# Let's require registration to ask a question

More than two years after http://meta.math.stackexchange.com/q/4569/, this may be the time for reconsideration. A brief update:

1. In July 2012, when the aforementioned discussion took place, the site received $11000$ questions. In October 2014, the number was $41000$. (Source) [In addition to the above numbers, about $500$ questions asked in July 2012 have subsequently been deleted, compared to $2600$ from October 2014. (Source: special moderator search abilities.)]

2. In 2012, only Stack Overflow disallowed questions by unregistered users. By now, several other sites followed: Physics, Electrical Engineering, Programmers, and Ask Ubuntu.

3. Unregistered users sometimes lose access to their accounts after asking. So they lose ability to edit or comment on their questions, which creates noise: updates and comments posted as answers, or a new identical copy of the question posted. All this takes the time and energy of reviewers, editors, and moderators; and the users themselves are not happy either.

4. Split accounts are a known problem for the closure of exact duplicates.

5. Unregistered users cannot delete own posts. So instead they deface or blank it, leading to chaos: votes to close as unclear, rollbacks, votes to close with custom reason "OP wishes to delete", votes to delete, flags for moderator attention.

6. The quality... well, you know. Back in the days, Qiaochu Yuan wrote

StackOverflow has less of a need for growth and more of a need to improve the quality of content, but I don't think we're there yet.

I think we may be there now. Registration requirement will not magically improve the quality of questions, but if there are fewer of them overall, the arriving questions can be attended to more thoroughly.

• My gut instinct is that that simply puts more barriers up to people needing help - we already ask for beautiful formatting and brilliant explanations of work done, and this is just something else for them to get tangled up in. Indeed, the vast majority of my students speak English poorly, and I worry that this would further hinder them asking question here (which I constantly encourage them to do). However, one of my students finally plucked up the courage to ask here. And they registered! So I have no issue with this proposal - if my students can register, then noone else has any excuses! – user1729 Nov 7 '14 at 9:27
• (Incidentally, the student in question registered and asked a beautifully formatted question...which was closed for "missing context or detail". My students have much to learn...) – user1729 Nov 7 '14 at 9:29
• Or perhaps those voting to close questions have something to learn. – Gerry Myerson Nov 7 '14 at 10:02
• @Gerry I am not going to deny that it was interesting being on "the other side", as it were, of the closed question line! The question my student asked was simply a problem statement: "Do this". However, knowing the context -knowing what I had told them- I don't really know what effort the poor student could have shown! (It was a deliberately hard question from a mock exam, and I was feeling particularly grumpy when I set it.) – user1729 Nov 7 '14 at 10:12
• I agree with this proposal. I hope that requiring registration would reduce the number of abandoned questions, which is a notable source of low quality content. People who have lost access to their question can also cause a lot of confusion and unnecessary work for others. – Joonas Ilmavirta Nov 7 '14 at 11:34
• I think we are certainly at the point where we need to focus on quality, not quantity. I am not sure where registration will just lead people to register more throwaway accounts, though. It would certainly simplify the other issues listed in the question. – Carl Mummert Nov 7 '14 at 12:03
• Could we get some statistics on the number of unregistered user questions vs. the number of registered user questions? – Thomas Nov 7 '14 at 13:22
• I support this. The problems have gotten worse, and this will alleviate them. If the asker cares about either the site or math enough, then registering is not a major obstacle. If the asker does not care for either, then I'm not sure they are entitled to answers. – Jyrki Lahtonen Nov 7 '14 at 13:58
• @Jyrki: The only person entitled to some answers is Lt. Daniel Kaffee. And he wants the truth! – Asaf Karagila Nov 7 '14 at 15:57
• Regarding "getting tangled up" @user1729 I think the much bigger risk is 3, 4, 5 in OP. I can easily see somebody asking quickly and then loosing access and being confused. By contrast it seems pretty hard to imagine somebody intending to ask things on the internet that will not manage to sign up. I mean all kinds of people manage to use all kinds of internet things that requier some sign up. – quid Nov 7 '14 at 16:43
• What I remember from other cases where this was proposed is that SE has not observed any increase in question quality after enabling mandatory registration. It might be worth it to lower the support burden for users that have lost control of their account, but I would not expect this to solve any quality problems. – user9733 Nov 7 '14 at 21:40
• While I don’t feel strongly about it, I am mildly opposed. I don’t expect it to have significant positive effect, and it strikes me as being yet another manifestation, albeit a mild one, of the growing atmosphere that caused me to stay away for almost a year. Frankly, if I were still teaching, I would be increasingly hesitant to send any but the best students here to ask questions. – Brian M. Scott Nov 8 '14 at 17:04
• @Brian: I too go back and forth about this issue. But registering for a website you expect to use more than once to post your homework, that's not a big deal. On the other hand, confused users that don't register and then post several comments as answers because they keep losing access to previous accounts; or users that have several unregistered accounts making it harder to track their previously asked questions (which is something I know I do when I notice people asking several questions); these are problems that will be easily solved that way. – Asaf Karagila Nov 8 '14 at 18:02
• @ArthurFischer I take this answer by Shog9 as "will happen, but not before the login system is redesigned". Fine with me. – user147263 Nov 19 '14 at 4:04
• Happy birthday to you, happy birthday to you! Happy birthday, dear let's require registration to ask a question feature request thread on meta.MSE, happy birthday to you! – Asaf Karagila Nov 7 '15 at 16:18

It's... uh... been a while since this request was made. We require a few things before turning a switch like this - it's pretty harmful to small sites to have what is a large chunk of potential askers to be blocked. But... Maths isn't a small site, and it hasn't been one for a loooong time. So when this request was first passed into my radar two years ago, it was pretty quickly approved.

Just... might've... forgotten to ask the right people to flip switches. And it didn't get into my radar again until this past weekend.

Sorry about the delay, but without further ado, these switches are being flipped as I'm writing this answer - registration will now be required in order to ask a question.

• Praise be! Hope this makes things easier for mods and active users alike. – J. M. isn't a mathematician Oct 10 '16 at 17:42
• HALLELUJAH!!!!! – Asaf Karagila Oct 10 '16 at 22:49
• Are there other feature requests, widely agreed upon, buried somewhere in the meta site, and not enacted for, say, more than a year? – Did Oct 11 '16 at 6:03
• So, did this actually happen? Has it had any effect? – Gerry Myerson Nov 3 '17 at 22:41
• @Gerry: That's a very good question, but it should be asked separately so that it can be answered properly and not in comments. – Asaf Karagila Nov 4 '17 at 0:36
• @Asaf, done – math.meta.stackexchange.com/questions/27304/… – Gerry Myerson Nov 4 '17 at 5:48

I think this comes back to the fundamental question of how people here envision the purpose of this site, and if the history of such debates is any indication, addressing the issue will lead to acrimony, drama, accusations, frustration, and other uncivil behavior on the part of certain individuals...over what I would like to remind people is ultimately just not that important.

Online forums tend to attract certain personalities that, for whatever reason, seek to carve out their own sphere of influence in whatever area of expertise they feel to possess, and they guard that power fiercely, possibly because they feel they lack such power in their daily lives. An affront or threat to what they believe is "their" forum--or at least, their vision of what that forum should be--is a direct insult to them, and that's why these conversations get so poisoned.

To recap the essential nature of the divisions here: some people think that the goal of this site is to be an question-answering service: everything short of spam is fair game for those whose purpose is to accrue as much reputation as possible. Others think the goal of this site is to create a repository of valuable, concise information: only questions worth answering should be asked, and it is a waste of time and resources to have the incessant "noise" of students who merely want to get answers without demonstrating effort.

