It has been 6 months since the Facebook-StackOverflow partnership and mini-site were launched. Is there any data that can be shared publicly about whether the Facebook mini-site has been a success or a failure?

To me, it has generally felt like there has been little, and now dwindling, support from Facebook employees. Most of the questions seem to get very few views or answers, and are mostly from 1-rep users who don't return. There are large numbers of problem questions, including off-topic support questions (probably because Facebook doesn't offer much support to their 800 million users besides a FAQ).

Even the two small statistical items that are published on Facebook's weekly "Operation Developer Love" (a terribly misleading title, since Facebook doesn't love its developers) are bad (31% answer rate, 48% reply rate), especially compared to the site's average of 81% answered. But this is where statistics from Stack Overflow's side would help.

  • 70
    There was actually support from Facebook employees? I've never seen it.
    – casperOne
    Commented Feb 7, 2012 at 17:21
  • 45
    If their IPO information has told us anything, Facebook now exists for Zynga-like developers, which solidly puts it in the Axis of Evil. The Ayatollah Khomeini is more reputable than Mark Pincus...this partnership needs to be dropped.
    – user7116
    Commented Feb 7, 2012 at 17:21
  • 4
    While I agree that the Facebook tags have been flooded with garbage (see the jump in [facebook] in Bill's unanswered question rankings), I'm not sure I see a good solution for this. Beyond removing the filtered facebook.stackoverflow.com, we can't really stop a company from directing their developers here. Commented Feb 7, 2012 at 18:04
  • 9
    I must say I would like to see this Facebook scheme succeed, because this kind of cooperation is potentially a way for SE to make money in a win-win situation - if it works, of course. Either way, let's not forget looking at how bad are the Facebook tag's stats compared to other tags with a lot of low-quality questions like PHP or Android. I have an inkling things won't look that bad in comparison.
    – Pekka
    Commented Feb 7, 2012 at 18:07
  • 8
    @Pekka'sOrganicRepFarm yes but cooperation is a 2 way street and Facebook doesn't seem committed to it. It was more of an outsourcing / cost-saving to them.
    – bkaid
    Commented Feb 7, 2012 at 18:09
  • 5
    If I didn't know better, I would suspect that you do not love the Facebook mini-site.
    – Pops
    Commented Feb 7, 2012 at 18:09
  • Re 2 way street - yeah, fair enough. It won't work that way.
    – Pekka
    Commented Feb 7, 2012 at 18:10
  • Wonder if the FB sub-SO having its own dedicated moderator(s) would help the situation any?
    – chown
    Commented Feb 7, 2012 at 18:34
  • @sixlettervariables -- what do you mean "axis of evil"? I've never heard of this before for programming/tech companies ... is Zynga in it? Commented Feb 7, 2012 at 18:54
  • 4
    First thing I did was add all the facebook tags to my ignore list. The death of this nuisance will be something I shed no tears for.
    – Incognito
    Commented Feb 8, 2012 at 0:33
  • 6
    @OffBySome - I'm a Facebook employee and try to answer a lot of the questions too; though haven't had as much time to do so recently - personally i think adding a higher bar to entry of questions would be best, it's frustrating answering poor questions from first-time users who never return and never bothered to check if their question was already answered
    – Igy
    Commented Feb 8, 2012 at 10:37
  • 4
    @Igy I think it is important to work with the moderators to come up with a stricter policy but what is important is the transparency of what and how many Facebook employees are actively on your site. As stated under facebook.com/support: "Facebook and Stack Exchange have partnered to support the Facebook Developer community on Stack Overflow. Many Facebook engineers actively participate on Stack Overflow, along with some of the most knowledgeable Facebook developers, making it one of the best places for you to get help with your programming questions."
    – phwd
    Commented Feb 8, 2012 at 16:58
  • 12
    @Igy the part in bold may have been true at the start of partnership but currently it is not. If the Facebook Dev Relations Team feel overworked by the high number of low quality one timer questions, I think it is Facebook and SE Inc. responsibility to devise something to fix this because everyone on Stack Overflow is highly aware that Facebook questions represent some of the lowest quality questions on the site.
    – phwd
    Commented Feb 8, 2012 at 17:01
  • 11
    I'm the all time Facebook answerer and I can say that the mini-site is a mess. I gave-up on it ~3+ months ago! (my activity on site dropped significantly) Before the partnership, I was enjoying answering Facebook tags where I was getting a fair rep. (which is a good motive). Also when I see a question that I answered almost one year ago got asked 30+ times afterwards...it kinda put you down...
    – ifaour
    Commented Feb 11, 2012 at 7:43
  • The [facebook] tag -- the recipient of 50% of my downvotes, close votes and flags. Commented Mar 28, 2013 at 18:43

