What’s on your mind?

Question

You. Yes, you: What’s on your mind?

When you think about the future of the network - years from now - what worries you? What makes you anxious? What do you look forward to? What might make you hopeful?

---

I’ve spent the last few months thinking about the state of the Stack Exchange network and where the future might take us. My reflections have meandered down many inconclusive pathways, and prompted questions about the network that have no easy or simple resolution. I eventually decided that my long-term aim is to help answer three fundamental questions: Who are we? Where do we want to be? And, what do we need to do to achieve it?

These are large, heavy questions, and they all revolve around a shared sense of common identity and purpose. (I suspect it is uncontroversial to say that the feeling of a common mission has felt somewhat degraded lately.) In many respects, these questions are not mine to answer – but they won’t be answered unless someone stops to ask them. I might as well get us started. But it’s best to take it one step at a time.

“We” might seem a little strange for me to use here. The last few years have not been the most harmonious, I’ll freely admit. So I will clarify. I mean “we” to refer to the three major roles: consumers of the knowledge stored here, those who create and store knowledge, and the people who operate the place itself (not limited to employees), who all together produce the thing we call “Stack Exchange.”

---

You might be inclined to believe I am preparing you for some specific change, a product release, or maybe a policy shift, and I could hardly blame you for thinking this. But I assure you I am not. I ask out of my own curiosity and a growing restless sensation that tells me questions like these are becoming ever more necessary.

For this reason I feel confident saying there will be more questions and opportunities for discussion in the future. Please do answer, but please don’t rush to answer. You won’t miss your window in the here and now. Stack Exchange isn’t going to keel over and die tomorrow, and there is no hurry: time enough, certainly, for a moment of contemplation, which I would encourage you to take before hurrying to respond.

The answers below should not be a popularity contest, and I’d ask you not to treat them like one. (If I could disable voting, I would.) I really want to know your answer. I want to know what your thoughts are, free from worries about how well it plays to a Meta audience. I know the scripts and boilerplate arguments like the back of my hand. They have merit, but I’m not as interested in rehashing them here (unless you have a particularly useful rendering of them in mind). I’d ask you not to worry about its reception here, and for similar reasons, I’d also ask you not to self-censor too much.

I would like to make explicit that I want to hear from atypical voices on the platform. I want to hear from a wide range of people gathered here. So treat this post as an explicit and open invitation to people who do not normally participate on Meta. If there’s someone on one of the Stack Exchange sites whose voice is not represented here, please send them a link and encourage them to post.

Finally, while I’d encourage you to read existing answers, I’d also encourage you to post even if it feels like the discussion has already played out. Six weeks from now, I’ll still be interested in your answer.

---

Please take some time to think. All of your thoughts are welcome, but I’d encourage you to focus specifically on the question at hand: When you think about the future, years from now, what comes to mind? In the long view, what weighs on you, and what gives you hope?

Join me in Chat, if you think the format there will suit your thinking better.

Answer 1

When you think about the future of the network - years from now - what worries you?

I worry that what you said is true:

Stack Exchange isn’t going to keel over and die tomorrow

But not because it will happen after tomorrow. But it has already happened.

I am directly looking at the company here. After layoff after layoff, I find myself losing hope in the people on "the other side" of the network. The employees and the company. Generally, taken as a whole.

For years, the technical product has been getting somewhat weird, somewhat bad, somewhat incomplete updates. But lately, it feels it has been getting worse. Project after project starts and gets abandoned partway through. We have had things released with an announcement coming hours later of what to expect. When there are already heaps of bugs reported for it. And are never addressed or just partly. The follow-up after something is released and users find it unsuitable is more and more lacking.

It is hard to even pinpoint exactly when this started feeling as more of a problem than just "well, these things happen". Before the layoffs it was already happening, for sure. But after the layoffs it has gotten perceptibly worse. New redesigns or new features or even fixes are done that clearly show whoever is in charge (and that would be multiple people involved) did not actually have an idea what they were redesigning, creating, or fixing. Because they get it wrong.

It is kind of like seeing somebody draw an animal they have never seen but only described. You can squint and maybe see what they were going for. "A yellow horse with black spots, long neck, and horns. Oh, it was supposed to be a giraffe. I can kind of see it." You know - that sort of feeling. Clearly it is not the thing. But it resembles it.

That is becoming increasingly common. Staff have also outright admitted they had no idea how something worked after redesign/fix. When some user pointed out that it is missing some common feature we would take for a given.

Simply put, it feels like whoever is left after the layoffs does not understand the product they have. Neither on what the design was and why (so, technical side of this), nor definitely not on the usage side.

For each new step taken, it seems sometimes there are up to two steps back. And the site is on a very slight downward slope with the occasional slight bump. Like a boring roller-coaster.

And yes, there are things that happen here and there. The Community Asks sprints, for example. They suck, by the way:

I am not trying to be ungrateful. I am trying to say that the company has had the time and resources to address these things multiple times over. Yet it did not. Few things that were implemented were great. But also, few did the step back and forth dance. The ones that need more than 30 minutes - that can have a measurable complexity. The inbox "improvement" from the first sprint? Still not addressed. I hate using the inbox now. And I have to do it every single day. The nonsense that was done with the tag synonyms search. Yeah, it was a yellow spotted long-necked horned horse.

Again, some things were great. Only...I am one of the people who tinkers with the sites. Not as much as others, but I write userscripts and userstyles. Occasionally, even publish them. I know adding an asterisk to the close/delete votes did not take long. I had a userstyle for it. Yes, simple CSS only. The whole infrastructure for determining when you had voted for something was already there. I know it cannot have taken long to just add an asterisk to maybe few places.

And this is what I noticed from the Community Asks sprints. A lot of very low-hanging fruit were picked. I am not ungrateful. I am indifferent. I cannot even get excited for any of this.

I used to be excited. I used to look forward to Stack Exchange trying out a new tech. Maybe even ML, AI - I used to imagine the possibilities. Now any time something is announced, I can only imagine how badly it could fail or how it would be misapplied, how it would be mishandled.

I have lost almost all my faith in the company delivering a product. Which has huge implications as many long-standing users have left over this. Or maybe not "left" but reduced participation by a lot. On one hand, you could say it is over some UX. But the whole thing runs deeper. It also damages the connection with the community.

What do you look forward to? What might make you hopeful?

There are still employees within the company that seem to get it. They use the sites, they do actual conversations. They understand the community.

Like you, Slate. Quite honestly, I decided to answer because you asked this. I might not have written an answer if it was posted by somebody else.

I wish so much that some day I would start thinking of "the company" in similar terms. And not "the company" as one big entity that is clueless, and a handful of employees who are not.

Answer 2

I worry that the business model for the public Q&A network might not be sustainable in the future, and that this will lead to a much bigger conflict between the interests of the company and the community. The entire internet is getting shittier every day, the SE network is one of the few places that still has mostly well-curated content that isn't written simply for SEO or marketing. I worry that complex, difficult questions alone could not sustain the business if the vast majority of traffic vanishes because that need is handled by LLMs and similar tools.

I worry that the community has gotten too rigid in their views on how to enforce our quality standards, and that this might result in sites run by bitter vets that slowly die because no new users take up the work. I don't think there is a fundamental conflict between helping people and creating a library of curated content, but this is a view many experienced members hold.

I worry that the divisions between the community and the company will prevent necessary and maybe risky changes to be done, or done well. The community will likely react badly to many changes and assign ulterior motives to them, and SE will stop listening to the complaints because they will consider them unreasonable. But difficult changes to core systems will not work without community input.

I am cautiously optimistic that AI and LLMs will not be purely a threat to the network but also a chance as there is a niche to fill for authentic and curated human-generated content with high quality.

Answer 3

Necessary preamble: Yes, I am an employee of Stack Overflow. In fact, if you go by original start date, I'm the oldest remaining employee here (but I did spend a few years elsewhere, so I'm not the most tenured). But I'm a simple software engineer who sometimes plays the resident historian, not a decision maker.

I am writing this not because I was asked to. Nobody even pointed me at this question; I just found it on Meta. I also didn't ask anyone for permission to post this.

To be perfectly clear, this is my personal opinion as a member of the community out of which I was hired many years ago, and it's not in any way representing the opinions or policies of the company. I chose to use the word "we" when talking about the company merely because it felt disingenuous to use the third person when in fact I am part of said company.

Finally, I'm writing about Stack Overflow specifically, but most of what I'm writing also applies to the rest of the Stack Exchange network.

