Recently, I was involved in a rather passionate back and forth with a user who insisted that the functioning of MSE as a website has led to the erosion of fun and general math interest in many students. This user is also an ex-moderator of the website, and did command support for their stance over that tenure : hence, their comments cannot be regarded as fringe in any manner (at least this is my opinion). Having been off the main site for some time, I was not able to comment on this sentiment, but I did defend the culture of the website somewhat passionately before our comments were rightly erased by moderators as it would have led to an unproductive discussion. However, this led me to several questions about the functioning of this website.
Is it true that the website lacks purpose? What is the website's purpose?
Does this site have enough people that allow for the pursual of this purpose, and is the purpose realistic?
What do other websites look like, and should this website diversify more in order to fulfil more needs?
I think the first thing that reflects the purpose of a website is its design, and by the team that handles meta-matters; the moderators in this case. The design attracts participants towards certain behaviours which it hopes the majority will replicate; and consequently the website is shaped in the image of these contributions. Moderators work behind the scenes, enforcing site-optimal behaviours, mopping up exceptions and keeping the site flowing. So : what does SE's design tell us? Also, should the functionality of the website be respected, or should it not? The second question is a debate which I'll come to later.
On any question, the two biggest fonts are reserved for the question and the answers. The comments; notice how they are not only in a much smaller font, but they're restricted in size, and once there are too many of them, the system either forces comments to go into a separate chat or compresses them. There's no doubt that the system wants you to see the questions and answers, and wants to keep any additional discussion aside or minimized. So people who like long-drawn discussions will eventually be impeded by the system. This won't be nice, but it's exactly what the intention is. Discussion being discouraged should be a turn-off to so many people, and it is.
The site consists of a front page, which by default shows questions that have most recently had activity on them, but can be adjusted to show the newest questions, the most highly scored questions, "interesting" questions, or questions having only certain tags that can be user-curated. Thus, by site design, questions are jostling for attention and need to be attractive enough, or "high quality" to answer. Here arise two questions : what IS "high quality", and what deserves attention? More on that later, too.
Apart from this, there is the moderation system, which significantly relies on pruning and editing of content. Content can be edited, deleted, fixed and questions can be resuscitated. Content can be organized as related, or duplication of other content, or as inappropriate or spam. Content is king, not the contributor.
Now, let's imagine that people did use the site as intended, and that everybody agreed with the above reasoning. What would then change?
Teaching assumes the contributor is king, NOT the content. So Stack Exchange is NOT the best platform for teaching or learning, because that's person-specific, not content specific. Discussion forums and mentoring websites would do a far better job. Since content is king here, you want the most informative and organized content to be retained and more visible than the other kinds of content e.g. pedagogical and recreational. That should attract the experts, right?
Discussion should ideally be minimized or nil. This is what is meant by "questions should be asked only after sufficient research", and that this research should be demonstrated in the question, so that no discussion is required : a question comes in, and it's ready to answer. No chit-chat, get an answer and move on.
The question is : who is this convenient for? Who exactly benefits from the site being organized in this manner? Well, one clearly sees that the answer is someone who just searches the site. They search on Google or ChatGPT or wherever, and the first hit that comes up is a high-quality SE post. They find what they want really quickly since the organization of the site and reliability of the answers is excellent, comments are removed or minimal so they're quickly able to scroll to the meat of the matter, and they can use whatever they found after exiting the site.
Every day, many, many users are doing exactly this. As of last year (the graphic seems to be absent now on the main site) ninety-four percent of visits to this site were via Google and only six percent were via other websites including the main site itself, highlighting that its major user base is visitors, not contributors. For a site that professes to help as many people as possible, it's clear that it aims on doing that by making contributors willing enough, to help visitors. The entire reason behind the reputation system is to somehow force contributors to stay on the website and chase a number, because any other reason to stay on the website would be purely intrinsic (e.g. love of math, or having a social circle on the website, or attachment to the website's purpose or whatever).
