Negativity specific to CR [closed]

Question

First of all, some simple statistics.

codereview.stackexchange.com

26490 - questions asked
345 - rated negatively
1.3% (negative of the total)

tex.stackexchange.com

94800 - questions asked
150 - rated negatively
0.16% (negative of the total)

I counted pages, this is why the number of questions is a multiple of 15 and 50

In absolute numbers this doesn't sound like much, but another way of seeing this is to see a tenfold increase compared to another similar site.

---

Problem

I participate in several other SE sites, and until today I only had this vague feeling that TeX is kinda welcoming and Codereview is not so much. I had my hunch as to what may be causing it. I don't visit CR frequently, but the general impression that I had was that, quite often the answers may be even worse than the question they were improving. I also encountered several instances of moderators making absolutely bizarre and irrelevant changes to questions, especially posted by novice users. This becomes particularly strange when someone with no expertise in a language of the subject edits the question to best match his (incorrect) intuitive understanding of what the question is asking.

All in all, my impression was that of unscrupulous and harmful attitude resulting in "Lord of the Flies" effect (aka Bystander effect). The typical scenario when the individual is harmed goes like this:

1. The "victim" posts a question which to a non-expert looks as deserving bad karma.
2. This opens the gate for other non-experts who hesitated to vote before.
3. "Victim" with high probability never posts again.

To prove my point by ways of example, I found an interesting problem, which seems to fit my scenario very well. It sounds almost trivial, if you don't pay close attention to the problem statement. And the solution isn't really that hard to come by for an expert. And I found it reading a CS coursebook, which is almost mandatory reading for any CS graduate.

The aftermath of my test: the question is closed as "off topic" / "broken code", while it's perfectly on topic, and the code actually works! It received three good answers, one of the answers is really in-depth, and with good measure of understanding. And it was downoted at least twice. So, someone who is, apparently, an expert in the matter took his or her time to lay out an extended answer with diagrams (note that this person has very few karma points), and people who voted on closing the question are proudly displaying their achievements in SE careers profiles! :)

---

Possible solution

First of all, I believe that the reason for this kind of behavior is psychological and has very little to do with the expertise level. So, the solution needs to be psychological too. As a limited time experiment, I would suggest that CR would change the "rewards" for some activities. Most importantly, downvoting the question must deduce twice as many points to the first person downvoting the question compared to the person who asked the question. Next person downvoting the question will restore some points to the first person downvoting and will only sacrifice, say, 1.5 times as much as the original poster. The whole idea is to balance out by the time there are at least 5 downvotes. The person who "casts the first stone", however, will need to think very carefully (if they care about karma).

I didn't do the math, but if you are interested, I can do it.

Answer 1

Nice rant. My turn now.

---

"Perfectly on topic"

The aftermath of my test: the question is closed as "off topic" / "broken code", while it's perfectly on topic, and the code actually works! It received three good answers, one of the answers is really in-depth, and with good measure of understanding. And it was downoted at least twice.

The question is not "perfectly on topic". It's very much on the fence, bordering unclear what you're asking. I agree that the close reason that was picked, doesn't reflect that. The system isn't perfect, but it works for the vast majority of cases. I'm sorry your particular question wasn't in that majority, except as Zak mentioned, "I'm posting this in good faith that my code actually works" sounds very much like you didn't bother testing whether it actually works or not: turns out, asking "Will this code work?" is blatantly off-topic on Code Review. But you saw our Help Center and already know this, right?

This is a good regex question. This is another one. This is yet another. What's the common difference? The regular expression is shown in full context, and that context allows reviewers to go as far as recommending a different approach than using a regex pattern, if that's appropriate; we know what the regex is used for, and these questions all have an appropriate language tag. Regular expressions are a tool, not a language. Depending on the language you're using them in, the implementation changes and some things aren't supported - regex is pretty much like the sql tag: without another tag to further specify it, it's pretty vague.

Every single top-voted regex question includes a language tag.

