-4

There was a question about StackOverflow productivity: Has Stack Overflow saved billions of dollars in programmer productivity? I answered it to the best of my ability. Rather than asking a question, or for claification the moderator closed it presumably because he didn't like the answer. That's an egregious use of an answer-close. Moderators should not being use their delete/censure button in replace of a down-vote. This answer isn't back pocket calculation. It uses all of the official data made mention and explains how that data should be adjusted and in which direction. I'm unsure at why the moderator things it's a "back pocket calculation" and I can't even ask him a question on it.

Please uncensure this post. You can't just delete it because the conclusion isn't of the same bias that you want, or because you want your question to ferment as you wait for more upvotes.

To make matters even worse, Sklivvz the mod is a paid employee of StackExchange. It's clear he is protecting his company and not acting in the best interests of an objective reduction of a broad grandiose valuation to something reasonable and obviously (imho) absurd.

5

FWIW there's a transcript here of a chat which hints at the kind of answer Sklivvz wanted:

for example take a group of programmers and split it in half, one half being denied the use of stack overflow. give them the same problems and measure KPIs

He's looking for published results available via Google Scholar, and/or experiments from an economist.


I think that yours was a relatively low-quality answer.

Downvoting it instead of deleting it might have been sufficient; except that the question was (unusually) attracting 70,000 views from developers who are naive to the kind of answers expected on Skeptics, therefore the http://en.wikipedia.org/wiki/Broken_windows_theory applied more strongly than usually.


I might prefer it if moderators would refrain from moderating their own questions (to avoid any personal bias on their part, and to avoid complaints on your part).

For example it might have been better if Oddthinking instead Sklivvz had been the one who 'policed' this particular question. The end-result might have been similar though (other mods might have agreed with Sklivvz's decision to delete this answer)..

  • 2
    Why is my question low-value? The title isn't asking for a published Google Scholar article -- and, if that's what he is wanting it's off topic for this StackExchange. I provided a sourced answer to his question. His appeal to a single source, or the product of a single site is outside of the domain of a broad Skeptics community. – Evan Carroll Dec 6 '13 at 18:35
  • 4
    +1 for moderators refraining from moderating their own questions. – rjzii Dec 6 '13 at 18:37
  • 2
    @EvanCarroll 1) Words like "ridiculous" make it look like you're trolling or hostile 2) It's original research, which is deprecated 3) It's too long in that it's mostly unreferenced 4) The only substantial part is calculations which say "total alleged money divided by number of answers divided by money per hour implies 6 hours per question" 5) It fails to mention let alone answer the central claim, which is that it is very valuable ... – ChrisW Dec 6 '13 at 18:53
  • ... because it has allegedly been of service to (e.g. saved time for or increased productivity of) an average of 450 viewers per question. – ChrisW Dec 6 '13 at 18:55
  • That's just predicating the answer on a bad definition of value which I reject. You may as well assign that value of the additional 449 viewers to Cisco, the Internet, Google, or Tim Burners-Lee. – Evan Carroll Dec 6 '13 at 18:59
  • I would never suspend a user were I personally involved, but vote to delete is a 10k privilege, and certainly not a diamond moderator power. – Sklivvz Dec 10 '13 at 9:30
  • @Sklivvz FYI users with 20K+ privilege can vote to delete a down-voted answer. The only users on this site with 20K+ privilege are moderators (except Oliver_C who was last seen 8 months ago). Is "Flag for moderator attention" an option you could use? – ChrisW Dec 10 '13 at 13:21
  • right sorry, 10k is for questions, 20k for answers. In any case, it's not a diamond moderator power. – Sklivvz Dec 10 '13 at 15:21
5

I would have deleted your answer as well if I were a moderator of the same reasons that were outlined by the others. This really is a research level question that someone could get a solid paper and maybe even a thesis out of and your answer is very hand-wavy and mildly editorialized (i.e. "Jeffs bureaucracy") in way that doesn't really add much to the answer.

At a minimum a solid answer to this question is going to need someone that can apply standard economic methodologies and justify their use since the value of stackoverflow.com is in the long tail of questions which means that you can't even assign the same economic value to each question since an answer to an especially hard question will have greater value than "How do I sort a list of strings in Python?"