So, by design, the site is a high-quality math content repository, or a library of high quality question-answer pairs which is efficiently organized and ideal for visitors to search, find, obtain what they want, and exit. But it isn't ideal for contributors, however we try to spin it. Then again : much of the internet relies upon voluntary contribution, so I wonder what might be able to replace reputation... oh, hang on. It's obvious. Let's talk about this in the next section.
It's obvious : if I, in my free time, have to contribute somewhere, I do that because I like it. Phrase it all you like : I like helping people, I like answering questions, I like being a useful cog in a globally recognized website, whatever. Sometimes, we may need to do things we don't like, to ultimately to do the things that we like : for example, we may like gardening but may not clearing up the rotten weeds or fighting bug infestations. We may like the piano but not want to play boring scales and chords. We may like math, but not writing out the math carefully.
We'd still do it if it's worth it ultimately. So people might like to "contribute" here if the reason they attribute to their participation and contribution is worth more, in their opinion, than whatever site functionality and moderatorship may think about that contribution. It appears self-centred, but it's not, and here's why. Remember, self-centredness is morally an issue for a large number of people, but unless it's legally an issue one can let it slide as an expression of opinion or an undesirable but permissible behaviour. On a site like SE, the very lack of clear definition of multiple keywords ensures that many behaviours are just that : undesirable, but very much permissible. Perhaps even defendable. Maybe more common than many might think!
Let's see. We want a website that prioritizes reliability and organization, because that's what the website reflects. But
1) When a question is asked, who decides whether it's reliable? Some would think only the original poster knows the true meaning or what they're asking, others would believe that they know what the original poster is trying to say, or that it's obvious. So "lack of clarity" lacks clarity! (who would've thought). Then comes breadth : when we decide questions are broad, what if there are two extremely similar questions in one post, or two questions that amount to a general and a specific case? What if a question appears opinion-based, but its first answer ends up being highly fact-oriented and absolves the question of apparent wrongdoing? So, the lines are grey, and we have things to argue over here. We've now outlawed problem statement questions (PSQs) but one argument in favour of them was that they were crisp and to-the-point deliverables for visitors, no corking and playing around, straight to the point. So even those questions carried a sliver of merit.
2) Who decides what kind of content is organizable? Today, the standards say that context must be provided. I can think of three reasons (a) somehow, it convinces multiple people that the original poster isn't simply getting their homework done (b) it actually adds to the question, makes it attractive to read and might help answerers with their attempts (c) it may contain keywords, which make the question easier to search. But what about a question which contains only keywords and no "effort to solve it"? What about something that has minimal, sufficient context, which an answerer believes is enough for them to answer the question, but not enough to really allow that question to be accessible to search? And then there's the question of duplicates : how general or specific a duplicate target should be, whether duplicate targets can afford to be questions that are "good then, bad today", whether duplicates should be older than the original question etc.
These are debates that, in my opinion, are in line with the website's purpose, but still capable of creating fault lines. I don't think we can avoid, stop or even discourage such debates; we must come to uneasy consensuses, enforce them consistently, and then debate away the remainders or leave them to the voting system. So site functionality and moderatorship can be respected, and yet we may still have very little in common to cheer on. Contributors have to really rummage through their reserves of desire, interest and commitment to persevere over these rules, but ultimately marvel at what they're creating (and maybe take a little bit of fun collecting rep as well : it's not an evil number if used correctly).
But are there enough people who are taking about these things, having debates and helping with the website's running? I don't think so, although the site has seen an alarming dip in questions recently so the number of people needed is also down. The thing is, if the site needs to diversify and cater to multiple "purposes" around the main purpose, people need to work hard and moderate, say, certain tags or posts separately, and perhaps if each question can receive more curative attention then the site can flourish in different directions too. More people need to step up, though. I can imagine that, if the number of questions dips below some threshold, then we can start treating each question separately and wriggling it into the system through edits or discussion. That would bring up MSE's versatility massively.