---

The original purpose of Stack Overflow was to build a collaboratively edited repository of programming knowledge.

In that sense, I actually disagree with your (Slate's) comment here, where you say

Fundamentally, we are a pile of questions with a bunch of answers attached to them: the perfect haystack.

Yes, arguably, we are a pile of questions and answers, but in my opinion that is not fundamentally so. The Q&A format was always a means to an end. A couple of points to support that opinion:

• We made concrete questions about real-life programming problems the building blocks of the site because it ensured that the collected knowledge is relevant and timely, and the signal doesn't get drowned in the noise of random discussions about anything that could somehow be construed to be programming-related.

• Duplicate questions were strongly discouraged, with tools to deal with duplicates built very early on, because a repository of knowledge needs a canonical place with potential solutions to a problem collected in one place, instead of a haystack.

• The main purpose of the whole voting system was to make sure that answers can be sorted by what the community thought was the best answers, instead of a cronological list of forum posts.

• Jeff Atwood considered Stack Overflow to have reached 1.0 status at the moment when anonymous users were allowed to edit posts, because everyone was supposed to be able to make the answers even better for the next person with the same problem.

• When a question was posted, that moment was always supposed to be a tiny seed for what was to become the canonical resource for anyone who might have the same problem in the future. The question itself was important because it was the starting point, but the answers were the pearls. Yes, the original asker should ideally get an answer, but the priority were the thousands of future people who had the same problem, and who now no longer had to ask the question because it was already answered.

So, the original vision was first and foremost to create that archive of knowledge, available for everyone and not owned by any individual entity. Community-curated Q&A were the means to achieve that vision. And it was widely successful.

The founders spent a lot of time thinking about things like community management, often quoting a book on the topic called Here comes everybody. They made decisions about things like gamification, reputation, moderation, and elections. You can argue about each and every decision, and I myself had a healthy amount of disagreement with Jeff on details.

Decisions have consequences, and they will always involve tradeoffs. For example, the strict rules of the site were made to ensure quality and long-term relevance, but they also lead to a culture of rule-lawyering and gatekeeping that we observe on Stack Overflow today, and that give it a reputation of being an unwelcoming place.

But ultimately, the fact that Stack Overflow became an overnight success shows that the decisions seem to have made some sense.

At the time.

And that ⬆️ caveat, to answer your question, is what's on my mind.

Stack Overflow was built for the world of 2008, for a generation of programmers who had often learned programming from books or from trial and error in their IDEs. This was the time that marked the early days of Web 2.0, when Twitter was brand new, and people used the internet much more in a transactional way. Heck, the idea of a web-based chat on the side seemed crazy to many.

But we live in a different time now. People learn new things online. People live online. We're no longer only talking about experienced programmers who learned the ropes elsewhere and come to Stack Overflow for specifc advice. We're talking about generations of programmers who are learning from Stack Overflow, who never experienced a world where Stack Overflow did not exist. People who learn programming from YouTube videos, who discuss their front-end libraries in Discords, who have never installed an IDE because they started their programming journey on Glitch.

(Yeah and there's also the LLM-shaped elephant in the room. I don't want to talk about it.)

It's easy for old-timers like me to dismiss those younger programmers with a giant dose of "you're doing it wrong", but all yelling at clouds is not going to change the fact that the world has evolved, and Stack Overflow has not.

I don't have the answers to what Stack Overflow needs to be like, in order for it to be a home for today's programmers. But I'm pretty sure the answers are different from the ones of 2008.

We need to figure that out, and if there are things about Stack Overflow that no longer make sense in 2025, then let's bring the sledge hammer.

Some people (myself included) are going to be unhappy because those things were important to them. Let's be empathetic with those people, let's allow them their grief, and let's explain to them why we believe we're doing the right thing in order to fulfill that original vision. And let's not do that explaining with a press release and platitudes, but by talking to them eye-to-eye.

Then pour one out, shed a tear, and swing that hammer.

And then build the Stack Overflow of today.

Answer 4

What worries me the most is that step-by-step, the company will destroy the core value of the sites and drive away people that come here, day after day, to ask, answer, curate, and engage with others creating wealth of knowledge.

I am worried that experts and learners that are primarily driven by desire to help and share, as well as learn and improve their own skills, will be replaced by ones whose primary desire is collecting some virtual Internet points at all costs.

I am worried that everyone knowledgeable will eventually leave, and that all sites will turn into remnants of old human knowledge covered with AI junk.

Answer 5

I would first answer your questions with a question of my own: before people sink considerable energy into responding to this in meaningful ways, since you mention this is just you asking out of curiosity, what weight is there behind this initiative? If we come to you now with deep, impactful concerns that drive at the root of some existential problem, is anything going to happen to address that problem? Especially if it interferes with other interests by the company's board or leadership direction?

I ask because we already have a bunch of the answers you're soliciting here, strewn about in questions and answers here on MSE or on MSO or other network sites, bemoaning the state of things, highlighting critical bugs in software or in workflow processes that go outright ignored by the only people who can address them: the company. And they are ignored because they don't align with (and in some ways directly conflict with) the company's current vision, which is: cash in on the (freely donated) work of the community to sell data to Google in an effort to make a bunch of money. Instead the only work on the core product (Q&A) anymore is a one-week sprint every few months (which sometimes get canceled anyway) or random UI/design overhauls no one asked for and which break half the site functionality, to boot.

So I guess what worries me about the future is: how bad will Stack Exchange be after <insert number here> more years of soulless abuse of the original purpose of Joel and Jeff's vision (of a high-quality, curated open network of Q&A sites populated by human experts) by owners who not only don't understand but also don't even try to use the product they now control, or listen to their traditional customers (SE communities), instead opting to shift to the business-minded view that the only customers are those who pay, not understanding that it's the other kind of customer (the user base) that made the product and company worth paying anything in the first place.

Jeff and Joel did a great thing with the re-launch of Stack Exchange 2.0 when they licensed all future content under Creative Commons, but the main mistake they made was not making Stack Overflow a non-profit or similar structured business entity. Because it is just a general for-profit corporation, I'm worried they doomed it to a future of ownership by value-extracting MBAs who care about shareholder value and press releases rather than whether the product is any good or the customers are actually served. That's certainly been mostly what we've experienced since the sale to Prosus, and I'm worried that the company will continue along that trajectory. Stack Overflow doesn't need to make lots of money or partner with AI or whatever else. They just need to have a few core offerings like ads, enterprise instances, and job listings (whatever happened to that, eh?) to pay the bills. The company should focus on finding its niche and fitting there for as long as it can, not constantly growing or constantly reinventing itself like toxic MBA/private equity culture demands.

Answer 6

My worries aren't really confined to Stack Exchange; I have a broader concern about Enshittification of the parts of the internet I've grown up with and found useful.

Google search is probably the center of my disappointment about the internet lately. You can't find good things from there these days. As others have noted, that often includes not finding Stack Overflow at the top of Google search results anymore. I don't know how much that is SO's fault and how much is Google's.

Instead of useful stuff, most of what rises to the top of search is junk, and even the good stuff is padded with crap to show up in search. That means every recipe page is padded with silly stories no one cares about. If you search for public health information, you're just as likely to find the top results polluted by local clinics and dentists (not even necessarily near you!) who paid a SEO firm, not to mention all the scams selling you on some pseudoscience supplement. Purveyors of honest product reviews are getting crowded out by disguised paid offerings and trusted/familiar brands are being gutted, effectively selling off their reputation without any concern for quality.

This junk all predates the new GenAI craze, but GenAI makes it that much easier to make junk. I don't know if Google will turn their ailing search around, or if they want to; someone who finds what they need right away isn't loading as many Google Ads pages. I'm not finding their competitors' products much better. They're all dealing with a deeper and deeper pile of junk to sort through, and either they don't have their own deep pockets or their own deep pockets come with their own baggage.

Given that it's so hard to find good stuff, it's hard for anyone new to make good stuff: they can't bring the eyeballs to it that they need to grow and survive. What we have left are a dwindling number of individual sites to rely on for decent content. For me, that includes some useful subreddits, it includes Wikipedia, it includes some other more focused Wikis, it includes Stack Exchange, and it includes some more private/local spaces like Discord channels that are either local spaces for a few friends of mine or communities built up around a particular game or modification or software tool.

So, I'm worried about Stack Exchange. I see how catering to business needs often destroys the underlying product. I see how common it is for companies to decide that the most valuable thing they have is their reputation and then subsequently decide that the best way to extract that value is by pilfering it.