---

"Unwelcoming"

I participate in several other SE sites, and until today I only had this vague feeling that TeX is kinda welcoming and Codereview is not so much. I had my hunch as to what may be causing it. I don't visit CR frequently, but the general impression that I had was that, quite often the answers may be even worse than the question they were improving.

I've been a regular (read: addicted) member for over 2 years, and I'm not even sure I've seen more than 2 or 3 users (including you) making that remark. This is what we're seeing:

ckuhn203 Aug 5 '14 10:23 AM I think I've learned more here than anywhere.

Alex L Jan 5 5:54 AM I've learned more than I ever have by reviewing

CodeX Jul 21 '14 3:36 PM Since I've been here I have learned more about PHP than in the 4 years of googling

JaDogg Nov 6 '14 8:20 AM I learned more in the time I participated in this site(100 days) than what I've learned during a 3 year degree.

user3344977 Jul 18 '14 12:38 AM @nhgrif Thank you for all of the help. I was just telling someone the other day that I've learned more in the past 2 weeks posting on codereview then I have in the past 2 months.

syb0rg Feb 17 '14 8:05 PM @DavidMcDavidson If you do stick around here, you will find that you can earn a lot of reputation very quickly, much more so than Stack Overflow.

Marc-Andre Feb 17 '14 5:22 PM Indeed, CR helps a lot. I'm reading a lot of code, review. And doing a lot configuration of projects and others. My first year so fare has been very good so far! I've learned a lot!

And it goes on and on and on and on. And I haven't even mentioned how The 2nd Monitor (the site's main chatroom) is perceived.

I think you're getting the wrong impression, out of your own personal frustrations and misunderstandings of what this site is all about.

---

/mod-hat off

I'm not going to address your other "points" and the other boiling comments written in frustration, because they are not making you look good at all. Just know that Jamal and 200_success have been pro-tem moderators in this site for a long while, and they were recently elected by a crushing majority of community members - and the recent CR election was the biggest election to ever happen for a graduating beta site: our mods know this site and its rules inside out, and are deeply trusted by the community. If you have a problem with the rules, state what the problem is respectfully, don't start shooting dirt in all directions - that's not how effective communication works.

Answer 2

This question was, in my opinion, improperly closed, because it could have been an allowable question with some minor changes. I applied those minor changes and reopened the question. You rolled it back. I've closed the question again.

As for extrapolating this closure into general accusations of negativity… that's not appreciated. Please focus on the problem at hand rather than ranting.

I was going to explain more about… but oh screw it. I have a meeting to attend right now.

Answer 3

Speaking as somebody who joined CR very recently and didn't do a great job finding out all the rules beforehand, I've found the community to be very welcoming.

We have relatively few rules, and as a result we are quite strict about enforcing them.

• The code must be your own
• The question must be clear and actually be a question (even if it's just "how good is my code?")
• The code must be working as intended

I think the last point is where you fell afoul of the CR community.

A statement like

I'm posting this in good faith that my code actually works

Says to almost everyone reading it "I think this'll work but I don't actually know".

After reading your question for the third time, I finally saw how you might've meant "My code runs, but I'm not sure it is actually catching all the cases it should".

Your question is difficult to distill a clear question out of.

I'm posting this in good faith that my code actually works, but I lack the confidence. Plus, if there is an easier way (the diagram for this DFA is really simple), I'd like to know about it.

We are not mind readers. Easier way to do what, exactly?

CR is moderated by the community at large. To the vast majority of people who are not trying to read extra meaning into your question, it was relatively unclear, and sounded like you didn't know if your code worked or not.

So we voted to close it, like so many other questions we get which are unclear, or break the simple requirements we have.

In this particular instance, that may have not been the case, but can you really blame us?

Answer 4

To throw some more accurate numbers into the mix, here is the query for the following results.

