We just posted to our blog about our The Loop: Our Community & Public Platform strategy & roadmap for Q2 2021.

Please see the post for details on the projects that our Community and Public Platform teams aim to deliver this quarter.

Public Platform Roadmap

Product Pillars April May June
Content Management • Review Queues: New reviewer onboarding • Outdated answers: allow users to flag an answer as outdated • Review Queues: Workflow updates
Inclusion ← New User Onboarding Discovery → ← New User Onboarding Discovery → ←New User Onboarding Discovery→
Grow & Scale • Automating site elections: Final incremental improvements Stacks editor alpha: Incrementally release improvements to alpha test → ← Stacks editor alpha: Incrementally release improvements to alpha test

Community: Strategy Roadmap

Product Pillars April May June
Community at our Center • Top Mod-Only Page Guidance
• Redefining Community Promotion/Open Source Ads →
• Moderator Messages Template Revamp →
← Top Mod-Only Page Guidance →
← Redefining Community Promotion/Open Source Ads →
← Moderator Messages Template Revamp →
← Top Mod-Only Page Guidance
← Redefining Community Promotion/Open Source Ads
← Moderator Messages Template Revamp
Community Builders • SLA & Process Improvements → ← SLA & Process Improvements → ← SLA & Process Improvements
Inclusion • Dev Survey 2021 → ← Dev Survey 2021 → ← Dev Survey 2021
Grow & Scale • Close Votes Project →
• Compiling Reasons for Closed Questions →
← Close Votes Project →
← Compiling Reasons for Closed Questions →
← Close Votes Project
← Compiling Reasons for Closed Questions
• Area 51 Lifecycle

We are interested in your thoughts and questions regarding our upcoming projects and priorities on the roadmap.

  • 2
    I'm particularly in favor of the evidence gathering strategy regarding out-of-date answers and look forward so hearing about what it kicks into the light also to see what that might mean for curators on the various sites. Thanks for this.
    – W.O.
    Apr 23, 2021 at 16:49
  • What happens to an answer that's successfully flagged as "outdated"?
    – Richard
    Apr 23, 2021 at 17:55
  • 2
    @Richard There's more info here on MSO - meta.stackoverflow.com/questions/406675/… Essentially, the flagging is to collect data about how many answers on SO are outdated and in what way.
    – Catija
    Apr 24, 2021 at 4:45

6 Answers 6


Compiling Reasons for Closing Questions

There’s no consensus or one single source of truth for question closing on Stack Overflow. There are lots of scattered Meta posts and discussions in a sort of “common-law” way, but no digestible primer on the topic. Nor are there concrete guidelines on when to close versus not.

Is that initiative limited to Stack Overflow, or is it planned for more communities / the network? I'm asking because it's sometimes used (especially by staff) to refer to the entire network (e.g. in the section "Moderator Messages Template Revamp", though that sentence is probably true if you interpret it as Stack Overflow the site, and perhaps even intended that way).

(Also: your question proves the need for multi-column cells in Markdown tables ...)

  • 3
    We're starting with Stack Overflow only. It may be something we do on other sites too but it may not, there's no concrete plan there.
    – Cesar M StaffMod
    Apr 23, 2021 at 21:01

It was tucked away at the end and easy to miss:

Area 51 Lifecycle

As we are planning for growth in both the Community Management team and in the Public Platform team, we need to decide how much time we can invest in moving forward with Area 51, in terms of:

  • Cleaning up tech debt and ongoing maintenance (from a dev perspective)
  • Supporting new communities as they graduate into sites (from a community management perspective)
  • Supporting the continued expansion of network sites… or not (from a product perspective)

That last line scares me.

Read it again. "Supporting the continued expansion of network sites.. or not" It could be taken two different ways:

  1. Whether new network sites will be created from Area 51.
  2. Whether sites that aren't SO are supported.

It isn't clear. The second interpretation is the one that all of the StackExchange sites have worried about for years. Ever since work on site graduation and site designs stalled, it seemed like at any moment the company could decide that it just wasn't worth it to keep on supporting all of the StackExchange sites.

I will be anxiously awaiting what is to come. Even if it means that the Sword of Damocles is hanging above the sites that I use.