Others have held onto the idea that if Stack Exchange does bad enough, some other Q&A will pop up in its place. Maybe that's true, but it's really hard. I think it's gotten a lot harder to bootstrap a good quality site. There already exist alternatives in content (like subreddits) but not really in structure - other Q&A sites are already themselves infested with junk or just don't have enough of a following (sorry Codidact).

I don't think as a company Stack Overflow/Exchange is quite at the pilfering step yet, but it seems like it's coming soon. The new "Jobs" product seems to me like a pilfering move. Other attempts to save the site in the face of GenAI competitors seem doomed to produce a worse GenAI product while abandoning the actual benefits of this one.

Answer 7

I've recently been thinking a lot about long-term preservation of accessibility of information.
Lately, I got university access to a whole lot of information, such as research papers, which really helped drive home how frustating it is to have that knowledge and information locked behind a paywall or simply gone.

Luckily, freely sharing and storing information is really one of the primary goals of the Stack Exchange network as a whole. People ask questions, other people answer, and that knowledge is now available for the public. Other people edit and improve the content, creating a repository of knowledge that is freely accessible to all.

Unfortunately, having a centralized, free, repository of knowledge available doesn't necessarily ensure that it will stay that way.

The Internet Archive recently went down due to malicious actors, highlighting the issues with relying on a single source for long-term preservation - or even two sources. The Internet Archive is the primary tool that most people rely on to back up information here on SE, from preserving the SE webpages themselves to the off-site resources linked-to by them. Assuming a worst-case scenario, if both Stack Exchange and the Internet Archive are compromised, get taken down, or decide to paywall their content,¹ then we are left without free access to all of that information that we've built up over the last fifteen years.

We do have the data dumps. However, the process for accessing Stack Exchange data dumps was recently changed, which has made obtaining and preserving a backup of the entire network much more difficult. (There are community-developed tools to get a network-wide dump, but it now requires being able to run such a script, dodging the SE rate limits, and creating a profile on each site - a relatively significant amount of effort.)
Before it was changed, the data dump lived at the Internet Archive, which meant that it didn't rely on SE itself, but it was (is?) then vulnerable to the same problems as everything else on the Internet Archive. The data dump even currently also doesn't back up images, which are often integral to understanding the post, and the alt text is, unfortunately, usually insufficient to replace the image in long-term data preservation.

The data dump also only addresses half the problem. Long-term storage of the content has other issues; link rot and decaying images can remove supporting evidence or references, and information becomes less relevant and goes out of date as advancements in various fields are made.

So while the data dump is a good step in the right direction when it comes to preserving the knowledge on the network, it doesn't entirely solve our issues. There's no real way to guarantee that the backup will remain free and accessible without to a degree abusing the current data dump system, dedicating gigabytes of personal storage space to storing it, and then somehow allowing external access to it. We also need to better make sure that the information is transformed into a format where it can be preserved long-term, such as lessening the reliance on images, improving the alt text process, and making it clearer when information is likely to eventually go out of date.

To sum up: our current system in theory caters towards creating a long-term repository of information, but a lot of the content on the network simply isn't designed for long-term preservation, and we don't have any good methods of ensuring that the knowledge is preserved in the long run.

---

¹ Incidentally, one of the reasons for creating the data dump in the first place, in case the company abandoned the goal of freely accessible information.

Answer 8

Indifference

This is not meant as some hyperbole based on recent events. Well, there is a bit of pouting perhaps, but that is really just the icing on the cake. When you ask me to think about the future of the network, years from now...

[radio static]

Now, hold on rolling your eyes for a second while we stroll down memory lane...

You see, SE (actually, SO initially for me) was an idea to invest in. Building a library of knowledge and all that. Using a library of knowledge, just to get that point its often-missing spotlight. Yeah, that good feeling when you are finally given a broom because you are trusted to help keep the place tidy behind the scenes.
That warm feeling of familiarity when you open SO at 10 PM, find an answer to your problem and realise you already upvoted it a year ago. Because SO got your back no matter how forgetful you are. Because on SO, DenverCoder9 does share their wisdom.

This spark is gone.

I could (and did) come up with things that worry me, that make me anxious, that I might look forward to, or that might even make me hopeful. But those are, to be completely honest, all technical thoughts. Whether AI makes SE obsolete, how the community might change, licenses and acknowledgements and what it means to be a library and how that could evolve and yada yada yada.
But none of that is "on my mind". It is in the spreadsheet-land of my head, the tinkering-with-a-challenge section, the trying-to-crack-a-puzzle domain. A metaphorical piece of paper that you know had some interesting scribbles on it but if you lost it, well...

What is "on my mind" is... meh.

SE losing the space AI race? Well, too bad, what else is on? The community duking out an epic battle of curationists versus helpdeskians? Welcome to the internet. The ultimate solution to acknowledgment of intellectual impact in an age of evolving rationality of information provenance given coevolving diffusion of knowledge dominion? Sounds like fun (NOT). An irreplaceable vault of knowledge jeopardised by the whims of modern techno-society? Welcome to the internet, again.

What is on my mind? On my mind is "I don't care".

And this is worrying me about the future of the network. Because the network does not run on "I don't care". Because I am decently sure I am not the only one who has to admit to themselves "Actually... I don't care".
But at the same time, it makes me hopeful. Because, you know, for every other swing of the axe to the network by SE inc or vocal members of the community or the public opinion or people silently quitting... I don't care anymore.

🤷‍♂️

Answer 9

I'd like to thank you for asking us these questions. They're really important ones to re-ask every once in awhile with a project as old as Stack Exchange is. The world's changed a lot in these last few years, and the network's changed right along with it. People come and go, features rise and fall, but this massive encyclopedia of knowledge is still here, and still widely useful.

I'm going to try answering the "big", core questions. I anticipate that I'm going to say things that are either flat-out wrong, or that you disagree with. I fully expect to miss the mark at times. I imagine this may be because I haven't been around here nearly as long as a lot of others. I still want to throw them out there.

Oh, and let's get rid of a few readers now. You can just stop reading and downvote if you disagree with this statement, because after a lot of reflection, I don't personally respect this line of thinking anymore:

Stack Exchange, Inc. is not evil.

---

Who are we?

We're a large, multicultural arrangement of users working together, knowingly or unknowingly, towards the betterment of an impossibly large encyclopedia of Q&A-style knowledge. We believe in a self-regulating system that empowers its users to act like stoneshapers, carving away at rock until it resembles something worthwhile to look at.

... No. That's a bit of an antiquated definition, isn't it? Much as we'd like for it to remain as it used to be, this place has grown to be far bigger, incorporating the interests of far more than just a core userbase. You're a software company whose flagship product of public Q&A no longer feels like its flagship product, because you've added more products to support. We have to expand our definition to include the creation and maintenance of software that's at the benefit of your sponsors via Collectives, your customers via Teams and OverflowAI, and your shareholders. That's a long list of people you need to do right by. So much so that... Sometimes, it feels like you do so at the expense of the community that's been there since the start. The community that spawned when Stack Overflow was tears in Jeff and Joel's eyes; where the words "Build it and they will come" were first floated. That public Q&A community has changed too, though. It's grown, shrunk, expanded and contracted in interests and desires, flared up during controversy and quelled during times of ease. It's acted as your contributors, your feedback mechanism, your research audience, your QA testers, your moderators, your confidants, and... Dare I say, at times, your friends.

We're more than we ever were, now. That makes things challenging. For us, and for you.

Where do we want to be, and what do we do to achieve it?

Where do you want to be, Stack Exchange? I don't think you've told us lately. I'd love for you to answer that question. You've, in a roundabout way, been telling us what you want to do to achieve it with some of your actions, but sometimes that feels... Off to us. Decisions and product movements that just don't feel right. When you do state a goal for a change, it sometimes feels like you either aren't working towards that goal, or that what your doing may even run contrary to that goal. Maybe that's because we don't know what the heck you're trying to get at.

I really think that we seriously just don't see your vision, and when we point out issues with what you're planning to do, we don't feel as heard as we feel we should be. Were we all on the same page, maybe it would make more sense. Instead we have to guess, and that leads us down toxic, and sometimes conspiratorial, lines of thought. I hate being there. I hate listening to others follow those thoughts. It feels shortsighted, rude, unproductive, and pointless. But how can we be less shortsighted when we don't know what your vision for this place is? Tell us what you're trying to do, how you're going to do it, and why. Lay it all out for us in a way that truly unveils what your core goals are, and try and listen to us when we point out that what you're doing runs afoul of that goal. We're not as unreasonable as you might think.

