Stack Exchange is failing its community

Question

I guess this is very much an intervention.

There's a few policies in the pipeline - here's one, and there's another one that's not been officially announced that a good many moderators are opposed to. I'll update when (or if) it goes through, but it's not great.

Tied in with the new 'push' to AI (getting 10% of staff on that) and the loss of community team members including one who was a pillar of the community he founded (which on the whole resulted in a 10% reduction in staff), makes me feel that the company doesn't really have a long term plan and that there's a loss of trust in the company's ability to understand the needs of the community and that we've a real voice in these decisions.

Change is often give and take - we've contributed thousands of hours to moderation and working with our communities, and to an extent, it feels like we're underappreciated.

We've got features we don't really need being prioritized and features that we've asked for left deep in the backlog. We've asked for more community hires to give the company better insight into what the community needs - but practically we still have the same policies that exclude anyone who is still interested. We've asked for better accessibility to chat and meta, and work on chat as well as notifications and it's not really anywhere in our future.

We got new voting buttons instead.

I've brought up many of these concerns in the past - and made no real headway. Listening to fellow mods, I feel like the current course of action will lead to irreparable damage to the community. We've made headway over the past few years, but I feel like that a lot of that progress could be lost.

After so many rounds of being let down, SE feels like the dad in this ad, with quite a few smashed piggy banks and broken promises littering the floor.

Many of these policies seem aimed at appeasing the 'worst' part of our communities, rather than supporting the folks doing the curation and other tasks of actually running the community. Yet we're having things like 'we're suddenly overruling the communities on ChatGPT posts' without any real support happening with no real warning.

We've also had multiple rounds of 'we need to downsize to keep the company healthy' which generally has ended up in the community losing the voices we had in the company, and leading to a much more difficult time in it.

In the time since we first posted this post - the data dumps, which essentially are our insurance policy should the company go evil, were no longer updated, though this might no longer be the case. There was no consultation or even prior announcements of this. It doesn't even stop use of SE data in genAI tools since most of the big companies can just scrape the information... Which feels like an unnecessary bridge is currently on fire. Since then, due to a lot of pressure this move was reversed, but the damage to trust has been done.

Stack's always been 'We need to do this now' and not 'how do we build competencies and trust for the future'. It's been short-sighted - admittedly in the same way that most tech companies seem to be, focusing on the next quarter rather than the long term.

I've... been working on problems that started a decade ago, chipping away at them before suddenly having a giant pile of new problems dumped on it. I'm royally frustrated because eventually there's going to be a point where the things that keep moderators going are simply outweighed by the problems caused by policies that really don't reflect the realities of the sites as they are.

I guess if you got this far, you're actually interested.

• Are there actually long term, 'we're sticking to this' plans to get to profitability and community growth? I mean if y'all are firing people, like Meta (who made a bad bet on VR), Twitter (which was bought by someone who made a bad bet on Twitter)... It means something is 'wrong'. We're invariably told everything is fine, and then things go wrong. This isn't the first time, and it's a case of 'say hi to the new management, same as the old management'?

• Are there longer term plans to make working for SE more attractive to community members - and to make a greater effort towards hiring in the community? Seeing multiple rounds of downsizing clearly doesn't help.

• Is there an actual plan for community growth as far as getting people to stick around? Sometimes it feels like it's a marketing term rather than an actual desire to do better. The community is people, not AI.

• I feel like there's a major disconnect, and a lack of understanding of the mechanics and culture of the communities. Are there any long term plans to improve this? We shouldn't see changes in core mechanics and community policies without consultation, and many of these seem to cater towards the 'outside' and the 'complainers' rather than the people who're active.

• Is there any plan for popular community requested features to get on the roadmap any time soon?

• we need a way to ensure community interests are protected

Answer 1

A related issue: the company does not know that it does not know how to be inclusive.

Accessibility

To abridge the Introduction to Understanding WCAG 2.1:

Anyone who wants to use the Web must have content that is:

1. Perceivable - Information and user interface components must be presentable to users in ways they can perceive.
2. Operable - User interface components and navigation must be operable.
3. Understandable - Information and the operation of user interface must be understandable.
4. Robust - Content must be robust enough that it can be interpreted reliably by a wide variety of user agents, including assistive technologies.

There are many general usability guidelines that make content more usable by all people, including those with disabilities. However, in WCAG 2.1, we only include those guidelines that address problems particular to people with disabilities.

WCAG isn't the be-all and end-all of "accessibility", but it's a pretty darn good start.

Blockquotes

Using images from Changes to block quotes in the pipeline and a screencap from Shakespeare's “say sooth” vs. “tell truth”.