  • 5
    A bit over a year ago, they fired the person responsible for new site development; the last two sites to be launched are a year old this week. So, effectively, the decision has been a firm "not" for a while now... The agenda is a question of whether that will change.
    – Shog9
    Apr 24, 2021 at 5:09
  • 3
    @Shog9 We're launching one in the next week or so, actually. More Crypto... Anyway, we have no plans to start shutting down sites now any more than we did when Ana announced that low-activity beta sites weren't at risk of being shut down as long as there were mods willing to keep them up. The elections work is nearing completion which will make holding more elections at the same time easier. But... we also feel like a lot of the recent sites have been underperforming and we're making a lot of crypto sites... which can be fleeting.
    – Catija
    Apr 24, 2021 at 7:20
  • 4
    The crypto trend is an artifact of who is able to garner the necessary momentum to blow through the Area51 process before aging rules set in, @Cat - the intent of those rules was to ensure sufficient interest to actually launch, but made some major assumptions about both the level of interest that external communities would have in a presence on SE, and in the level of promotion that key people would do for the idea; it was designed in an era when (e.g.) spinning up a Discord server or GitHub discussion board wasn't dead easy. Area51 needs to become a community building tool vs the hurdle.
    – Shog9
    Apr 24, 2021 at 14:18
  • @Shog9 ... I'm conflicted about that. The last two sites that we put in beta weren't crypto. They were Drones and Operations Research, so we definitely have had tons of Crypto, it hasn't been the only type of site we've made recently. But we used to intend it to be more for building communities - the deadline was a year and now it's ... three months per phase? But sites that should have been high-volume never got that volume because there wasn't enough existing interested parties, like A&C - there's so much scope there for that site to have been huge but it languishes still.
    – Catija
    Apr 24, 2021 at 17:00
  • So, the current focus is that A51 is a place for existing communities to petition to create a new site on SE. We even require that they link to said community - an existing forum, discord, OS project, etc... They are their own outreach wing - if the people proposing the site aren't willing to put in the effort of seeking out members, the site will not succeed... and considering our current ability to invest in new sites - that's... reasonable. We don't do outreach for sites that currently exist - we certainly don't have the bandwidth to do it for sites that don't exist yet.
    – Catija
    Apr 24, 2021 at 17:02
  • 1
    I certainly agree that there's room to do better - understanding why sites like Cooking, A&C, Photography, Game Dev... that should appeal to good numbers of people are so quiet and underused. But, as much as I appreciate the ability to create sites and grow new communities around a shared subject of interest or expertise... I think it makes more sense for us to start by better supporting the sites that we feel could most benefit from a boost and that already exist. Hopefully, once we have it figured out for existing sites, we can start looking at new ones again
    – Catija
    Apr 24, 2021 at 17:06
  • 1
    The fascinating thing to me now is that niche communities can sometimes be more successful than broad ones. MathOverflow is particularly niche and much of what's on that site is also valid on Math... and yet, that site is a recognized place in the Math community (or seems so at least) for Q&A for their level of math... and so it does pretty well. But I think that's because a site can be successful when it becomes the place to go for a subject. Even though there's tons of options out there for devs, SO is ubiquitous... so how can we get more of that for other sites?
    – Catija
    Apr 24, 2021 at 17:09
  • 5
    I think it's worth noting that the SO business model has now swung back to roughly what it was in 2009, @Cat - hosted Q&A. That stands in contrast to the model that motivated the creation of SE 2.0 (and Area 51), which was roughly "build prestige sites that can double as portfolios for their membership". To me, this means two things: 1) there's no value in aiming for the "big bang" stuff anymore, and 2) y'all REALLY need to double down on community building. Like, yesterday. It is the future of both the company and the network.
    – Shog9
    Apr 24, 2021 at 20:25
  • That sentence is crystal clear to me, they're shutting down Area 51 and won't work on any alternative to it. Apr 25, 2021 at 8:53

While I had some thoughts about new site creation, which I shan't repeat - where's some things talked about here that might be worth addressing specifically

  • Cleaning up tech debt and ongoing maintenance (from a dev perspective)

If memory serves A51 is very much a fork of the SE 1.0 - we've gone quite a ways from there. There were plans to replace it, but other priorities kept getting in the way.

Whether or not a model for new communities follows area 51 - its worth having a clear plan and following through on it.

I really do hope that the end result is a maintained/maintainable solution for new site creation though.

  • Supporting new communities as they graduate into sites (from a community management perspective)

Not properly bootstrapping new communities would be a mistake. Even if they have established communities, getting norms established properly and guiding them through the initial stages is essential. As far as SE goes, supporting communities, new or old should be a pillar of community management, as is working with them to help achieve graduation

Supporting communities should never be an if, it should be a how can we do better

  • Supporting the continued expansion of network sites… or not (from a product perspective)

The or not bothers me. While I don't expect the massive growth we had in the past, It seems most of the new communities are cryptocurrency focused - and I'm personally doubtful that they're individually sustainable or that we need a separate site for each one.