One thing you're not lacking in here is a long line of people who want to see you succeed. We differ in our views of what success looks like because we seem to have an entirely different finish line than you do. My Dutch moderator colleagues have repeatedly mentioned that they worry their Dutch bluntness can come off too strongly at times, but I urge you, as I would urge you with all feedback, to pierce through that bluntness to the core of their message, because it's so often what needs to be said or considered. I value their input deeply, no matter how well it's dressed.

The community, if I dare to represent it here, wants to move towards a Q&A product that empowers its contributors to make the content here the best it can be, with as many well-meaning users contributing as it can. We want to be the place that people go when they want answers to their questions, feel rewarded when they get those answers, and feel safe in the knowledge that they've been vetted by helpful humans. We want new users to be informed of how to best fit in with the established norms and guidelines, and feel welcomed into the space. We want the content on the network to be human-verified at a minimum, accurate, and helpful.

To get there, we want the public Q&A product to receive a lot of community-sourced attention from SE on the technical issues we've been asking for for years. We haven't even begun to consider the "societal" problems we have because we're so distracted by all the technical problems we'd like you to address. I hope that was one of the takeaways when Phillippe asked his magic-wand question, because he explicitly asked us to err away from technical asks, and yet 90% of the responses there were technical asks.

What does the future look like, years from now? What weighs on you?

To be honest? Minus the recent AI... uh... stuff... and the potential for that to leak onto the site... For the most part, I would imagine the network looks the same as it currently does. Y'know... That worries me a lot. I'm really worried that not enough meaningful change is going to come between now and, say, five years from now. Whether that's a result of a lack of resources or a lack of direction is unclear. What gives me this impression is a "Two steps forward, three steps back" feeling at times. That isn't meant to make you feel as though your efforts are worthless, it's just a feeling that I have. I'll readily admit it may not be fully rational.

I'll take a moment to mention something that weighs on me a little, and it's related to the Community Asks sprints. These sprints are fantastic and I'm really happy to see some old, long-requested, low-engineering-lift asks are being completed by you guys. I will never stop giving you credit for doing them. So... Why does this weigh on me? Because the nature of it makes me feel (rightly or wrongly) like our Asks aren't otherwise being prioritized. We've got a long laundry list of technical adjustments that we'd love to see. You guys don't have the engineering bandwidth to tackle a lot of them, and I completely understand that. But when we have all of these outstanding asks and your resources go towards extremely unpopular features that we didn't ask for and/or gave feedback about how poor an idea it was/is, it feels really bad.

I'll toss some recent examples out there not to be mean and punch down on you, but because a lot of the feedback on them involves a bunch of genuine confusion* about the choices made, which points back to my previous point about how either your goals are miscommunicated to us or your choices in pursuing those goals are misaligned:

• Homepage updates for logged-in users (Removing HNQ?, A proposal to mass-edit the tag wikis to account for the change)
• Tag hover experiment wrap-up ("Everything about this is backwards")
• The most recent Discussions announcement ("What purpose (do) they serve", "It's time to admit Discussions have failed", and this call to stop moderating it because it doesn't have the bare minimum the community needs to support it)

Now let's un-minus that recent... (sigh) - Generative AI stuff... Ugh, I'm seeing AI absolutely everywhere nowadays. It's in my browser, my favorite websites, my operating system... My email service, if you can believe that. Each iteration that's presented to me across the internet, and I promise you I am giving them an honest try, is marginally helpful at best. So many of these applications seem like they could've just been a feature that relied on very few metrics instead of some LLM. All of that primes me not to like it when I see even more of it, but believe me when I say I'm trying to keep an open mind.

Rant aside, let's talk about AI and SE. I'll warn you, I fear I may have written the following a bit unkindly. Not very mod-like, eh?

Since its explosion, you've loudly and proudly been saying you want to try and be a huge advocate for ethical AI, and you've tried relentlessly to get AI on the public platform. You're even putting it in the SG for crying out loud, the SG being something I was so happy with I hopped on with Ben Popper and basically did PR for it. I stand by everything I said in that podcast, but your reckless abandon to pursue AI for the sake of AI is turning me off (the network) really hard. You may think your foray into AI isn't reckless. That it isn't just putting it on the site for the sake of putting it on the site. But that's how it feels to us. Cutting past our reactions to that is going to be extremely difficult.

So obviously when the Answers Bot leak happened, I just... Got sad, I guess? I say this all because SE is the place I go to for human-generated content that's worth reading to help me with a problem I might have. If I wanted AI, I'd go somewhere that has AI. You've heard this before, so let me just carry on.

In the response to the AI bot leak, Philippe says that the main reason for this work is to prepare for a future LLM that is, and I'll exaggerate to the extreme here partly because I'm having fun and partly because I feel like it may demonstrate my point, "The Almighty Understand-er of All Questions that Can Answer Everything Accurately".

Let's assume that's an achievable goal. Let's assume that there will one day exist an LLM that's accurate and "good enough" to pass the smell test of an expert on any given question (because that's one of our bars for participation, by the way!) What the heck use will SE have then? Why would I want to come to a site for a human-written response to an issue when I can just ask the LLM in this futuristic dreamscape that's near perfect and remove the step of looking at SE? This is likely a bit of a shortsighted conclusion, but ultimately you're creating the on-ramp for the kind of content that would herald... Your own obsolescence? Kinda? That's probably stretching it a bit, but assuming we one day have an LLM that needs very little human review... What then?

All of this has been framed from a technical perspective; how the site will function and appear. What about from a social perspective? How does the relationship between the company and the community look five years from now? Do users feel like their asks are being met? I don't have true answers to these hypotheticals. All of it deeply depends on company direction and the world changing around us. Will we continue to squabble about (in our view, half-baked) features? Will we still feel unheard about some things, but heard and respected in others? Will we still have conspiratorial thoughts about you guys killing the data dumps?

What worries you?

I'm worried that the community will never be able to trust you guys again. Not fully, anyway. Like I mentioned earlier, there are folks who are "black-pilled" (warning, Urban Dictionary link) on Stack Exchange. They're convinced that the network can never be where they view it can/should be. They're under the impression that you guys are doing good things to shadow the bad. They're convinced you don't care, that you're going to kill the data dump, that you're going to force AI down our throats, and that you're stealing our data and selling it to the highest bidder.

I don't like to think that way. I like to think that you have far more nuanced motives. But what worries me is that, if these kinds of negative sentiments are mainstream... What the hell is the point in continuing? If we've completely discarded the assumption of good faith, how can we ever hope to move forward in an environment where the community is meant to be at the center of everything you do?

What gives you hope?

The people.

What I mean by that is, there are a bunch of people who really try their best to make this site the best it can be in whatever avenue of greatness they can leverage. Their efforts are felt and appreciated, even if they aren't outwardly acknowledged.

After the Strike debacle, I was hopeful that things could start to repair and improve. I will say that on several fronts, SE has been more communicative when they have the time to do so. Becoming a mod here let me take a peek into the Mod Team and the TL, and in the recent months (in comparison to the months prior to that that I had read the backlog of), staff have made several posts that really dropped the veil on how things work internally, the current status of some things, and a few other posts that were just really genuine. I really, really value those posts. Even as I write this, Slate wrote a really illuminating post (mod-only link) that I really hope makes it, in some form, onto Meta not only because it's important to the network, but because it was a very genuine and informative response from an employee. I wish I could talk about it here. I'm sorry for mentioning it when I can't.

Further, after my appearance on the SO podcast, Philippe sent me an email. He said he listened to every word I had said, and I believed him, because he went on to point out several bits from the rough cut of the podcast. He then offered an open line of communication, stating that if there ever was anything that I felt needed pointed out to him, I should do so through e-mail, and he promised he would read it. While I'd say that I'm a bit sad (and maybe a little disappointed) to see that Philippe doesn't participate around the network as much as when he became community VP, I can tell in my communications with him that he really does care about the network. I wish he showed us that side of him more often. Even still, that genuine love for what works here gives me hope.

I love seeing questions like these. I love it when employees talk openly about what they're working on. I love being involved in the process of crafting network greatness, because I feel heard and respected. My involvement with the SG beta showed me what community-driven development could look like, and I haven't seen a similar effort since. I want more of that.

So... There are really good people here who work to try and make this place great. Using that drive in a manner that serves the community (yeah that's hard to put a target on, I know) is your best path forward.

I'll close this out with a thank you to those who read through everything, especially SE employees who made it through. I wrote entirely too much, and probably said a bit that you disagree with, but you read it anyway. Thank you for listening.

Answer 10

I fear that the need for stackoverflow.com as it currently is structured has more or less passed.

It's not necessarily that people don't need answers anymore, or that people aren't willing to answer, but as a whole I don't feel it really serves the needs that real developers have today. It's great as a resource to find solutions to already solved problems, but... you can only find them using 3rd party tools and those tools now provide our solutions without introducing users to Stack Overflow, meaning we've lost the primary source of new users and without new users we aren't getting new answerers or askers. Pair that with asking questions on SO being such a daunting and difficult process and it's easy to assume why people aren't participating. I believe it is far too late to do anything meaningful against this problem given all of the historical data has already been consumed by these 3rd parties. No amount of blocking or partnering with them will reverse this. The best we can do is make asking less of a process and more likely to result in a positive outcome for those that do decide to participate.

When I joined the network, my purpose for participating was primarily helping other users with a secondary goal of exposing myself to as many problems as possible to grow my own ability as a developer. I don't feel like either of these two purposes really exist here anymore. The former maybe, but that would require motivated askers (with non answered questions) who turn into answerers... and without a reoccurring influx of new users every year things are only going to diminish.

Answer 11

I haven’t been around nearly as long as most, but I still have my opinions on the company.

I think the company cares about numbers far too much.

They care about the amount of users in total. The amount joining per day. The amount of questions and answers a day. They care about how many people visit their sites.

But do they care about the 1% of people who actually allow the company to function and thrive?

The consistent users, who answer and edit every day. The moderators, who moderate. Even the people who just upvote and downvote every day.

Do they care about how many spammers mods destroy? How many questions I edit and close every day?

I think less. They care less than they should.

---

Because guess what, StackExchange? If you lose that 1%, you basically lose your business model.

Give mods more tools, give the community more of a say, give the 1% credit for what they do.

---

I love the network and it’s respective communities.

But I shouldn’t have to love it less because SE has the wrong priorities.

Answer 12

Hey, it's starball speaking. Slate, I'm very happy you're starting this discussion and listening.

I feel like I need to preface my thoughts with a bit about who I am, just so it's established to the reader where I'm coming from, since no matter how hard I try, what I think of and "for" other people on this platform, and what I want for the future will be polarized to my own experiences and values.

So- about me. I'm a lover of music who somehow found myself studying computers. Among other resources like documentation, youtube videos, and blog posts, I benefitted from Stack Overflow significantly in my studies and continue to do so. When I first entered my program of study, the first time I went on Stack Overflow and created an account, I remember thinking I'd be crazy and depraved to get sucked into this gamified system. After I graduated, I did... Cheers to being crazy and depraved, I guess 🥂.

There was a period of ~2 years where I was heavily involved in answering questions about VS Code and CMake. I climbed the rep ladder and enjoyed it (too much for my own good). I participated in a few burninations, hung out in the SOCVR chatroom for a while, made bug reports and feature requests, sent a few charcoal reports, and took part in beta tests. I earned a dup hammer and tried to make good use of it, often going doing rabbit trails in web searches to point people to canonicals. I drive a humble, steady stream of traffic here- several pages of link-sharing-badges-worth. I was a hair-width away from being elected as a moderator on Stack Overflow earlier this year, which I'm quite proud of.

These days, due to being busier (and that's a good thing), I've reduced my participation to deleting unnecessary comments and non-answer posts, and sometimes downvoting content in my main tag that I don't consider useful. Sometimes I review in Staging Ground. I am still somewhat involved on MSO and MSE, though less than I was a few months ago.

All that is to say- I have spent a fair amount of time and mental energy into trying to learn the ways of this communal library of information, and to make it a cleaner, more useful place. Playing the game made me care about it. Caring about it made me invest in it. Investing in it just made me care more about it (positive reinforcement(?)).

---

Who are we?

We are problem-solvers and makers. Curious people and experts. Readers, authors, and volunteer librarians and janitors.

The regulars are just total weirdoes. *avoids eye contact with my reflection on my screen*. I half joke, but that's genuinely what I think too. The deepest layers of this place seem to attract a certain genre of oddballs.

In our (speaking of the contributors as a whole now) idealized form, I see us as something like the study group (think private group chats or walled gardens like Discord), the office hours (think maintainer-operated forums like Discourse), and the helpful colleague (think LLM) but designed for time and energy efficiency at massive scale.

Where do we want to be?

Our idealized form, I hope. Or as close to it as we can get.

(My following responses to this and the following question will be a bit self-centered)

As for myself, I don't have the time to be an "author" anymore. I just want to be empowered by the community library's sponsors as a librarian and janitor to clean up garbage and organize the stuff around me. I want a library where volunteer librarians can flourish, because I have come to want this library to be a place people love: An unmaintained community library will grow into a dysfunctional mess. Readers don't go to dysfunctional libraries unless they're desperate enough to gamble a bunch of their time and mental energy. Authors won't freely give their books where readers don't go. (okay, the anology is stretching- have mercy).

what do we need to do to achieve it?

In reference to my above response,

You (the company) trusted me enough for me to build a track record as a librarian and janitor- trust me for my track record.

Recognize the value of the sum and continuation of my contributions. Make me (and others like me) feel like an asset rather than an obstacle. Maybe even a bit as a partner.

Fix my broken tools. Regularly devote some time and money to give me good tools. Don't just fix the stuff that's easy to fix. Fix the stuff that's easy to fix and which makes my everyday contributions easier and less frustrating.

Put up nice signs in prominent places that encourage visitors to have good etiquette like "lower your voice" and "no littering" (think JIT guidance and better onboarding- things I have been mentioning for years now, and other people long before me).

Stop delivering broken things and then running away to deliver the next broken thing.

Come say hi once in a while.

This is not the first time I've said words like these. Other people have said them before me too. But I'm happier to be saying them in this context than the others.

I've focused only on what I want the company to do because I'm lazy, and it's the easiest thing to do. The company holds the platform and its mechanics. The community holds the norms, and I'd say a larger share of the guidelines (though those go in hand with the platform mechanics). Thinking and talking about evaluating and changing norms and guidelines is hard, and we (the community) are still a disparate group with different opinions and personal variations of the shared vision. It also doesn't feel as important to talk about how to do better when the mind is occupied with wanting things to not (in one's own eyes) keep getting worse outside of one's control.

---

When you think about the future of the network - years from now - what worries you? What makes you anxious? What do you look forward to? What might make you hopeful?

Symptoms of derilict

I am not good at long-term thinking years ahead. But looking at Stack Overflow's site analytics, the consistent downward trend in new posts and votes seems to suggest a combination of entering a semi-maintenance mode (many questions have already been asked), a steady decline in the general population's inclination to seek answers to their questions here, and a decline in the ranking favourability that SO has in web search engines. It's not so much these that directly worry me.

Searchability and concentration of useful content

I'm worried that the content people want to find will not turn up for them when they look for it. There are a lot of things that threaten this.

• Questions that are just unlikely to be useful to other people due to lacking an appropriate amount of issue isolation, generalization, or abstraction. These pop up endlessly and dilute the more valuable content.
• Bad question titles. These frustrate the user experience of going through search results.
• Lots of bad questions and questions with bad titles, and the continual growth of this problem.
• Poor in-site search for technological(?) reasons.
• Poor web search engine search for content quality reasons like the two categories I mentioned above.

It gives me hope that the Staging Ground has graduated from beta. It shows a commitment to quality. Quality now has its own phase before release to experts for further review / answering. It weighs on me that it seems like reviewers are much more on the ball than askers.

Ignoring the problem of goodness/badness being in part in the eye of the beholder, I wish the stuff that I consider bad was easier to get rid of so that it would bother people less.

For the good stuff, I think it would be really interesting (where applicable) to be able to capture what search queries led a user to something that they try to upvote. As an answerer, I'd love to know how to optimize the searchability of content I write that people find useful. Or even of content I didn't write but that I often see duplicate questions of. As an expert, I have blindness to the mental model of non-experts ("curse of knowledge"). To an extent, that information is available for frequently duplicated questions through their duplicates, but I don't see anything incentivizing or encouraging people to improve searchability of canonicals given that information. I know you didn't ask for feature-requests here, but thinking about problems made me feel a little inspired.

External search and voting

I'm worried that people will stop voting.

Voting is a core piece of this platform's model. Voting is what makes the good stuff rise to the top. That's a core part of this platform's value.