As of 1 Sept 2015 14:26 UTC-4, there were the following breakdowns for questions (on SEDE, which is always behind):

Code Review:
26507 Total questions
24876 With a score of 1 or more (93.85%)
1289 With a score of 0 (4.86%)
342 With a score of -1 or less (1.30%)

Stack Overflow:
10030251 Total questions
4699034 With a score of 1 or more (46.85%)
4773153 With a score of 0 (47.59%)
558064 With a score of -1 or less (5.56%)

TeX - LaTeX:
94592 Total questions
86109 With a score of 1 or more (91.03%)
8352 With a score of 0 (9.70%)
131 With a score of -1 or less (0.14%)

Database Administrators:
36706 Total questions
24759 With a score of 1 or more (67.45%)
11089 With a score of 0 (30.21%)
858 With a score of -1 or less (2.34%)

Christianity: 
7220 total 
6253 With a score of 1 or more (86.81%)
427 With a score of 0 (5.91%)
540 With a score of -1 or less (7.48%)

So, with all of this in mind, let's compare them:

Stack Overflow and Christianity have a much higher (3x+) percentage of questions downvoted to the negatives than Code Review. Database Administrators has a higher (almost 2x) percentage of questions (and over 2x number of questions) downvoted to the negatives than Code Review. TeX - LaTeX has almost 1/10 the percentage of questions downvoted to the negatives than Code Review.

So, when comparing Code Review to the other SE sites (one of which you picked the other two of which I picked at random, one of them being the pilot site for SE), it's really irrelevant how many or what percent of questions are downvoted to the negatives. By your logic, the DBA site is much less friendly than Code Review, and the SO site is even less so (though we all may be able to agree on that).

If we look at the number of questions which are voted non-zero, though, we have a whole new story. In fact, this changes everything. Only TeX - LaTeX and Code Review have non-zero question percents that are less than 10% out of this list. In fact, less than 5% of Code Review questions are voted at 0. Meaning, the Code Review users vote a lot. We aren't afraid to vote.

You describe the problem as if it's some conspiracy, or as if we are trying to downvote questions a lot. In reality, the question just happened to be a sub-par question according to the standards. We can't be assumed to know whether or not your code works. So when you say something to the effect of, "I haven't really tested it, so I can't vouch for it" in a question, you should expect the downvotes to pile up. And guess what, this isn't the first time it happened, and it won't be the last. The problem isn't that Code Review is unwelcoming, the problem is that not all questions are good.

If I went to Stack Overflow and said, "I have a Regex that isn't working" and simply posted the Regex with no information about what I consider "working" to be, you can bet the house on it that the downvotes would pile up immediately.

This is effectively what you did here, just for a different topic.

So, by your logic, should I send Christianity this message?

---

More data!

Just as well, here's a query that will rank all the SE sites in order of the percentage of downvoted questions.

Code Review ranks 100 out of 147. This means that only 46 SE sites in that query have lower percentages of questions voted below zero. And 99 of the SE sites have higher perecentages of questions voted below zero. And Workplace has 12%+ of their non-deleted questions voted below zero. So, is Code Review really that far off?

---

My point by all of this is not that there are "worse" sites in the SE network. It's that you cannot just take numbers (which are meaningless on their own, not to mention when together) and put some "spin" on them to make this community look like a bunch of bad guys. Everyone (or at least, almost everyone) is here to help each other. We're not here to put people down, throw mud, or insult each other. We're here to help each other improve. If you are not, then you can continue to expect backlash like this.

Answer 5

First of all, some simple statistics. (...)

I participate in several other SE sites, and until today I only had this vague feeling that TeX is kinda welcoming and Codereview is not so much.

I understand that you have good intentions by posting statistics, but there are a couple of problems with statistics.

First of all, the statistics does not include deleted questions. If we would delete all negatively voted questions, would we then be 100% welcoming? Of course not. Unfortunately, there's no easy way to find out how many questions each site has deleted.

Secondly, the votes are not everything. I am personally a big advocate for comments. Unfortunately, your question did not get any good comments about its topicness. I admit, that's one thing where we could have improved the handling of this question. The off-topic reason that was chosen, 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. is likely not explanatory enough in this case.

Thirdly, quality. We get a whole lot of questions about "How can this code be improved?" that contain very little to no pure-English explanation about what the code does. Such questions, while on-topic, are not good quality. There are plenty of such questions on Code Review, and I believe most of those questions have a comment suggesting the asker to add more information about what the code does.