We need new communities, but rather than 'just' communities staking a claim - it might be useful to try to 'match' communities with the same basic interests. Instituting a 'common' cryptocurrency site might be a good start for that

  • Redefining Community Promotion/Open Source Ads

    Additionally, the program uses code no one maintains.

This is kind of a theme - we've a lot of tech debt from when public Q&A (or Q&A for that matter) wasn't a priority, and these tools are a good way to get goodwill and cross promote across the network.

While not a direct observation - I'm wondering what plans there are to *keep* technical debt like this to a minimum and ensure that tools like this are *maintained* incrementally, rather than periodically rebuilt?

  • Compiling Reasons for Closing Questions

There’s no consensus or one single source of truth for question closing on Stack Overflow. There are lots of scattered Meta posts and discussions in a sort of “common-law” way, but no digestible primer on the topic.

While probably a good idea - the reason we have 'common law' rules is because we often end up finding certain things work/don't work. As much as finding the reasons is good - it might be worth trying to find a more universal way to keep the close reasons updated and turning the meta-discussions into something more concrete easily.


Thanks for the update.

I can't help but notice that this was posted on April 23rd, at the end of the day - meaning that almost a third of the quarter has already passed.

This appears to be somewhat of a recurring theme. Perhaps you could focus on starting these roadmap discussions somewhat earlier next time, so that announcements like this can be made before or at the start of the quarter.

  • 2
    This is good feedback, thanks! Usually we are focused on finalizing the roadmap post at the end of the previous quarter, so writing about the finalized roadmap.... that comes next. We can definitely work on doing it faster, though, we're also working on kicking off new projects in the beginning of the quarter. I see the desire to get these out earlier, however. Feedback heard! Apr 27, 2021 at 14:34

I notice two of these roadmap items pertain to updating documentation:

  • Top Mod-Only Page Guidance

  • Compiling Reasons for Closing Questions

Something we've spent a lot of time thinking about where I work when considering documentation: Less documentation is better, but harder. Whenever we update documentation (or create new documentation) we try to take a hard look at what's been written before, with a goal to end up with less content than we started with, while still accomplishing our goal. A few ways to do this:

  • If we're creating new pages, are there old pages that can be deleted?

  • Is the existing documentation too focused on details that didn't end up being important?

  • Is the existing documentation scattered in too many places? If we bring it together there are probably overlapping bits that can be deleted.

It's easy to write up instructions about how to do a thing. It's harder to go through all the old documentation that talks about the thing and update/delete/abbreviate it. But in the end, you end up with clearer, more organized documentation that doesn't take as long to read.

  • 3
    For the closing questions, it's definitely the existing documentation is scattered in too many places. While I don't think we should go deleting any old meta posts, it's possible we clean up the Help Center to point to the compilation after it's done.
    – Cesar M StaffMod
    Apr 23, 2021 at 21:03
  • 1
    @CesarM agreed, deleting meta posts is a non-starter. But I wonder if after the outdated answers project is completed for SO, something similar could be done for meta.SE?
    – LShaver
    Apr 23, 2021 at 21:07
  • 1
    Something worth thinking about - is very much 'why' this information is scattered and working out a meta -> 'inline' documentation workflow to keep it updated. Apr 24, 2021 at 3:53

The roadmap, in generic terms, looks good. Everything on there is stuff that is either valuable, or pain points.

Couple of points of feedback, though.

Onboarding discovery

How will you discover this? Is the intention to go by action data performed by new users, or to ask them? If it's the latter, how will the cohorts be defined and what site will it happen on?

The post itself

Not sure where to provide the feedback, but - would it be possible to get a link to the transcript on the podcast at the bottom of the page? It's impossible to find where it is from directly and not possible to do anything but listen to it.

  • The podcast dosen't have anything to do with the post I think Apr 23, 2021 at 16:16
  • @JourneymanGeek As I said - I wasn't sure where to provide the feedback as I could not find a place to actually do so. :-( Apr 23, 2021 at 16:20
  • 1
    I'm pretty sure the actual podcast page should have a transcript but... wouldn't hurt to ask as a meta question on its own Apr 23, 2021 at 16:26
  • 2
    @SébastienRenauld discovery will include data analysis, user research, and a review of qualitative research we already have (meta feedback, new user feedback, etc). This is similar to how we've tackled recent projects like review queues and outdated answers. Re: cohorts, we are focusing on new users (those newly signed up) as a starting point.
    – Des StaffMod
    Apr 26, 2021 at 13:16

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .