This event has ended, for more info, please see our recap in The Loop, August: Community-a-thon and in our MSE feedback post.

As was announced in the Q2 roadmap and subsequently discussed on MSE, we will be hosting an internal company Community-a-thon event from Tuesday, June 16 through Monday, July 13 (inclusive).

Some of the goals of the event:

  • Improve Empathy:
    • Increase the degree of identification with the SE Network sites and communities, their tools, challenges, motivations, and dedication.
    • For experienced users: create a new account and try to feel again what it's like to be a new user.
    • For less experienced or new users: gain empathy for the different Communities and sites, and a greater understanding of this aspect of our products and services.
  • Participation Targets:
    • Aiming for a very high percentage of employees to engage on a site during the course of the event.
    • Engagement 1 and 3 months later is at higher level than before the event.
  • Product Familiarity & Feedback:
    • Help employees to know more about and to experience the way that the public platform product works: its strengths and weaknesses, its quirks, its pain points and rewards, etc.
    • Collect at least two pieces of feedback from each participant relating to their experience on different sites.

The event itself will include a number of initiatives aimed at encouraging all employees (regardless of previous experience level on the network) to visit and engage on any sites that appeal to them. Engagement for our purposes here includes posting questions and answers, commenting, voting, and editing, chat participation, and more. There will be an informal contest (structured to avoid too much gamification) set up to encourage employees to visit and engage on new sites and to participate consistently during the event.

We will be directing employees who are inexperienced on the network to use the resources that already exist for learning the ropes as new users. That said, we are also going to be setting up a few different avenues of internal support for new users, to provide mentoring and guidance when appropriate.

The main purpose of this post is to let the Community know about more of the details and goals surrounding this event. There are no official parts of this internal event that will require interaction with specific sites or Community members. That said, we would like to encourage interaction between staff and users whenever possible. To that end, a new chat room has been created. This room will be advertised to event participants as a friendly place to go and meet Community members, to receive advice on places to go on the network, or to just chat.

This post, along with the chat room, will serve as a place for interested Community members to talk about any ideas that they have related to this event and its goals, ways of increasing positive interactions between staff and the Community, and ways in which you can help us increase the overall level of empathy and understanding between the company and the Community.

We’re looking forward to this event, and hope that it will bring us closer to our communities by allowing Stack Exchange employees to experience and identify with the dedication of our users, their frustrations, motivations, and triumphs.

  • 2
    Sounds like a great idea. Looking forward to meeting you guys! May 21, 2020 at 14:40
  • 12
    Will there be hats? ;) May 21, 2020 at 14:49
  • 9
    Will there at least be cake? :P
    – Zoe
    May 21, 2020 at 14:55
  • 4
    @ZoeTheLockdownPrincess No that would spread the 'rona
    – LemmyX
    May 21, 2020 at 15:13
  • 11
    Cake and grief counseling will be offered at the end of this test protocol. May 21, 2020 at 15:13
  • Roughly how many staff are there?
    – LShaver
    May 21, 2020 at 19:33
  • 4
    @LShaver there is around 250 staff at the moment May 21, 2020 at 20:00
  • 15
    I participated in a similar event three years ago. I found it very rewarding on a personal level even though I was already very much involved with a number of Stack Exchange communities. One suggestion I have for y'all is to commit to follow-through. Last time we came up with a number of points of friction that were never addressed. May 21, 2020 at 21:14
  • 6
    @JonEricson we are committing to go through all the feedback and incorporate what we can into our roadmap. Cannot of course commit a priori to implement fixes immediately for any and all points of friction that are reported. I will also try to plan at least one follow up post here to report on how it went and on things that we learned that may be relevant or interesting to the Community. May 22, 2020 at 10:19
  • How about involving experienced excellent/selected members from sites/communities into the staff rather than involving staff into the sites/communities?
    – Pandya
    May 23, 2020 at 8:29
  • 3
    New Chat Room Jun 16, 2020 at 8:28

5 Answers 5


I'm happy to see this happen again! It's been far, far too long.

One of the observations that came out the last time y'all did this was that too many people didn't know how to get started: they struggled to find questions to answer, to think of questions to ask, or to find other forms of participation that fit in with their skills.

That problem is in some way inevitable when your primary goal is the broad "participate" instead of something more specific like "learn a new platform" or "build a corpus of good information for my field on the 'Net".