So statistics don't tell the whole truth.

---

I had my hunch as to what may be causing it. I don't visit CR frequently, but the general impression that I had was that, quite often the answers may be even worse than the question they were improving.

Depends on what you mean by "worse". Answers on Code Review may touch on any or all aspects of the code. It's hard to ask reviewers to not focus on something.

Answers on Code Review might not always be about what you want, but it is what other users feel that you should hear.

And if you see an answer that you feel is bad for any reason, feel free to down-vote it, add a comment, and/or ask for opinions on the answer in chat

---

I also encountered several instances of moderators making absolutely bizarre and irrelevant changes to questions, especially posted by novice users. This becomes particularly strange when someone with no expertise in a language of the subject edits the question to best match his (incorrect) intuitive understanding of what the question is asking.

There is no such thing as bizarre and irrelevant changes. There are either improvements, or non-improvements. Be aware that not even suggested edits are anymore rejected for being too minor - that rejection reason has been deleted.

There are some users which are known to edit a lot of things on the site. They do this with good intentions. If you find a bad edit, provide a link to the bad edit so that we can discuss that specific edit.

---

The "victim" posts a question which to a non-expert looks as deserving bad karma. This opens the gate for other non-experts who hesitated to vote before. "Victim" with high probability never posts again. To prove my point by ways of example, I found an interesting problem, which seems to fit my scenario very well. It sounds almost trivial, if you don't pay close attention to the problem statement. And the solution isn't really that hard to come by for an expert. And I found it reading a CS coursebook, which is almost mandatory reading for any CS graduate.

Okay, several different things here.

The general "problem" about massively down-voted questions is not specific to Code Review. This is something that applies to all of Stack Exchange. Meta Stack Exchange has a whole bunch of questions related to down-votes.

"Victim" with high probability never posts again.

"high probability" is very vague. Neither you nor I can say the exact percentage here. There is a risk that they never post again, yes. Which is why we prefer to add comments. I am convinced that it is possible to be friendly, while still down-voting and saying "this particular question does not meet our quality standard".

And I found it reading a CS coursebook, which is almost mandatory reading for any CS graduate.

Not sure if this is completely relevant to the point you are trying to make here, but either way. Not all users on this site are CS graduates. In fact, I know very few who actually are. And either way, you shouldn't need to be a CS graduate to make use of Code Review.

---

The aftermath of my test: the question is closed as "off topic" / "broken code", while it's perfectly on topic, and the code actually works! It received three good answers, one of the answers is really in-depth, and with good measure of understanding. And it was downoted at least twice. So, someone who is, apparently, an expert in the matter took his or her time to lay out an extended answer with diagrams (note that this person has very few karma points), and people who voted on closing the question are proudly displaying their achievements in SE careers profiles! :)

As mentioned by 200_success, the close reason was indeed not the best one. The specific problems with your question is:

• It is a regex-only question (note the answer by 200_success).
• It lacks context. It doesn't say why the regex is needed or how the regex is used.
• It lacks examples of strings that should be allowed or denied.
• It is slightly unclear, as evidenced by rolfl's comment "what's wrong with ([01][01])+"

---

As a limited time experiment, I would suggest that CR would change the "rewards" for some activities. Most importantly, downvoting the question must deduce twice as many points to the first person downvoting the question compared to the person who asked the question. Next person downvoting the question will restore some points to the first person downvoting and will only sacrifice, say, 1.5 times as much as the original poster. The whole idea is to balance out by the time there are at least 5 downvotes. The person who "casts the first stone", however, will need to think very carefully (if they care about karma).

This is an interesting idea that is not possible for Code Review to apply by ourselves. This would have to involve some implementation from Stack Exchange. If you really want to see this happen, the best thing to do is to post a feature-request on Meta Stack Exchange. Before you do that however, think about a few things:

• Will it really prevent massive down-votes? If I am the single down-voter on a question, it will affect me quite a lot. But the more down-voters, the less impact it will have for each down-voter, so it is essentially giving down-voters incentive to look for more down-voters.
• Be aware that down-votes on questions used to cost reputation, but was changed to be free