It was accepted and embraced by the creators of this platform that external search was/is a driver of incoming traffic. Now with LLMs that do web searches, there's less need for searchers/consumers to come directly onsite. That means less eyes on our voting buttons.

Ideas about integrating voting into LLM response interfaces are stuck with the messiness of trying to guess that the user would vote and how, or that they don't care to spend extra time to go cast a vote onsite, or don't have the privilege to.

I recognize that I'm echoing concerns behind company initiatives that I've challenged in the past year or so (the designs- not necessarily the concerns). I just don't know what a good solution is (though I've had related ideas that I still believe in and would like to see implemented/tried). At this point, I don't know what to hope either. Maybe that if we can do something about all the content badness making searches ineffective, people will click into the site more.

I have a hunch that what we do will still be needed, even if just to be a source that feeds the next generation of machines for the next generation of new questions. It gives me hope that I see in external discourse that significant chunks of the general population think the same way.

Divide between community and company

I started becoming heavily involved here in ~2022. When I saw this divide, it shocked me. I went down rabbit holes reading about things that happened in 2019 and it was a really unhappy experience (for me just reading about it) fueled by morbid curiosity. I regularly saw lots of discontent and plenty of anger on meta and in chat. I came to accept it as the standard of living here- at least on meta. Then I got to experience it with front-row seats with all that followed the release of ChatGPT. And though there have been some resolutions like the strike negotiation results, and some wins like the announcement of Community Asks Sprints, the net trend does not look up from where I'm sitting and what I want to see. Any lift forces I see are not strong enough.

I no longer worry that that trend will continue downward. I have silently and unconsciously come to expect it to. Maybe it's just that I'm no longer fanatically involved, and am now busier with the rest of my life, but recently whenever something new pops up from the company that goes in a direction that concerns me or that I feel opposed to, the upswelling of reactive emotion inside me immediately fizzles out into a disappointed sigh.

I worry that the trend of the company being less and less involved in the community and involving the community less and less will continue.

It also weighs on me that the company appears from where I stand to have a thing for laying off staff who act as bridges in the divide, representing the voice of each side to the other. I worry that this will continue and we'll lose the remaining people who do that. I don't really have hope that more such people will be hired- just hope that those who remain get to keep their positions.

New blood

I'm worried that slowly we will lose the incoming stream of rare, crazy people who get deeply involved in maintaining and improving the libraries- who dilligently operate the review queues and try to do a good job- who vote up/down/close/reopen/delete/undelete.

I have a vague but strong feeling that it's unnecessarily difficult for a new person who wants to help out to discover and learn how. That worries me.

I also worry that it will become harder and harder for those who still come aboard to gain the privileges to make valuable contributions. Though in some ways, it also doesn't feel like too big of a problem.

Getting enough rep to make edits without peer review is so difficult. I'm not sure yet whether I think that's totally a bad thing. I have other ideas on how to make little improvements, but the edit queue still needs more people operating it.

I feel hope when once in a blue moon, I see a new face getting deeply involved- performing moderation, doing reviews, participating on meta, appearing in chatrooms.

I'm worried that new people will see all the shouty, angry, or just generally annoyed voices on meta and think they need to be like that to fit in. Have some grace for me when I say that that's what I thought for a while. I can very much be a people-pleaser. I think I've learned the voices of the regulars of meta and chat, come to distinguish those that I admire and want to emulate from those I just observe and sometimes carefully interact with (and those in between), gained enough confidence in myself from building up somewhat of my own reputation here, and come to feel less intensely about this platform- all that enabling me to (hopefully for the most part now) have a more neutral tone. I'm not perfect, but I think I'm going in a direction I'm happy with.

I feel hope when I see or remember the people I admire continuing to be themselves and continuing to be good librarians amidst... everything. I hope they keep going, and I hope I become more like them as long as I'm still active here.

---

My reflections have meandered down many inconclusive pathways, and prompted questions about the network that have no easy or simple resolution.

Well, now that makes two of us :)

Answer 13

The company has no attention span.

We have core components of the network that were worked on - like the new editor, that seems to have been suddenly abandoned. There's other examples of this.

SE's focus has jumped from public Q&A to careers 1.0, to Teams, to now trying to shoehorn AI somewhere. In between, we had things like Documentation which was an attempt to move beyond Q&A. But Q&A is still the core of what SE's about and what it's good at.

We've also had very positive steps to better the network and relationships that suddenly got dropped cause something else is more important to the management.

Apologies to people with ADD but the company has corporate ADD, and once a good, solid Q&A system was 'complete', has been floundering about the 'next thing' to the detriment of 'what it was good at'.

Answer 14

Creating a canonical reference / library, and solving specific people's problems, are at odds. Canonical Q&A pairs are an exceptional situation, often requiring out-of-band negotiation (on meta, in chat – occasionally it's handled in comments) to establish: and then once they do, bye-bye personalised support on that topic.

These things do not need to be at odds. We can't really afford for them to be at odds. To build a useful reference, our collective awareness of problems needs to be up-to-date; and to build a library, our canonical references on each topic need to not be diluted by a dozen badly-answered "halp i dont unersand, fix teh codez plz" questions; and to be genuinely useful, we need access to the partial and iterative knowledge represented by those badly-answered "halp i dont unersand, fix teh codez plz" questions. (I swear: the number of times one of those questions, even unanswered, has contained the information I needed…)

I'm not saying the solution's tooling, but the solution is tooling. A radical proposal (aimed at the Trilogy):

• All questions get an extra (required) meta tag: "help" or "reference".
• "Help" questions are allowed to duplicate each other (by current standards), though the context needs to be sufficiently-different that you can't copy-paste an answer. (MCVEs / minreprexes are still expected.)
• Marking questions as "reference" is an earned privilege. (New user questions will necessarily be "help" questions.)
• Marking a "help" question as a duplicate is not considered closure. In fact, duplicate votes can be Accepted. (New answer type, maybe? Or else, tooling sugar, like the original duplicate system / current auto-comment system.)
• Sufficiently-canonical "help" questions can be turned into "reference" questions by changing the tag.
• A "reference" cloning an existing "help" is not considered a duplicate, if the "help" question could not be edited into a useful "reference" within the bounds of usual edit guidelines (e.g. not invalidating existing answers, not conflicting with the author's intent, not putting words into the author's mouth).

I've been thinking this for a while, but never had the opportunity to write it up. This solution's off the top of my head, and is by no means perfect: but it should somewhat convey my view of the problem.

Answer 15

Well - the community model is broken, and there's a deep mismatch in expectations.

We occasionally have long delays in community, and community moderator requests. The community has made specific requests in terms of the model and sometimes composition of the community team but these get ignored, and either we get promised things will get better when there's more resources (which is sometimes temporarily true), or from my perspective, we're gaslit, and told that the actions taken seem to be the right thing.

And well, the people who're most passionate about the network, and want to see it thrive are also the ones most affected by this. I'm asking myself "will this be taken in the spirit its written, or seen in the worst light possible" or told "this is how things have to work".

At the risk of repeating myself (in various different ways)

• We don't seem to have enough people in the community team to deal with the workload. I've written a post about this in the moderator team (I could write one for meta if/when there's interest and I feel its the right time), but I've heard (privately) entirely contradictory information. At the very least periodically downsizing people for short term survival doesn't quite work when at least part of the issue is long term trust.

• many aspects of the social contract are broken, and there's a disconnect between the desires of the 'core' of the community, and what seems expedient for management to try to meet their goals

• perhaps it would be a worthy idea to have a small fully front facing subteam within the community team - dealing with short term, quick fixes and feedback the community has. I tend to get pushback for asking for the 'old' ways of community management being deeply embedded in the community, but I don't feel its something that's incompatible with the 'professional' model of community management. In a perfect world, I'd like to see both folks embedded in new communities, and working with 'key' new ones but I'd consider at least one person whose job is primarily community communication a massive win. SE's historically had people with these skillsets, and probably still does.

Consider the responses that 'active' devs and SREs get vs say, various AI related announcements.

• we need investments of time, effort, inconvenience and money in rebuilding trust sustainably. If something is of benefit to the community, but isn't 'what we do now' - rather than dismissing it, it would be nice to see some of these things happen. This probably ties into the 'other' core issue I see - that of lack of focus.

• People in positions of leadership are not engaged in a visible way with the community. I do realise people have different leadership styles but after all these years I'd expect folks to at least occasionally engage in the company's alleged flagship product.

Answer 16

First of all, thanks for asking. Unlike previous attempts this one appears genuine and coming from the heart.