However, this is not a struggle that is unique to employees of SE, Inc! Indeed, it is something I've seen frequently over the years from folks who are entrusted to support their communities on Stack Overflow. I've even seen it in my current job, where a small group of people are working to help folks in the PostgreSQL community via outreach on these sites.

So with that in mind, I offer y'all a couple of sections from a guide I wrote for my colleagues. Perhaps if someone gets stuck, or despairs of where to begin... They might find this guidance useful.

Excerpt from: "A guide to participation on Stack Overflow"

What sorts of participation are valuable?

First and foremost, taking visible actions which reflect our values:

  • Posting answers which are helpful, informative and guide the reader (not just the original asker!) toward a better understanding of the subject. For maximum impact, prefer answering HOWTO questions (those asking for help accomplishing a specific goal or understanding a specific concept) vs. “debugging” questions (those asking for help resolving a problem in a specific application or configuration), but bear in mind that even the latter can be answered in a way which guides readers toward a holistic approach to problem solving!
  • Editing questions (or answers!) to improve clarity, readability, and use language and grammar that reflect well on both the asker and any answerers. Remember: the majority of readers will be future learners, and will follow the example others set for them!
  • Commenting on questions or answers to provide suggestions for improvement or clarification.

All of the above activities allow us to lead by example, both in technical acumen and social grace: we should always bear in mind the precepts of our Code of Conduct, and strive to reflect and encourage the best that our community has to offer!

There are also more subtle - but equally useful! - ways to participate:

  • Voting serves two purposes: elevating the visibility of useful information, and encouraging authors. The latter is particularly important from a community-building perspective: positive reinforcement for people who take the time to participate constructively helps to keep people motivated and enthusiastic! Down-voting is less important in general, but - when encountering posts with harmful or misleading information - also a useful tool to both motivate authors and safeguard readers.
  • Recommending helpful Q&A in other venues (Slack/IRC/Twitter), highlighting the contributions of authors and providing them with more widespread recognition.

A detailed exploration of these varied forms of participation follows...


Answers are what brings developers to Stack Overflow. A good answer can reach hundreds, even thousands of developers beyond just the person asking the question. It is therefore important to be mindful of where time and effort is spent when answering - and in maintaining the information conveyed in answers. Some tips:

  • Generalize problems. Most questions get only a handful of views; they are the "long tail", each helping a handful of people but in aggregate reaching many. In light of this, it is worth being judicious about where time is spent when answering: aim first to answer questions which are specific, but not so niche as to be uninteresting to anyone besides their authors. There's nothing wrong with helping someone track down a performance issue in their thousand-line query, but if you can instead provide an answer that shows them how to identify it themselves, you could help them and others with less effort. A specific answer that links to a guide or tutorial on a broader topic for more details can be an elegant approach, providing an immediate solution for readers in a hurry, and a deeper understanding for those with the time to learn more.
  • Choose topics that interest you. Look for topics you know about, want to learn more about, or otherwise pique your interest. This attitude comes across in your writing, and helps to engage readers.
  • Show your work. Keep a list of resources you draw on when answering: reference material, blog posts, relevant source code, etc. Citing these can demystify the material for readers, allowing them to learn not just answers but the thought processes involved in finding them. Many readers won't care, but for those that do this lowers one barrier to entry in their focal community. If you (or someone else) has previously written a tutorial or guide to the topic at hand, linking to it after providing a solution can be a great way to break down the process as well!
  • Edit the question. More on this below, but... Think of each answer as a mini blog post - with a prologue written by the asker. Taking a few minutes to help their work look good also helps yours to look good!


One of the more uncommon features of Stack Overflow is the ability for anyone to edit any question or answer. Among the most important uses are...

  • Tagging. Most askers pick 1-3 tags when posting their questions, for instance "postgresql amazon-rds-aurora". These tags ensure that the question is seen by people following each of the relevant topics, while also improving searches. Adding relevant tags to questions improves the chances that both the question and any answers will be seen.
  • Titles. A good title is essential if others are to find a question that matches their problem. Titles should be specific, descriptive, and reasonably short - for example, a question concerning an error message seen while upgrading PostgreSQL should include at minimum the error message itself, the versions of Postgres involved, and "upgrade". If the asker chooses a less descriptive title, a helpful edit can and should add these details!
  • Grammar and formatting. A question or answer that is clear and easy to read stands to do more good than one which isn't; helping authors to make their posts attractive helps them get attention, and improves the experience for readers learning about Postgres.
    Technical inaccuracies. This one needs a bit of care: fixing mistakes in an answer (or typos in a question) is useful, but it's generally considered a faux pas to change a post in a way that completely invalidates its author's intentions. Obviously, "fixing" code in a question such that the problem being asked about no longer exists isn't terribly useful, but for an answer this gets a bit trickier: generally, one can alter code or descriptions to fix unintended errors or provide a more robust implementation, but should refrain from replacing the author's preferred approach with an entirely different one in most situations. For example, if the author recommends using cursors, it would generally be inappropriate to replace their solution with a recursive CTE - even when the latter is demonstrably a better solution; instead, a separate answer should be provided which presents the preferred solution and explains its advantages relative to the existing answer.


Voting is how we indicate what is useful. The system uses votes to calculate a post score, which affects the order of answers, the visibility of questions, the rank of results within on-site search, and a few other things. This score is also used by search engines such as Google to identify useful answers, which may be highlighted in various ways in search results. Votes are also a signal to authors that their work is appreciated (or, in the case of down-votes, possibly problematic). Finally, votes are used to calculate the "reputation" score associated with accounts, which gates access to various privileges on the site. Any user with a reputation of at least 15 can cast up to 40 up-votes per day; those with at least 125 reputation can cast down-votes as well (the total # of up- and down-votes per day cannot exceed 40, and may be restricted to fewer in some situations).

Given the ease with which the voting privilege can be obtained, and the disproportionate effect it can have on both the visibility of posts and the attitude of participants... It makes a great deal of sense to vote as often as possible!

One cautionary note: DO NOT vote on questions or answers from co-workers. If done regularly, this is easily detected - and the consequences can be dire. More on this below.


Comments are Stack Overflow's token nod toward non-Q&A discussion; any question or answer can be commented on (by its author, or by anyone with >= 50 reputation); comments are generally considered expendable, even temporary, and don't contribute much to long-term value - but they can be very useful as a quick way of asking for clarifying details or simply providing encouragement to authors whose work you appreciate!

Beware: as with so many other sites on the 'Net, comments on Stack Overflow can be quick, thoughtless, and even crude; whenever possible, prefer answering or editing to commenting, and avoid being drawn into anything that looks like a pointless argument.


There cannot be answers without questions, and while there is certainly no lack of volume when it comes to questions there can be gaps in what is asked about. Asking a good question is an uncommon and too often unrecognized skill - so if you happen to be good at it, asking a question on behalf of someone else who is struggling can be a great way to help them out. You're even allowed to self-answer if you happen to find the solution yourself (this should be done sparingly however)!

Recommending (suggest questions to others)

Sometimes the most effective solution is just knowing that a solution already exists: if you find or know of an existing answer that addresses a given problem, pointing to it when someone encounters that problem (either on Stack Overflow or elsewhere on the 'Net) can be a great help! It's also a lot easier than trying to reiterate such a solution in, say, Slack or Twitter.

This can be another good "show your work" opportunity, helping new folks get over their apprehension and acclimate to the community.

On Stack Overflow itself, if a new question has been previously answered it is possible to flag these as duplicates, thus leaving a path for future readers to quickly find their way to the existing answers.

What sort of activity should be avoided?

As with any venue, Stack Overflow has its own norms that must be respected. Here are a few common pitfalls to avoid:

Voting for colleagues

My co-worker writes some great posts on Stack Overflow - I wanna go vote for all of them! But if I do that, the system will detect it and invalidate the votes. If I try to get clever, there's still a good chance I'll trip a heuristic somewhere and a moderator will find out and invalidate the votes... possibly weeks or months later! Voting "rings" or other fealty-based patterns of voting are considered fraud on Stack Overflow, and if discovered the consequences can range from mere invalidation of the votes, to suspension of accounts, to account deletion - definitely not something worth risking!

There's a more subtle aspect to this as well: it makes it harder for your colleagues to learn what sorts of participation are actually useful to others. Since you can't see who votes for your posts, there's no easy way to tell the difference between "my answers are useful to folks" and "I have co-workers who upvote everything I write"... That can (and often does) lead to some bad habits.

If you vote for one or two posts from a co-worker without realizing it, no big deal - but try to avoid it, for everyone's sake.


It can be so, so difficult to walk away when Someone is Wrong on the Internet! But, try to do so anyway. It chews up a lot of time, and generally produces little of value; present your argument and your evidence, and leave others to make up their minds. And if a conversation devolves into personal insults... DEFINITELY take that opportunity to close the page and do something else (or maybe flag such nastiness for a moderator first).

Long discussions

As with arguments, a lengthy discussion tends to chew up more time than it generates in value. When, exactly, this is true can vary - for a particularly hard-to-reproduce problem, it might be worth spending hours trying to draw out enough detail! But try to be self-aware here, and use your best judgement.

Overt promotion / partiality

Nobody likes spam, but... Nobody thinks it’s spam when they're doing it. Folks who appear to be using Stack Overflow only to generate links to or leads for their product/service tend to get a backlash from the communities they interact with - and that's precisely the opposite of what we're looking to accomplish! Don't be afraid to link to something you or a co-worker has written when it is directly relevant, but also don't be afraid to link to anything else on the Internet when it is relevant - and above all else, avoid linking to irrelevant material.

Good luck, and... As always... Have fun!

  • 5
    One excellent tip; comments on Stack Overflow can be quick, thoughtless, and even crude; whenever possible, prefer answering or editing to commenting, and avoid being drawn into anything that looks like a pointless argument. That's an UV on its own.
    – Chindraba
    May 21, 2020 at 18:08
  • 6
    @Shog9 Can I get a TLDR please? May 21, 2020 at 18:43
  • 11
    TL;DR: "Life is pain - keep looking up!"
    – Shog9
    May 21, 2020 at 18:51
  • 1
    That's a great excerpt from "A guide to participation on Stack Overflow". Where can we find the full thing?
    – Mast
    May 22, 2020 at 7:55
  • 1
    Just come work with me, @mast 😉
    – Shog9
    May 22, 2020 at 13:09

Drinking your own champagne is always a good plan. I hope it will taste well!

Question: what will the visibility of these new accounts be? I can imagine they won't have a [Staff] indicator (that would reveal that somebody is a 'Michelin Inspector' and would spoil the experiment), but what will happen after the event? If a (seemingly) active user suddenly stops contributing and doesn't react on comments, other users might be worried (though this happens now and then already.) Will the new accounts be merged into existing accounts?

Other users have conducted similar (small scale) experiments (starting a new account 'for science') and wrote about them here or on Meta Stack Overflow (I can't find a reference right now). If we can find it, it might be a good read upfront.

  • 2
    jlericson.com/2017/07/26/race_to_1k_1.html - SE staff did this in May-June 2017 as well
    – Mithical
    May 21, 2020 at 15:56
  • 3
    The event in 2017 was similar, but only included folks who already had accounts, who would start from 0 rep. So different (though overlapping to a degree) breadth and goals from this one. May 21, 2020 at 16:37
  • 6
    I would hope that if someone on staff finds a home on a site, they'd keep it up after the end of this event rather than dropping off - of course, I'd expect participation to dwindle some, even based on personal experience - I rarely participate on sites other than the Metas despite being 10k+ on four sites. I'm still here in general but I have a lot going on. In the case that someone creates a sock to get the new user experience, we could probably merge the accounts (if that's what the staff member would like). But I think many staff members will participate on their regular accounts.
    – Catija
    May 21, 2020 at 18:09
  • 3
    Staff indicators don't always come with diamonds, and most staff names aren't recognisable to the average user. If someone looks like a normal user and smells like a normal user, most people interacting with them probably aren't going to click through to the profile and see the "Staff" indicator anyway. May 21, 2020 at 20:00
  • 3
    @Randal'Thor point taken. And we aren't going to be going out of our way to identify people as staff who do not want to be identified as such (that would be beyond the scope of what we are doing here and probably isn't appropriate anyway). May 21, 2020 at 20:02
  • Nothing wrong with dog food ;) May 21, 2020 at 22:00
  • 1
    small scale experiment post
    – rene
    May 22, 2020 at 8:29

We will be directing employees who are inexperienced on the network to use the resources that already exist for learning the ropes as new users.

Just so I get this right:

  • Will you be tasking them to use existing resources?
  • Or: Will you be directing them to the existing resources?

If the goal is to make the new-user experience as authentic as possible, I would certainly avoid the latter¹ – after all, lack of direction to the existing resources is an ongoing problem². Even the first choice may be too much.

¹ and particularly a guide as posted by Shog9
² I acknowledge that this is difficult to solve and you made a lot of progress recently.


I read that there is going to be an informal contest. I guess that means no prizes for the winner?

What are things you are going to "score" participants on?

I would suggest that you are going to score beneficial contributions to the network. Perhaps the average number of (bronze) badges per site scored. As to prevent people from joining all sites and scoring the informed badge on all of them, to reach the highest total number of badges.

  • 2
    The points system isn't finalized yet, but it will reward things like visiting or engaging on a unique site during the contest, on any site once per day, points for votes received (up to a max), for participating in chat, etc. Was originally going to include points for unique badges received, but our data isn't aggregated on a network level to allow me to query it for the leaderboard (and I am not going to do something crazy like query all sites every day for stats). May 22, 2020 at 10:15
  • 1
    There will be small prizes for participants who reach certain point thresholds, but nothing too big. Mostly for bragging rights and to encourage folks. May 22, 2020 at 10:16
  • @YaakovEllis I feel a reinstatement if Swag comming ;)
    – Luuklag
    May 22, 2020 at 14:09

An interesting, though radical, element to the experience could be introduced into the Community-a-thon.

Include real new users who can, and will, freely communicate with a single staff about their experience.

The details, while incomplete from lack of planning on my part, should be reasonably simple to complete by the involved staffers.

Each staff member most likely has someone they know, in-home being the best, close family a good second choice, who has never used a site in the network, probably not even knowing any more about it than I do TikTok. (I've never even visted the site or looked into it more than how to spell it just now.) This person could be engaged to join at least two sites, one they are known to be a good fit for based on the staffer's knowledge of the person, and another which the subject personally finds interesting as a hobby or mere curiosity.

The staffer, after obtaining their contact's willingness to engage in the expirement, can join the two, or more, sites and begin engaging on them. The staffer should be willing to allow the subject to find their own way, offering now technical help with the site usage. (An unknown new user probably wouldn't have anyone to ask either.) The staffer's communication about the site should be limited to getting feedback about their experience. If the subject gets really stuck, the staffer can help them past a rough patch, and then back off again once past that point.

The staffer and subject should have their communications about the experience out-of-band. If possible, the subject shouldn't even know the username of the staffer, or be on sites where the staffer doesn't have an account yet, if not both. The out-of-band communication is intended to avoid the communities from knowing that this new user is part of the expirement and receiving different (better or worse) treatment from the community.

The staffer should be getting feedback on several subjects, including site-specific information.

  • Opening the accounts
  • Posting questions, answers, and comments
  • Finding and joining chat
  • Experiences with reactions to good/bad postings
  • Perceptions about "community" from their interactions
  • Reactions to the UX from a technical, usability PoV
  • Reactions to the site experience from a "fitting in" PoV
  • Other feedback needed to help the company make decisions

Sorry about the incomplete nature of this "proposal." I'm sure it could be fleshed out much better. I'm not going to try adding details for two reasons. First, I'm not going to put the thought energy, and time into doing so as I'm only pssing on thoughts to the team who might be interested. Second, the people who would use this will have their own ideas, surely have their own goals, and most likely would only use the concepts from here, not the details anyway.

  • It is my impression that the staff coming to these communities are new enough and don't require any further interpolation of an experience than the one they'll receive naturally.
    – Makoto
    May 21, 2020 at 17:21
  • 3
    A very significant percentage of the staff joining will fit your definition of the "new in-home user". And from the rest, many will be paired up with the kids experienced users in a very similar way to what you describe. Though we won't be able to formally include this in the event, your ideas will overlap with many of our feedback and mentorship related activities. Thanks! May 21, 2020 at 17:26
  • 1
    The "mentoring" aspect is the biggest one to avoid. I read mentoring as one staffer, established, helping another staffer, new user, navigate the supposed minefield of new membership. Random new users will not have that resource.
    – Chindraba
    May 21, 2020 at 17:36
  • 3
    @User777088 I hear and understand. We are still trying to figure this part out. But the goal with mentoring would not be to allow the new users to completely avoid the minefield. Rather, to help them get unstuck if they do find themselves there. The "get feedback from new users" goal is only one part of this. Another important part is to get more SE staff as engaged users (who will stay engaged after the event is over), so trying to plan out ways to help folks out while they are in the midst of onboarding issues is one thing that we are trying to plan. May 21, 2020 at 18:42

You must log in to answer this question.

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