@YannisRizos - When people are upset that questions were closed and they fear they might be deleted you say it's just to prevent new answers. When people are upset that closed questions were deleted you say that's what happens to closed questions that aren't improved.
@psr All closed questions are on a path to deletion. On ProgSE, we tend to avoid deleting closed questions that add value to the site, regardless of whether they fit our guidelines or not. If you have evidence to the contrary, Meta is where you should post it.
@psr Where's the contradiction? Closures themselves are nothing more than a way to stop further answers, what happens to a question after it gets closed is a different issue. Some will certainly be deleted, some will be re-opened and most duplicates will be merged. "not worth having around" is only relevant to the first category.
...and it's not like we don't have quite a few years old closed questions around...
...but if it's guarantees you are looking for, then the only guarantee that a closed question won't be deleted is getting it re-opened.
@psr When you get to 10K you'll get access to a list of recently deleted questions. At that point you'll realize that we only delete crap. The only time that questions that weren't necessarily crap were deleted was during STCI, but those deletion candidates were advertised on Meta for at least two weeks prior to deletion (and if I remember correctly, only @Rachel stepped up and salvaged a few).
@YannisRizos - During STCI I didn't salvage any because Mark Trapp's criteria for what would constitute salvaging were incomprehensible. What was wrong with them in the first place was frequently incomprehensible.
@WorldEngineer: That's actually a good point. // Although I've said that at least half a dozen times that if @Rachel and @YannisRizos went out for a beer (in the real world), they'd walk away with a newfound appreciation for each other.
@WorldEngineer: I understand why moderators are expected to uphold consistency; but I also know that Ralph Waldo Emerson said, "A foolish consistency is the hobgoblin of little minds." // Even if Stackexchange wanted most sites to be strictly Q&A, J think that it could carve out a niche for more permissive sites, which is why I'm often sympathetic to @Rachel's point of view.
@JimG I'm going to jump in on your previous discussion if you don't mind... you have to remember that SE is a network for expert Q&A. That doesn't mean that beginners aren't welcome but it means that the network is trying to be one of the most reliable source for info on everything. Subjective questions don't have any home here because there is no real answer. That's why you don't see "What language should I learn next?" In an Encyclopedia.
It's not that I'm against the questions, but that's why forums haven't completely died out...
@Dynamic: I really can't disagree with anything you just said.
@Dynamic: During 2010-2011, I feel like many borderline questions were abruptly and unilaterally closed, and I feel like that sucked some of the life out of Programmers.SE. For instance, some very excellent users from that time period no longer contribute here.
Fortunately, the current roster of moderators try to rehabilitate borderline questions. They understand that abrupt closures discourage new users.
But I'm often sympathetic to @Rachel 's point of view because I definitely shared her concern about the abrupt closures (approximately 1.5 years ago).
@JimG Abrupt closures are sometimes needed to make a statement: "This question is NOT going to be dealt with on the site". I feel like we (as a community) do a good job of sending the message when needed and letting the questions play out if we aren't positive.
Number of folks who need to be hit by a bus before the system is effected
granted lacking gnat it wouldn't have a significant effect, but lacking gnat, walter and chrisF the effect would be VERY significant on the way the front page of P.SE looked in regards to close votes
Or maybe it wouldn't. Maybe others would pick up the slack because enough others agree with them
But we will never know if they are a minority in opinion on what should be closed or not without abstinence from them for a test which won't happen I don't suspect. Not a huge deal frankly, the site isn't bad necessarily right now, I can still enjoy it. I just think it might benefit a greater deal of people more than the 3 lost if they abstained from executing their opinion upon the site.
@JimG. Honestly it was like an hour or something after I had a protracted argument with Mark Trapp on Meta that he left. I know people seem to miss him but I feel proud and that I may have been the straw that broke the camel's back effectively benefitting the site as a whole
@JimmyHoffa Two mistakes: 1) Closure stats are pointless if you don't count deleted questions (and you can't), 2) You're assuming that people who don't vote to close disagree with the closures. For everyone else but yourself that's pure speculation.
If you want to influence the site more, all you have to do is... do more. If you disagree with a closure, try and get the question re-opened. Everything else is just a waste of time.
Also, I wouldn't be surprised if our more active close voters are also our more active re-open voters... Some people are just a bit more active than others, and I have no idea why you've chosen to focus on the one activity that you perceive as negative, and are ignoring everything else.
@YannisRizos: Right - A devil's advocate would say, "If @gnat and @Walter hadn't voter to close question 'A', then somebody else would have." - And I actually buy that argument. // The unilateral closures are a different story; but like I said, I don't think that's a problem for Programmers.SE right now (although it may have been at one time).
@JimG. Also don't forget that the query shows successful close votes, being at the top of that list is a very good thing, it means that your instincts are mostly right (as normally it takes 4 others to agree with you before your close vote is successful).
@YannisRizos But as an example, run the query on some of the other smaller SE sites. You'll see different trends in the range of numbers in the top 10 close voters (Just need to type a SE URL in the box below the "run query" button)
I like to use Data.SE to view usage stats for some of the sites, however deleted posts are not included on Data.SE and I think this skews the numbers quite a bit, particularly on sites with a lot of deletions.
Would it be possible to include some limited data about deleted posts in Data.SE?
@JimmyHoffa Of course. For example, I've deleted 900+ questions this quarter, and I've closed more than half, that's about 450 successful close votes that are missing from your query (and only for the past 4 months).