Pre-history

December 2018

February 2020

Some improvements to blockquotes. -1 score, with many legitimate criticisms.

Now

This whole chain of changes has left us with a less-obvious, lower-contrast, harder-to-read blockquote. But the design is "cleaner", so it's okay, right?

Voting buttons

2022: A/B testing updated visuals on vote arrows. -288 score, with many legitimate criticisms, including the claim that the new design better complies with WCAG, which it does not. The company decided not to go through with this.

2023: Shia surprise! We are graduating the updated button styling for vote arrows

And there's more!

Here's a couple of other design changes that nobody asked for, received a lot of negative feedback from the community (and/or requests to revert them), but SE ignored and changed anyway:

• Please revert the line-height change!
• We are switching to system fonts on May 10, 2021
• Some changes to the profile while we make it responsive
• New post summary designs on site home pages and greatest hits now; everywhere else eventually
• New responsive Activity page

User agents (inc. screen readers)

The website is less usable than it was in 2008.

• Badge counts are now read properly.
• It no longer uses a table layout.

That's all I can say in its favour, really. I mean, just look at it in Dillo.

svn revision: 3240 (2009)	rev 2023.5.26.43462 via IA (with missing CSS)	rev 2023.5.26.43462 directly (with full CSS)

2,17855 gold badges2626 silver badges3030 bronze badges

Ah, I praised this too soon. Rather than implementing it properly, you used a brittle hack that just happens to work in the screen readers you tested. This doesn't happen with robust HTML.

(I have a transcript of Orca's output on this page, if you want to have a look. It's awful – and too big to fit in this answer.)

Chat

We are expected to move lengthy comment discussions to chat. Chat is not usable for all our users, and rooms created in this way are not exempt from automatic deletion.

New users

New users interpret moderation activities (downvotes, closure) as personal attacks, and get (understandably) upset. The solution is, clearly, to present our moderation in a way that doesn't upset them:

• Every "close" has its thorn: replace "close" with "on hold" for the first five days
• Let's send new users off to see the Wizard

And then, that stopped being the obvious solution. Instead, we got the welcome wagon. Here's a timeline; some highlights:

• Does Stack Exchange really want to conflate newbies with women/people of color?
• Come Take a Look at our New Contributor Indicator!: We were encouraged to be extra-specially nice to the newbies (reasonable), but the system remained unchanged despite requests.
• The Stack Overflow I wish to build and participate in is no longer supported

Now moderation (e.g. voting) is down. Also, users get frustrated when they can't upvote answers to their questions. The solution is obvious!

• Allow users to vote on answers to their own questions, just like how they can comment.

Sike! As per We are graduating the updated button styling for vote arrows:

• This means that 1 in 4 new users attempted to vote, but could not due to system limitations.

Because voting is one of the lowest barriers to participation, we felt that this was the best place to start to help users engage more on the platform.

I think you can probably guess what the plan entails.

Minorities

I don't actually want to get into this again. So rather than any of the obvious examples, I'll leave you with a subtler one.

• Introducing the Silver "Census" Badge!
• The "loop" has arrived. What does the community think about it?
• Why is "The Loop" survey asking about race, age, and gender?
• (UPDATED 2023 May 8) - Changes to the Census badge

You're not… explaining anything

2018-08-17 Makoto:

I want a satisfactory answer as to why it matters

2019-06-21 Makoto:

Stack Overflow wants to portray that they have metrics and data that indicate that questions are improving with the new wizard, but if there's no data to back up that claim, then there's nothing to prove that it's accomplishing anything.

What would be the real relief? Demanding better tools.

2019-07-29 amon:

An increasing rate of incomprehensible actions

You're asking for feedback, but you're not listening to it

How is that inclusive?

Diagnosis

Here's my speculation, Stack Overflow.

• You think you are more capable of identifying problems.
• Once you have identified a problem, you keep it internal, and do not consult the community.
• Once you have identified a solution to that problem, you keep it internal, and do not consult the community.
• Once you have implemented the solution, you keep it internal, and do not consult the community. Occasionally, you consult moderators, but by this point things are basically set in stone. (The new Code of Conduct was an extreme outlier: you decided on the entire approach, and didn't even explain half the urgent motivating factors, but at least we managed to get some of the wording changed.)

You have some kind of self-directed Gell-Mann Amnesia effect:

Briefly stated, the Gell-Mann Amnesia effect works as follows. You open the newspaper to an article on some subject you know well. In Murray's case, physics. In mine, show business. You read the article and see the journalist has absolutely no understanding of either the facts or the issues. Often, the article is so wrong it actually presents the story backward-reversing cause and effect. I call these the "wet streets cause rain" stories. Paper's full of them.

In any case, you read with exasperation or amusement the multiple errors in a story-and then turn the page to national or international affairs, and read with renewed interest as if the rest of the newspaper was somehow more accurate about far-off Palestine than it was about the story you just read. You turn the page, and forget what you know.

Despite providing the infrastructure for tens of thousands of genuine subject-matter experts who are more than willing to share their expertise, you think your small overworked group of employees is going to do a better job alone than by asking for help.

Just, please. Tell us what you're doing. Run stuff past us before you've invested months of work into it. Heck, if you're embarrassed about it, run it past just the moderators: I promise we'll be kind, if you stop approaching us qua Dictator.

---

Have I missed anything important?

Answer 2

At the moment, Stack Overflow looks an awful lot like Twitter:

In spite of all this discussion of celebrity and checkmarks, my problem with the new Twitter is the utter incompetence of the new CEO when it comes to community management. It offends my professional sensibilities. You never know when he’ll do something off the rails like hobbling posts that link to a competitor.

Of course that’s also why I stick around. Everybody understands the sick attraction of watching a train wreck. But imagine what it must be like for railroad experts to watch one preventable accident after another. There’s a feeling of utter helplessness, but also morbid curiosity. The good news is that the running string of Twitter disasters is mostly only going to hurt people invested in Twitter. Honestly, most of us would be better off just giving it up.

I think insanely competitive people have a leg up when it comes to becoming leaders in a "tech company"[^1] but competition isn't always the most effective strategy when it comes to community work. It's especially harmful when the community was built on the foundation of collaboration:

Stack Overflow is, as much as I could make it, an effort of collective programmer community.

You might remember this was the post that included this image:

That picture, even though it was a throwaway joke, helped build a community that's substantially smarter than any one of us. Indeed it built 180 such communities. It takes a certain misplaced confidence or wild desperation to pivot away from that sort of community to a (so-called) artificial intelligence model. At the very least, take a moment to appreciate the goose before digging inside it for golden eggs.

---

Also like Twitter, there's no easy solution. If everyone picked up and left for Codidact, that would probably be the best way to preserve what these communities were founded on. It's, unfortunately, not as easy as a mass exodus (which is already very difficult) because so much of our history lives on this network. I suspect these communities will suffer slow declines until everyone suddenly notices the party is over—the Usenet route.

It's tempting to think a change in leadership will help. But remember it's unlikely you'll find anyone who has just the right characteristics to pull off what Stack Exchange/Stack Overflow needs:

1. commitment to financial success and
2. investment in community health.

Much better for the existing leadership to listen to the people who made this place worth investing in in the first place.

To that end, I recommend firmly, but politely, reporting outcomes. What that means is:

• Focusing on data whenever possible.
• Praising wins, even if they are not what we'd like or ideal solutions.
• Sharing negative outcomes in a non-judgmental manner.
• Show your work, whenever possible, when connecting an outcome to a cause.
• Don't be discouraged if it seems nobody is listening.

---

Nothing lasts forever. When I joined the Stack Overflow beta, I thought it was a goofy experiment that I'd play with for a while and move on to something else. I was right! It's just the experiment lasted for far longer than I ever could have anticipated and became a much larger part of my life. I'd love to see a change in trajectory. I even hope for it. My optimism comes from seeing what you all have been able to accomplish when you work together for a common goal. My pessimism comes from seeing how impenetrable the company has been to your charms so far. :-(

---

[^1]: Huge aside, but when will companies learn that technology always exists in service of some human need? Uber isn't a "tech company". It's a taxi company that uses mobile phones to replace dispatchers. Stack Overflow will flail away until it figures out what need it actually fills.

Answer 3

You're mixing a few things here that while somewhat related, are rather distinct issues. SE losing staff with deep experience with the network, and not hiring (intentionally) new staff with network experience is an entire separate subject. And the general prioritization of features has always been a bit iffy, it's understandable as it makes sense to prioritize based on business reasons to a large degree. Though as a whole the neglect of some basic issues with the network combined with the recent layoffs paint a picture of the future that doesn't look all that nice.

I think the main issue right now is that SE is pushing two extremely unpopular changes, and is doing that in a way that removes any hope some might still have had for a collaborative way of resolving issues. The GPT announcement was an edict, SE wasn't asking for our input, they just handed us the new rules. They didn't even hint that this topic was something that concerned them, we went from not knowing there was an issue at all to a very broad and categorical rule that doesn't leave any room for mods or communities to decide the rules.

This decision, if it is held up as proposed would remove any agency from the communities here, they are no longer free to decide how to handle artificially created content. That is a very intrusive step that came out of nowhere.

Overall the impression is that SE is making the decisions, and then telling us about them. We're not part of them, when we hear about them the decisions are already made. Which is not that unusual for a company, but one big difference right now is that they're touching core areas of the network with vast implications. And the community has no say at all here.

Those actions are seriously problematic on their own, but they also indicate serious issues in the relationship between SE and the mods and community.

Answer 4

Is there an actual plan for community growth as far as getting people to stick around?

There is no plan. They are just winging it.

I could be wrong. I really, really hope I am wrong, but I have seen this happening before.

Before the company was sold, there was still an idea or at least a need to maintain sites' reputation, quality, and sense of community because those were the selling points and integral parts of the product.

Now, we have entered a new stage of development, where chasing number figures on paper is the only metric for the CEO and upper management, along with throwing hot phrases and trendy features around to keep the appearance that there is some "important" work being done.

Quality, long-term sustainability, and therefore community engagement no longer matters. The sites, especially Stack Overflow, have some momentum that will keep them going for a fair amount of time, no matter how badly sites are managed, and whatever bad happens as the result of mismanagement will not happen overnight.

From now on, it will be just squeezing money out of sites as much as possible until everything completely falls apart, but at least numbers will be great while going all the way to hell.

And you don't have to worry, before we hit rock bottom the current CEO and management will be long gone as they will fish some new, better jobs with the great numbers they had on this one.

Answer 5

I haven't seen a lot of that from Stack Exchange in the past...well, few years, but let's focus on the last month. We've been dropped edicts in the Teacher's Lounge and then left to argue between ourselves with no follow up.

- Me, nearly 4 years ago

The quote above was related to the code of conduct change. No, not the one that just happened. The previous one (or maybe the one before that, I've forgotten). The point is, the criticism is still valid and applies here too.

On a holiday when the bulk of the staff was out, a new edict was dropped. An edict that should be taking effect immediately, but wasn't posted publicly. An edict that will fundamentally change how I am to moderate my sites and support my community.

And then they just left moderators to guess what was really meant. Do we assume Stack Exchange has thought of everything we are raising as counter points? Do we assume they missed some of these? How do you assume good intentions if absolutely no discussion was had ahead of time, so we don't have any context?

---

Moderator and community (especially community) input is being given lower value than it should. I think this is putting it generously. In another meta post after the awfulness of 2019's problems, I suggested 4 ideas to help rebuild community trust.

Bring the new CEO forward on Meta to answer questions from the community. I have not seen anything from the CEO since they started in October. At least, not on Stack Exchange. I've seen a fair number of reports in the press about them. Come forward with your thoughts and present them to the community. Defend them to the community. Present your roadmap for Q&A, Teams, Jobs and anything else that impacts everything we do.

Obviously a couple products aren't valid here any longer, but the key element is that the CEO hasn't talked with the community. Instead we get blog posts that the community dislikes. A lot.

Stop saying "legal action" and providing lawyered language for the CM team to parrot back to us. Someone who is actually giving the orders needs to come forward and say it themselves or it needs to stop. The community can smell it from a mile away and every time it's posted another round of this begins. So, let the people you pay to manage the community actually do their job. They are here to build bridges and support our growth. You are crippling that. The distrust you've built up is very rapidly making this a situation that can't be resolved in a way that keeps the meta community around.

I'm not sure where this new edict and ideas are coming from. The CMs are still relaying information.

Select a project or two the community really wants to be completed and complete it. Rebuild some of that lost goodwill. There have been plenty of betas, announced projects, half demos, collaborations and "good ideas" over the last decade to write a book. Pick something that is feasible, non-trivial, and is wanted. Work from there. Knock a few of these out and watch that support rebound.

I think Journeyman covered this with the original post. This hasn't happened.

Stop issuing edicts without justification or further communication. Just stop. I, as a moderator, am not going to be the next person to parrot your words.

And here we are. It's still happening. I still refuse to parrot your edicts to the community. Come justify this to the communities this network is built on. Defend it and explain how it will help make things better. Convince us - convince me - that I'm wrong and you've thought through these changes and it will work.

Answer 6

Is there an actual plan for community growth as far as getting people to stick around?

Oh, they think they don't need us. They just want us to be training data for an AI and then dump us as soon as its development is complete. SE thinks they just need us for a few months of training data and moderation. They don't care about the community anymore. They don't care what we think, what makes it harder for us, what the core values are, our trust, or frankly anything else. All they care about is making an AI and getting money.