  • 1
    This is nonsense. Where in the /faq or site purpose does it exclude the audience of this site from answering a research level question? What kind of elitist terminology is that? What economic methodology do you want applied? And, when is a specific method and not sources and transparency the way the skeptic community works? Perhaps he does want a more accurate thesis written on the topic, but that isn't what we do. I worked with averages, and he didn't explicitly ask me to offset for difficulty of questions or labor-price of different languages. – Evan Carroll Dec 6 '13 at 18:40
  • @EvanCarroll When I say "research level" question I mean that you aren't going to be able to find a suitable answer just by doing a review of the literature that is out there. Rather, the question itself is complicated enough that answering it require a significant amount of work on the person answering the question and that amount of work is likely enough for a publishable paper. With regards to the economic methodology, it's not my responsibly as the challenger to tell you which one to use, it's your responsibility as the author to justify your methodology. – rjzii Dec 6 '13 at 19:51
  • @Articuno This question strikes me as an edge case. As such, if someone were to write a journal quality answer and post it I would be willing to accept it if they also provided evidence that it is a novel question. – rjzii Dec 6 '13 at 20:23
  • @Articuno It's an edge case because at this time it really does look like a novel question. Journal quality is just want it sounds like, of sufficient quality to be published in a reputable peer-reviewed journal. – rjzii Dec 6 '13 at 20:49
  • @Articuno define "original research" those numbers come from the primary source over their domain (direct in StackExchange, government for labor statistics). You're just throwing around more baseless libels. – Evan Carroll Dec 6 '13 at 21:00
  • @Articuno Edge cases in the technical sense are always going to pop up from time to time. From a philosophical standpoint, I'd rather take a well researched, journal quality answer that is original research over waiting for someone to publish and perhaps not having an answer at all. – rjzii Dec 6 '13 at 21:23
  • @Articuno I never said that nor did I imply that. – rjzii Dec 6 '13 at 21:25
  • @Articuno If you have gone through the academic hazing that is graduate school then you get a feeling for what "journal quality" is without having to articulate it. In short, it's something you wouldn't be embarrassed to show others and meets all of the criteria for a solid piece of work. There are manual of styles circulating that spell out exactly what this is. – rjzii Dec 6 '13 at 21:28
  • @Articuno that doesn't make it so. Take for instance, "that the opportunity costs that you list are relevant" is not a critique of original research. It's a just a snarky adolescent insult akin to "well, prove that matters." If I could otherwise be doing something else, it certainly matters. That's why "opportunity cost" is a recognized concept in economics. – Evan Carroll Dec 6 '13 at 21:29
  • @EvanCarroll I do mean "Well, prove that matters." – user5582 Dec 6 '13 at 21:30
  • 1
    @Articuno At this point you are quibbling. – rjzii Dec 6 '13 at 21:30
  • @EvanCarroll You must not have sat through a peer-review, the way Articuno spells things out is actually better than I've seen some others do it. :-/ – rjzii Dec 6 '13 at 21:31
  • @rob I know you don't use chat, but since I agree that this is secondary to Evan's question, I wrote my position here: chat.stackexchange.com/rooms/11839/original-research (I've deleted my comments to you in this thread.) – user5582 Dec 6 '13 at 21:42
3

This answer is primarily unreferenced. Here are the things that would need reference if I would support its undeletion:

  • That $209 per question is nonsense (if referenced this would answer the question, but currently, it's begging the question)
  • That the opportunity costs that you list are relevant
  • That there is such a thing as "Jeff's bureaucracy"
  • That Stack Overflow is a leviathan
  • If someone answers a question and it doesn't help you, it just increases the cost of the site to industry and offers no value in exchange
  • The 33% of questions that are unanswered cost the industry something
  • Marx agrees
  • in the latter case the difference has social value
  • Let's not pretend like none of these questions are economically useless

Basically, you have used 2 references to support an answer full of speculation and unreferenced assertions. I would have deleted this answer also.

A comment asking for a reference was the most upvoted comment, so don't act like you didn't know the reason for criticism, or "presume" that the deletion was because Sklivz didn't like the answer.

You're right that he can't dictate the type of answer that comes out, but answers still have to satisfy the standards of evidence required by this site and by skeptics.


Sklivz does disclose his employment at his profile page and can be seen if you hover over his avatar.

  • That $209 per question is nonsense. That's kind of the point. To say that the site is worth a billion would be to state that each question is worth $209, which is surely entirely arbitrary, but also statistically unlikely. You seem to have a problem with words, but it wasn't closed for that. – Evan Carroll Dec 6 '13 at 17:52
  • 1
    Sklivz does disclose his employment at his profile page. It's not disclosing, in any useful fashion, if you have to look for it. – Evan Carroll Dec 6 '13 at 17:54
  • That's not "begging" is explicating. The value of 1E9 is removed from what people can mentally wrap their heads around to begin with. It's also attached to a large black-box. I reduced that billion dollar value to the price-per-average-question, and then I tied it back into the labor-cost by the audience asking the question to reduce that number again to hours. That's what each question has to save to be of value (assuming 0 in the way of offsets). I even defined the problems with the terms value in that context. – Evan Carroll Dec 6 '13 at 18:01

You must log in to answer this question.

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