@Rachel what do deleted posts tell us in the scope of who makes most close votes? If most folks doing close votes now were doing more with the delete votes it just shows the same thing, an extremely active minority, if it shows other users coming up to par with the same folks then you could say the minority who are actively close voting more than everyone else are doing so disproportionately on questions not bad enough to be deleted
I find it interesting that the top 3 folks closing questions participate in asking and answering them very little (way down the list in number of Qs/As). Would make it seem like they may be taking more away from the site than they're contributing. I actually expected to see them pretty high up there. That said, my query may be wrong.
You may be archiving old questions by closing them to keep people off of them, but when someone wants to ask a question on a related topic, they start writing the title, see the closed question and presume well dang they don't want this content
However that said, a great majority of closed questions are low quality for the site, and probably don't belong. But in that case I think we need to improve the way we market our site and introduce ourselves to new users
@Rachel New questions asked is terribly skewed because of deletions, and the answer count going down is actually a good thing. Actually, it's absolutely fantastic, there's nothing worse for the site than those crap questions with 20+ (almost) one line answers that said more or less the same.
@YannisRizos .... See this sort of attitude is why I had to stop participating with programmers. I can't argue with people who ignore stats, evidence, and users that try time and again to make the same argument against the system and just get sent away by the crowd of established meta users
@Rachel Hm, what? What am I ignoring? 99% of our questions are answered, and we've maintained that percentage for as long as I remember. The fact that we dropped from 8 answers per question to about 5 is a very good thing.
This isn't a forum, 7-8 answers per question was not healthy...
I agree with @Rachel. Yes, "the audience that remains has a higher percentage of experts in the field" - however, this trend may end up with a closed circle of experts who have nothing to ask, thus nothing to answer... :-( — Péter TörökFeb 20 '12 at 10:44
@YannisRizos Oh, I often don't bother using the "reply" arrow, I just type @Ya[tab] so it "replies" to the most recent message of yours
Well I'm off. I wish someone with access to the full database and a knowledge of stats and what makes a healthy Q&A site for the internet would run some queries and adjust some things accordingly, but oh well
I know this is coming at the discussion fairly late but @YannisRizos is right, you're focusing on one thing that you think is "bad". My bet is if you checked the data you would find that most of those mean close voters :) are also on the top of the list for editors as well...
I would also speculate that those top close voters are also the top commenters
Just saying you could be looking at a ton of different activity other than just focusing on close votes.
That's the minority opinion I'm trying to show statistics prove. The majority opinion is clearly otherwise considering how few people do close vote, and based on posts/comments/edits you can't call those users who aren't close-voting disengaged
I've been playing with R recently, so if you want me to try to make pretty graphs or do computations and get something that is useful in a CSV format, let me know and I can make pictures or do stats on it.
I think most people other than the minority recognize close-votes have two negative effects: They remove content which is potentially beneficial to 1 person (the poster) and possible many others (yes close doesn't delete but as Yannis said it will be) but more importantly than that: Close votes discourage participation by making clear that participation will be punished unless you're super careful.
Go ahead heh, could be fun for you for sure, see if R can give you greater insights into the data.
Generally questions that are deleted are those questions that are not useful for the site where they are asked, such as off-topic, not constructive, not real, and too localized questions.
Duplicated questions are generally not deleted because they can help who is searching the duplicated questio...
To quote: "Over time, closed questions that are not useful as signpoints to other questions may also be removed". Duplicates pointing to open questions are useful as signs and shouldn't be deleted. /cc @YannisRizos
Because of the count, if they've never voted for a closure, it'll still say 1 because that 1 record it's counting is the User record, while the Left join on post history isn't multiplying their user record to count more than 1
@gnat So I was apparently wrong all over the place and the vote invalidation does trigger on deleted posts. Don't take it personally - the script is dumb and can't tell your intent or judge the quality of the posts. At least in this case the posts were improved? :)
@Walter I participate a fairly bit I believe. And I admit the query may not be 100% correct, it's a bit of a complex query off hand and most of it was written late at night heh. I don't show up in the queries because my close-vote ability came less than a week ago and these statistics are over the course of a year.
@MichaelT yeah, interview question I've used is design a poker game. Hand Evaluator is just one piece but yeah that would be decent
I believe the poker hand evaluator has enough leeway in its design that each language can implement it in a way that showcases that language... other ideas for other problems that would work similarly?
You're thinking of stuff that will favor calculation strengths of languages
Solvers are going to have less display of data modelling strengths
they will do a poorer job displaying IPC
Ruby has a huge weakness with concurrency in lots of implementations
I recall reading a bit on some AI challenge and the guy was going to do it in Ruby... and fought with it alot... then switched back to java and had it running quickly the way he wanted to without fighting.
I'm more thinking about language/program structure. "Look at the neat things we can do with enums... did you know java had an EnumSet that works as fast as a bitfield?"
the leaderboard stuff I like because it requires calculating the games score to come up with the winner which shows calculation, there's plenty of data modelling, and the board itself would be IPC updated/accessed
and it's very simple to understand and do
could involve transactionality wherein you show WCF's distributed transaction abilities
@JimmyHoffa I never said that. I said all closed questions are on a path to deletion, but they can also be: re-opened, merged (dupes), or even stay around closed for quite some time. Not all closures will lead to deletion, but the only guarantee that a closed question won't be deleted is getting it re-opened (except dupes).