Coming back to the debates, though, and I think there have been plenty of debates that don't prioritize or respect site functionality and moderatorship. These include fundamental disagreement with question closure, deletion and downvoting (which are purely content rating mechanisms), for instance. So I don't respect that kind of debate, would discourage it, and would keep out of it if necessary ; maybe at one point I did, but I'm tired really. Users having these kinds of debates will be finding themselves trapped here, to be honest. They should seek alternatives... of which there are NONE. Zilch. So I'm actually sympathetic to their requirements, but we can't fulfil them.
Why no alternatives, though? That's what we'll discuss next.
A recent SE meta post highlighted how, even though opinion based questions aren't allowed, they are still sometimes answered very well on the website, making the question worth keeping. It seems that even though SE is set up for one purpose, it keeps doing other things pretty well too, which means that people keep coming here for purposes tangential to MSE's usage. That's a lot of versatility that the site needs to display, and it's functionality is eventually overloaded. You still find cranks posting rubbish, new users asking a duplicated question for the millionth time without having searched it, and one question with a comment chain longer than the chemical name for Titin each day. They all want to be catered to here.
Well, think about it. The site promises reliability. How many sites do you think collect experts on a voluntary basis? Quora and Reddit don't count because they prioritize curiosity, helpfulness and wonder over question-answer pairs, so there's no expert-curation, just a bunch of rag-tag posts with a diamond in the middle of ten turds and eighty bales of hay (and a few dollops of cringe and advertisement along the way). Experts would use sites in their downtime, and their attention will be grabbed by posts door-delivered to them, of their requirement. SE does that miles better than Quora and Reddit. Any other site I know which collects experts typically puts content behind paywalls and keeps its experts on its payroll, thereby losing accessibility to visitors.
The site also promises organization. Try reading any Quora post, even one that has just four decent answers, and scroll through three pages of concomitant nonsense, one photo of sultry eye candy and maybe, maybe a couple of decent answers which are massively voted up by random bots? (with the other answers probably being hogwash). Reddit isn't too different. Try one of the Discords though, it might work : until you realize that the place can probably hold about 40 people before someone distracts the whole group. It's Metcalfe's law, no one can stop the math from mathing!
So there are ZERO scalable, affordable and accessible alternatives to MSE when it comes to ANY of the following purposes :
Tutoring/Teaching with a reliable mentor.
High quality question-answer pairs, which is basically what MSE does.
Discussion forums where worthwhile questions are picked apart and debated.
People need to build these alternatives, but would need to find ways of keeping noise out of such networks. They also need to debate ethical issues like doing students' homework or exams by mistake, and how much they want to entertain crankery, spam and memes on their network. SE has succeeded, to some extent, at doing precisely this. Usenet could not, AoPS could but searching on AoPS is like trying to find a needle in a haystack, and well, the less said about Codidact the better. That website, though, it's pretty sad to see it not taking off.
For if it did, then I think I'd have less sympathy for the teachers, the ones who'd love to discuss fancy ideas and brianstorm, and the cranks : they'd have their space(s), we'd have ours. But no, as you see, they don't have a space. No one, anywhere else, would be able to give them the service that SE does. That's why we have to realize that the purpose of MSE is a question-answer site in spirit and design, but ends up being an everything-about-math site eventually. It still continues to prioritize reliability and organization, but also recognizes the need to cater to those who do not research, want to discuss, want to teach and want to read opinions. Eventually it ends up serving a much, much larger purpose.
So to all those who feel unfairly treated by SE, and are definitely at odds with its purpose : go create that dream website, and show the likes of Atwood and Shirky that they were miles off in their design. Oddly enough, I'd love that, because math as a subject deserves three or four quality websites that cover every single blade of grass on the field of student requirement and development. But no, SE should not be overloaded. Its functionality and moderation carries a narrow, but effective purpose, and let's have all the debates we need around that purpose without getting carried away on more tangents. Over to you all.