers because they find it interesting. They will never vote to close it because they tend not to know what is or isn't on-topic here to begin with. You can read through some of the questions around HNQ on meta here to get a sense of our love-hate relationship with it. :) One of the main issues is that HNQ will propel a question to mass visibility and a large amount of upvotes when perhaps it should have been closed or put on hold. – Lilienthal Apr 14 '20 at 16:49
  • Fair point. This should be happening on all sites every second. There should be very few people knowing exactly what is on-topic on each of the SE sites. So, the issue is with the platform itself. If it doesn't want off-site users to upvote any off topic question, it shouldn't serve the question to them through HNQ in its early stage. I shall back off now. Lets see if there are people who would reopen it, or else better to close this question as well. Thank you – Vijendra Parashar Apr 14 '20 at 16:59
  • I mostly agree when it comes to HNQ. In fact we tend to remove questions from HNQ that meet these criteria but it's not something we can do in advance and it's admittedly an arbitrary decision. Anyhow, I hope you got some useful input at least. If you believe that parts of your question weren't properly addressed or you have a better way to word it you could consider editing it or launching a new more specific question. – Lilienthal Apr 14 '20 at 21:11
  • @VijendraParashar - I almost never vote to close, didn't vote to close this one, and have already voted to re-open it. Still, you already got 10 decent answers, yet haven't accepted one. – Joe Strazzere Apr 15 '20 at 15:23
  • @joe I was just waiting for people's upvotes on the answers. After it got closed, I just decided to get it re-opened first. A little stubbornness on my end. Lets see if that happens, otherwise I am going to accept the top answer. There's no reason to worry about that and thanks very much for voting to reopen – Vijendra Parashar Apr 15 '20 at 15:28
  • 1
    @VijendraParashar - I guess my point is that I don't see much need to reopen the question. You already got 10 opinions. How many more are needed? – Joe Strazzere Apr 15 '20 at 16:42
  • Don't need more, but I felt it was unfair to close it, so I was maybe just idiotically fighting for it. Dont have time to continue with this struggle. Accepted the answer, moving on – Vijendra Parashar Apr 15 '20 at 16:49
  • @VijendraParashar If you want to start a dialog about the issues raised in your question reddit is probably a better outlet. People have very particular ideas about subject matter and format on stackexchange. But it sounds like got a draconian employer, who craves too much control. To be fair, however, you did get some thoughtful responses here and that's what's most important. – teego1967 Apr 17 '20 at 18:54

I wasn't one of the users who voted to close - although I can see why they might have done. The subject of the post asks a fairly specific question about a very specific metric - and frankly it's going to be difficult to answer that question without a great deal of information about the developer's workload, what technologies they work in etc. And even if you got there it's not even a particularly useful metric.

The body of the post however seems to be skirting around a more workplace SE-centric goal of how to push back on management expectations that seem a little arbitrary and generally unreasonable. Your opinion clearly is that they are unreasonable (and FWIW I agree with you, as I think most people interacting with the question did too) but I think you might be better off asking how to push back on that unreasonableness rather then just asking if we all agree with you.

I think if the question were re-focused to make a clear goal of pushing back on these metrics and asking how to go about doing that it would be on topic here.

You must log in to answer this question.

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