Took me whole night (well, the hours before sleep) to figure out what Stack Overflow was, how it was supposed to be when created, etc. And just now I noticed balpha already nailed it, and in an excellent way. I really couldn’t say it better and this saves me from writing couple of paragraphs. Also, same as balpha, I write “Stack Overflow” as that is the flagship and engine, but of course I refer to the whole network and/or the company.

Now that the background is set, let’s jump to the future, aka what comes to my mind when I think about the future of Stack Overflow. Sadly, the TL;DR here is not good. I see no future for Stack Overflow in its current form. Unless it would go through a really drastic change, 10 years from now it would not exist anymore as public Q&A site. (the paid version would stay.) That’s something balpha also mentioned in his answer, but naturally I have to elaborate and refer to the other questions of what troubles me and what gives me hope.

Note: all the below is my own personal opinion, based on what I see and hear. I don’t have data to back it up, and in some cases I really hope to be wrong and see things as worse than they really are. But the discussion here does call for honest opinions, so that is what I give.

---

Why wouldn’t Stack Overflow survive? Because critical mass of core users are going to leave and/or stop curating. This will cause the site to be flooded with low quality posts that would not get edited, off topic questions not closed or deleted, and it would become impossible to find decent answers. After a while, views would decrease, new questions won’t be asked, and new answers would add little to no value. By all means, that is considered a death for a Q&A site. And don’t get it wrong: this mass Exodus has already began. Still not critical mass, and that’s what gives me hope it can still be handled and perhaps even reversed.

Why the core users leave? Because they are unhappy with the direction the company choose to follow. Because their actual concerns are neglected at best, or plainly ignored at worst. Because the company is trying to cover the real direction with noise and bells, instead of being straight forward and address painful conflicts. This is, in big part, because of the huge elephant in the room, that balpha mentioned as well. Shiny and golden, sparkling with potential and promise, the elephant persuaded Stack Overflow to ride it, to put their future in it. That by itself might not be bad and possibly the only way to survive in the short run, however that is not the point at all. The point is that it was done without proper communication with the above core users, except for some very poor attempts, that (again, in my opinion) were never real. The course was set, is still being set, regardless what the actual users think. And that worries me and troubles me.

However, the GenAI hype is going to subside, like a bubble it can’t possibly keep growing forever. The company will have to change direction again, find a different course to sail. But by then, it will be too late.

So, how can such a future be avoided, and instead have a thriving site and community in 2035? Drastic changes are required, no doubt. The company needs someone who is both involved deep in the “new generation” of programmers, as well as having the “spirit” of the first days of Stack Overflow. Kind of “these days” Jeff Atwood, who can think out of the box. Another way is to harness the power of this very community, and start a separate, official, discussion calling people to give their suggestions how Stack Exchange should change in 10 years, including any drastic measures they can think of.

But no matter what changes will be done, it can’t work without complete collaboration with the community members. Each step must be discussed, and feedback not only collected, but actually have impact. That is the thing that is missed the most.

---

Thanks for reading, and I will sum things up with some extra. I do not think Stack Exchange is evil, I believe the company lost its original way when it was sold to Prosus, and Joel’s original promise has been broken. My trust in management is broken beyond repair, however I still see others who have some trust, and what keeps me around are the friends I acquired over the years, and the still existing desire to make things a bit better. The middle layer of the company aka the Community Management team also consists of people whom I trust, hence I am posting this answer, knowing someone from that team will read it, but I’m beyond hope it will actually have any impact. (For example, management changing their ways or setting a different direction.)

Answer 17

I feel like the company is slowly driving away community members one by one.

I know a lot of this has definitely been said before, but I'm worried about the trust issues between the company and us - the community - getting so bad that everyone decides to just up and leave. Actually, some people are already doing that. I'm worried that people genuinely devoted to the site, who spend actual effort investing in everything from clearing the review queues to talking in chat, will just leave, and have their spots replaced by people who just answer for the reputation points.

I'm worried because I don't want to lose the connection with a lot of genuinely great people I've met here on Stack Exchange. I've seen a conversation about the future - it was mentioned that eventually SE might devalue enough to the point where we can just buy it. I feel like that might actually have a chance of happening. Another thing said was that the eventual death of the site is basically inevitable unless it branches out into something new.

My mental image in maybe a decade is that Stack Overflow will remain on the internet, devoid of anyone willing to pay attention or moderate. The amazing, helpful questions and answers will remain, but in a graveyard that slowly fills up with spam. I'm trying to be optimistic. And this makes me sad because my entire generation, including most of the programmers I know, grew up with Stack Overflow. Got a problem? Stack Overflow it. Obscure bug? Look on Stack Overflow. Another thing might just be that almost all the easy questions are already answered - it's hard to ask a good new question that hasn't been answered.

A lot of people might just be willing to scrape chat for all the nostalgic moments and memories and then up and leave the site. I haven't reached that point yet, but I'm not sure I'm still that far away.

Answer 18

I'm bobble. If you've spent any significant amount of time on Puzzling Stack Exchange, then you recognize me. I am a very active curator there. I believe I'm the most active. Moderating on Puzzling has been my constant for four and a half years, from secondary school to grad school. I've made some public mess-ups, but I try to be better, to take criticism, to shift to changing public opinion (though I always end up on the stricter side of center), etc.

What do I do? Edit, vote, flag, review, push others to review, organize meta posts, run tag cleanups, write FAQs, comment to welcome users and/or explain rules. Day in, day out. I check the homepage at least twice a day (after waking up, before going to bed), usually more. I go through each post, editing and voting and commenting as I see fit. I go through the review queues. I try to keep abreast of major network updates that might be relevant to Puzzling.

Puzzling is a medium-sized site, standing at 3.7 questions/day as of writing. This makes it possible for me to get pretty near curating every single new post. It also means that I worry new users won't have a "chance" to start curating. I joke in my profile that I'm the "friendly neighborhood unelected moderator" - but if there's already a friendly neighborhood Spider-Man, and the neighborhood is small, who needs another Spider-Man? How do they start? How do they practice? How do they have room to stretch their powers, learn their limits, learn the community's limits?

There's an obvious answer to this worry, which is to just stop. My activity does drop occasionally, usually due to real-life vacations or similar. But... less gets done. Edits that I think would be useful, especially image descriptions/transcriptions, simply aren't made. Comments to explain rules or help users improve their questions get posted later or not at all. Most obviously, I've been the sole driving force behind every recent major tag clean-up.

I love my site. I genuinely enjoy curation. I want to see it done, and done right. But I have a nagging worry that what I'm doing is fundamentally unsustainable, because I'm too scared to let go of the reins to let others learn the ropes.

---

This worry is quite specific to my situation as a super-curator on a smallish site. Still, it's really and truly what's on my mind. These sites depend on a constant influx of new users, some of whom are supposed to take up the curation torch.

Puzzling is pretty different than the rest of the network, given how we're largely original puzzles, so some of the larger network issues (GenAI etc.) don't affect us as much. We're happy to chug along on our own under the radar. Anyways, Slate, that's my answer to your question.

Answer 19

You listed three major roles on this network in your question:

Consumers of the knowledge stored here, those who create and store knowledge, and the people who operate the place itself (not limited to employees), who all together produce the thing we call “Stack Exchange.”

My biggest worry is that those three roles don't agree on what Stack Exchange should be.

First - those who "create and store knowledge" - which I interpret as "the people who answer, edit, vote on, close, migrate, review, and flag posts" on a regular basis. Overwhelmingly, this group seems to be guided by the mission statement in the tour: "We build libraries of high-quality questions and answers, focused on each community's area of expertise." We care about the longevity of the information stored here - which can mean making good things, improving bad things, or filtering out the unrecoverable.

Second - the "consumers of the knowledge." I could see this as meaning "the people who find answers here here via Google," but that group generally won't be very engaged with this platform - I care much more about the people who ask new questions on SE sites. People ask questions for different reasons, but everyone who asks a question wants an answer - and in many cases, they want it soon.

And finally - "the people who operate the place itself." I'm going to focus primarily on the company here even though that's not the only group in question. I believe that many of the company's goals align with the other two groups' - particularly the first's - but as a for-profit company, Stack Exchange, Inc also has to care about making money. If the product stops being profitable, the company fails - so although content quality and UX matter, they kind of have to be second to making the numbers go up.

I've only been here for a couple months, but I can already tell that these groups' goals are diverging - and I worry that our communication isn't working well enough to prevent that.