I don't care to weigh in on which end of this spectrum I sit on--that, if anyone wants to know, should be easily determined by my (short) history on this site. What I do want to point out, though, is that there is no good way to deal with such conflicts of view: someone is going to get offended by whatever course of (in)action is taken. And because of this, I say that the actual direction of the site is just not that crucial compared to whether individual contributors decide to be adults and have a mature perspective. Whatever happens, remind yourself that this is Just Not That Big of a Deal. Step back and stop being so obsessively invested as if Math.SE is "yours." It doesn't belong to just you, and if things don't go your way, that doesn't mean you have a right to be petulant about it.

• I find it amusing that an answer which criticizes people who have strong views about how Math.SE should operate ends with such a strong statement as "that doesn't mean you have a right to". – fkraiem Nov 9 '14 at 12:26
• Note that there are two kinds of questions in view here. There is the question of mathematical content - what is the most efficient mathematical solution to the problem posed, and there is the question which involves the understanding of the person asking - how do I see my way round this problem. Two people may be dealing with the same mathematical content, but have different problems with it, or be operating at different levels of mathematical understanding and maturity. – Mark Bennet Nov 10 '14 at 21:19
• "everything short of spam is fair game for those whose purpose is to accrue as much reputation as possible." Or for those whose purpose is to answer questions at all levels regardless of reputation issues. – guest Nov 12 '14 at 6:51

I am fairly new to Mathematics Stack Exchange. I know very little about moderating, and I know very little about the details that go into managing a site like this one. As a response to @AAA I write my views on this proposition in full in this post.

First, I wholeheartedly resonate with Raff's 6th point and his observation about requiring registration. It will not cure cancer, but it will certainly send it into remission.

Last, quality vs quantity. Equating quality with growth is like equating a few dozen cheese burgers with one dinner at Fogo de Chao — it does not make sense. I would take the dinner every single time I had the choice, no questions asked.

Quality attracts new users: I hold the view quality attracts a certaind kind of user. Namely, those who are genuinely interested in math and learning, as opposed to those getting a fix for the pressing question of the moment so they can move on to higher and better things.

New users write quality post: I am of the view that a certaind kind of user writes quality posts. Namely, those who are curious, who being aware of their ignorance like to spread it around in the hopes others will find their inquiry interesting as well.

People answer here since they know their answers will be widely published and useful: I speak for myself when I answer this one. That is an extraordinarily bad approximation of my motivation to answer questions. While the conclusion is indeed true, the claim that users in general will answer with that in mind as their motivation seems misguided. I tend to answer question for one of four reasons:

1) They are interesting.

2) I either know the answer, or can contribute to a solution.(Which I hope I am doing here)

3) Recognition for what I know.

4) Sometimes...I am just bored.

People ask questions here knowing they will be widely seen and have a chance of being answered I have a subtle disagreement on this point, it is not enough that a lot of people will see it...the major reason I post question here is not because of the number of people that will see it, but because of the kind of people that will see it. Namely, people who know —and honestly care— a lot about of math. Again, quality not quantity.

To AAA: this is not so much a justification but the elaborate version of my comment.

In reading this answer keep in mind heropup's answer. I have no stake in the outcome, and I will be fine either way, these are my views, to be considered or dismissed as the community sees fit.

Yes, I agree with you that atleast email verification should be a must. Another important disadvantage is that it helps people build their own reputation. I've often noticed questions asked anonymously and then answered by registered accounts to earn reputation. I tried it myself and earned 236 reputation from just 4 self-answered questions.

• In addition requiring registration does not help that much to prevent this. It is not exactly hard to get one more email adress these days. – quid Nov 10 '14 at 17:41

Since 2012 it's also the case that pretty much any web app has moved the opposite direction...

You might say, but they're getting something free! But what is Facebook, Goodreads, Songza, Spotify, Coursera, etc., other than getting something for free? Hubris aside, we're no different.

Yes, you can even take Coursera courses without registering!

However, as soon as you want to save classes, have a record of your work, use one class as a prereq to another, etc., you'll want to.

Imagine you've never heard of math.SE. It's some sketchy site. And you want to try it out and ask a juicy question and you have to register? Excuse me, good sir. I heard about you from Google. You don't need my email, no thanks.

Growth and quality are the same thing:

• quality attracts new users
• new users write quality posts
• people answer here since they know their answers will be widely published and useful
• people ask questions here knowing they will be widely seen and have a chance of being answered

Trying to create a gated community - which isn't what the question is arguing but it is the tradeoff it's arguing - is not an option for this.

Therefore, those users who have never heard of math.SE will actually need to be attracted to it. That's really important. It's antithetical marketing to say, "You already know we're worth registering for" when they really don't.

• "New users write quality posts" - one of the main problems we face at the moment is an overwhelming number of low quality posts by new users! I don't think that requiring registration will affect that, however. – Carl Mummert Nov 9 '14 at 17:24
• @CarlMummert better than the issue where once your current leaderboard gets bored or retires, there's no one left answering any questions at all. – djechlin Nov 9 '14 at 17:48
• One can find out what the site is by browsing it, which does not require any registration -- and new users would be well advised to do so before posting. Also, since you mention Facebook -- have you tried posting there without signing up? Or on Twitter, Pinterest, Instagram...? Or to create a Google doc, or upload to Dropbox? Accessing existing content is often registration-free, posting content that goes straight to the front page of the site is rarely so. – user147263 Nov 9 '14 at 18:31
• About the line "Excuse me, good sir. I heard about you from Google. You don't need my email, no thanks." I may be wrong, and if I am I would expect @Rafflesia to know and to correct me, but math.SE doesn't know my e-mail address. Certainly, they do not know my password, rather, they get their verification from Google! – user1729 Nov 10 '14 at 9:23
• @user1729 Depends on how you signed up. If it's with Google or Facebook, etc, then the email address is shared: see here; note that the implementation has been changing on both ends (SE and provider's) over the years. To avoid disclosing email, one can sign up with a disposable email address instead of using OpenID. – user147263 Nov 10 '14 at 12:22
• @Integrator I do not understand your question. How do I think what gets verification? – user1729 Nov 11 '14 at 9:30
• I sincerely and deeply disagree with "Growth and quality are the same thing." – ReverseFlow Nov 11 '14 at 23:22
• @user1729 As a moderator, I can see your email adress. – Michael Greinecker Nov 29 '14 at 9:35
• @MichaelGreinecker Intriguing! I figured that the system could after Raff's comment - but that Mods can... do you see the complete profile that I see? Which contains my "real name" and stuff? I feel like my shroud of anonymity has been ripped from my shoulders! Do you ever stalk people on Google? What does Raff do for a living?... – user1729 Nov 29 '14 at 9:45
• @user1729 Yes, I can see that. We had to sign a non-disclosure agreement though. So the only person who has a right here to state who Raff is, is Raff themself. – Michael Greinecker Nov 29 '14 at 9:50
• @Genomeme do you... justify your belief? – djechlin Nov 29 '14 at 17:31
• @user1729 A practical implication: for SE purposes, one may want to use an email like "user1729se@gmail" instead of *FirstLast@gmail". It is possible to remove old emails and login methods from the profile, and delete "real name" field. This makes sense especially because there are more and more SE sites (MESE, HSM,...) with more moderators. It's only a matter of time until a less-than-honest person becomes a mod on one of SE sites. – user147263 Nov 29 '14 at 19:16
• @AsafKaragila I could, but I like to flag spam into nonexistence, or fix some other small things around SE. – user147263 Nov 30 '14 at 20:41
• So, Raff, what do you do for a living? (And thanks for the advice. Although I prefer to think of myself as the Stig: the moderators and a few others know my identity...but most people simply don't care and it wouldn't matter if they knew anyway...) – user1729 Dec 1 '14 at 9:21
• If one really cares about anonymity one should not only pay attention regarding the email, which is rather easy, but possibly also regarding ones IP. – quid Dec 2 '14 at 21:11