An easy example is the plethora of "why was my question closed?" Q&As on Meta. Each one represents a divergence of goals: the asker wants their problem solved, and the community doesn't think answering unfocused or off-topic questions will help create that lasting library of information. But it also represents a failure of communication - because without things like proper onboarding, there's no way for a new user to guess that the community will react so badly. And that, in turn, gets back to the divergence in goals itself - because proper onboarding requires company intervention, and if onboarding isn't really profitable, how feasible is it for the team to spend time improving it?

I see Stack Overflow's Discussions as another failure of communication. Not in quantity of communication - there was plenty of community feedback, particularly as the spam started ramping up - but in its response. Many members of the community spoke up about the spam problem, but it took a long time for that communication to produce meaningful change. And like with onboarding, some of that can be traced back to divergence in goals - to a community member, a page chock-full of spam is anathema to the site's core objectives, but to Stack Exchange, fixing it just might not have enough of an ROI to make it worth the dev time.

And, of course, there's the gen AI fiasco. To me, this is the clearest case of both goal divergence and communication failure. To a new user, gen AI might seem great - it makes it way easier to ask and answer questions, sometimes churning out a 30+ minute post in seconds. But to a longtime community member, that same gen AI tool threatens to inject credible-sounding garbage into the library they've worked so hard to maintain. And to the company, AI may be receiving negative reactions from core users, but it's also the biggest, flashiest technology of the last few years - and capitalizing on it now might be the best chance to attract funding for a long time in the future. Communication challenges magnify all that - new users often aren't informed of gen AI policies when they first join, for example, and we've seen community-company dialogue fail time and time again.

None of this is to meant to draw attention to any particular issue or to over-emphasize the Most Recent Calamities. Rather, I want to highlight the structural tensions that I've noticed since I first opened Meta. So if you're asking me what I'm worried about? I worry that in the long term, these goals will continue to diverge without communication to bring them back together - that new users will see the community as hostile, that the community will see new users as lazy, that distrust of the company will continue to pile up, and that many of the company's actions will continue to deserve that distrust. This site can't survive without the people who asks questions, the volunteers who answer and organize them, or the company that keeps the servers running, and it definitely can't survive if our shared goals and obligations get lost in translation.

But I also have hope that good-faith communication will continue, that there are people who remember those shared goals, and that people on all sides will keep trying to keep the network alive. I don't think that faith is misplaced, either. Onboarding may not be perfect, but experiments like SO's Staging Ground are a step in the right direction. Discussions improvements may have taken a while, but they're starting, and even small changes have markedly reduced the amount of spam. And I know that people in the company care, Slate, because you took the time to write up this question and engage with the answers. So thanks for keeping this site alive a little longer - because if distrust and divergence are what's going to kill it, genuine, open, and honest communication like this are what's going to keep it alive.

Answer 20

I'm thinking that I still like this place. Not enough to really invest substantial amounts of energy and time (creation of knowledge is rewarding but also very exhausting) but surely enough to hang out and comment but also simply to read. What kept me here above all were the incredibly wise insights of many people, above all the original founders of the company, past members of staff and users. Reading their thoughts was a delight.

The platform is well thought out by now, the downvotes are at the core of the high quality. And the biggest cause of friction is insufficient guidance. But we can't hold hands all the time. Q&A isn't the solution to everything. There is no free lunch, not even in voluntary work. LLMs might scale better there as first response anyway but again they aren't free.

For years people rather ignored the slow demise (of Stack Overflow). Endless discussions about small changes. We are trapped in a local optimum. Not the worst but maybe not the same if we would start fresh today. That's why big organizations sometimes outlive themselves.

But then the company has no clue often enough and favor their pet projects, even if it's quite clear they won't fly right from the beginning (say collectives, discussions). If only they would listen more. 10 years ago it felt as if they listened to their user base, or at least to data/statistics. Lately... I don't know. I hardly understand the management gibberish in the blog or I have the feeling they extreme whitewash (social AI are just buzzwords) and do not always produce coherent arguments.

But it's not completely true either. We got the staging ground and the ask wizard, but maybe it's simply too little too late. One week sprint per quarter year may simply be not enough for solving all that needs to be solved.

Many knowledgeable people have left the platform already. Where are they now? How do they spend their time? The knowledge lives by the people. It's much less alive than it used to be.

In case it didn't become clear, the mission is to build a knowledge library. What else could it be? I honestly can't think of anything else. I really wonder what the company sees as its mission beyond making money, but hopefully it's not something nonsensical like community-driven AI. And the knowledge also needs to be consumed, but as long as there is a library, people will always find ways to consume it (if the system is open enough), I assume.

My insurance was always the data dumps and the content license. Money is involved so there is always an interest to prefer money making aspects. But I trusted management until some point. That trust is long gone. Making data dumps freely and easily available is the least the company can do. We know the history.

A non-profit like Codidact might be much better in that regard. If knowledge and software is both open source and freely available it's unlikely that the service will be misused; somebody else would immediately step in.

The decline of SO predates the advent of generative AI but surely it's a strong disruption in the way how we consume knowledge. It doesn't mean there is intelligence and the current hype about AI will ebb away at some point but it means that knowledge generators are deprived of credit where credit is due. Is it really fair and legal that machines use the content of every website to make a product and sell its output without financially compensating the knowledge creators in the first place? And what would be a fair compensation? This is a non-trivial question that nevertheless needs an answer. It's a wild West currently and favors the pirates over the farmers, the ones who actually grow the knowledge. That's not good. It's even a trap. We might actually destroy our foundations by allowing others to harvest knowledge that they didn't grow themselves. We might end up with a wasteland and worse off than before.

Even if I had the time and energy to invest in public knowledge creation, I might think twice in the current situation. I might instead feel that my work is unfairly used by big companies (and Stack Exchange Inc. is definitely on the side of the big companies there), which anyway have lots of money. I might kind of be on strike (knowledge generation strike) for the time being. I'm not against AI technology in general. I might look more favorable towards it in the future.

Creation of knowledge still makes sense and it's a worthwhile goal in itself, but the way it will be presented and generated and consumed will all change a lot from today. Adaptation to change in conditions is kind of normal but it needs to be the right adaptation.

In troubled times I would concentrate on quality. High quality is something not many people (or algorithms) can achieve. Only collaborations of humans supported by technology in a smart way can achieve that. Producing the highest quality Q&A might be a good argument for remaining relevant.

And maybe the human factor is also important. AI generated content cannot offer a human touch, the feeling that there is someone who really cares and enjoys the interaction. And the backside of it is human friction, the feeling that this is a really annoying place. Maybe the system can be designed a bit more towards the first. Can one strive for highest quality and still remain respectful and understanding all the time?

So, get your priorities right and start experimenting because ... not much to lose anyway. Present your data and listen to feedback. Formulate straight to the point. Tell us where you are going. Maybe then people will follow.

Anyway, posting here feels a bit like escapism, given that much, much more is going on in the world. I wish you luck with the continued development of the platform.

Answer 21

Personally, I’m most worried about the company’s actions, particularly in regard to AI and listening to the community.

There is now an AnswerBot on WebApps which was kept hidden from the community for months until it was accidentally leaked by posting AI answers.

I imagine said bot will soon start posting AI answers on WebApps intentionally. Then, the bot will likely be deemed successful there, no matter how bad the answers are or no matter the community thinks. It will then be moved on to flood other Stack Exchange sites.

With an AI answering every question in 30 seconds and the answers incredibly difficult to debunk, even though they are all wrong, StackExchange will become useless. Knowledgeable answerers will move somewhere else, and eventually so will the askers. If I wanted to ask my question to an LLM, I don’t need stack exchange for that.

Not only that, but you no longer listen to your community. Many if not most company posts (that aren’t fixing bugs or something like that) are very heavily downvoted. That’s bad and that didn’t use to happen. It means that the company is now out of touch with the community. We want different things, and the company seems to feel it can ignore the community.

This is very worrying, perhaps even more than the AnswerBot. As long as Stack Exchange listens to and values the community, I believe they will almost always arrive at reasonable decisions and generally not harm the site. But if they decide, as they seem to have, that the community is disposable, just a large data creation project for some LLM, I have no confidence this site won’t be destroyed.

The only thing that gives me hope, really, is last year’s strike. That was really the one time the company listened to the community. Sure, it took thousands of users, hundreds of mods, a 2 month long strike, and a lot of spam on the homepage, but StackExchange actually listened! That, if the community all works together, we actually can get through to the company. If Stack Exchange did something incredibly stupid tomorrow, and we all went on strike, I do think we could likely get through to them.