7 Answers 7


The data do seem to show pretty clearly that the answer rate for Facebook-tagged questions is astonishingly low and that the participation from Facebook employees has dwindled to almost nothing. We plan to bring this up with the company at the next possible opportunity, and encourage them to live up to their "developer love" motto. We'll also build a few simple reports that we can share with them to track progress in this regard.

  • 10
    I just want to say I'm glad to see that The Powers That Be are willing to listen to the community and modify things.
    – user176326
    Commented Feb 8, 2012 at 18:53
  • 6
    I think the feedback, especially if automated, would be the most useful in helping them live up to their end of the commitment. Once they can easily measure it, they can hold their support staff accountable, whereas right now they have no easy way to see the failure.
    – Pollyanna
    Commented Feb 8, 2012 at 18:54
  • Kudos. We'd appreciate it if you keep us appraised as these things develop (and you feel it's pertinent to do so).
    – casperOne
    Commented Feb 8, 2012 at 19:44
  • 6
    Funny, some time ago I read a post from a high-level guy at FB who claimed that they were sorry for all the mistakes they made with their api / treatment of developers, and that they were committed to fixing them. Guess we all know how that turned out Commented Feb 8, 2012 at 20:03
  • 5
    @Joel, it should be just like a stackexchange, if it doesn't live up to the StackExchange standards it doesn't have a place on the network.
    – Jonathan
    Commented Feb 22, 2012 at 19:22
  • 3
    Has the "next possible opportunity" come up yet? It's been almost six weeks since your answer.
    – casperOne
    Commented Mar 20, 2012 at 12:15
  • 8
    We will be out at Facebook HQ next week. The only reports we shared with them are visible on the tag "top users" page -- see for example stackoverflow.com/tags/facebook-graph-api/topusers Commented Mar 21, 2012 at 1:31
  • 13
    Any updates, year later? I see the partnership still exists so guess the rate improved? :) Commented Jun 12, 2013 at 6:38

I say yes, absolutely. The influx of low-quality and off-topic questions as well as the declining level of contribution from the Facebook Development team means that the users of Stack Overflow are getting the short end of the stick.

Note, this is different from whether or not Stack Exchange as a company is getting the short end of the stick; we don't know exactly what goes on behind the scenes, so there may be money/incentives involved which make the cost worth it to Stack Exchange.

My recommendation would be to revive the Facebook Developers Area 51 proposal and then start shoveling migrating questions in that direction.

The scope of the proposal should be widened to include Facebook policy, as well as Facebook apps.

There will be some overlap with Stack Overflow, and if the question is about programming against Facebook, then it would be on-topic here, otherwise, it would be on-topic on Facebook.SE (yes, I'm calling it now).

Then, we can retire the sub-domain on Stack Overflow as well as put to bed a bunch of Facebook sub-domain related issues (like this one).

  • 6
    "...short end of the stick" - You mean "...short end of the stack", right?
    – cdeszaq
    Commented Feb 7, 2012 at 18:02
  • 13
    Vouching for OffBySome and his flags here. There's just way too many people who don't get it. Commented Feb 7, 2012 at 18:13
  • @OffBySome Good point, once I added the link, that aspect of the statement was obsolete. Corrected. Thanks.
    – casperOne
    Commented Feb 7, 2012 at 21:14
  • @OffBySome Now that you've provided user ids, I'll take a crack at it.
    – casperOne
    Commented Feb 8, 2012 at 17:13
  • 3
    I say facebook itself should be laid to rest already.
    – Yes Barry
    Commented Nov 11, 2012 at 23:16

A quick follow-up to Joel's answer:

Joel, Alex and myself met with Don Faul and Charles Dowd of Facebook last week, to discuss ways in which we could work together to improve developer relations.

We expressed the concerns laid out here: that questions aren't getting answered by Facebook, and that the (lack of) documentation makes it difficult for anyone else to answer them.

They've pledged to make a real effort to getting more questions answered on Stack Overflow, both by increasing the answers posted by the developer relations team at Facebook, and by working to keep the documentation more up-to-date. We may also do some sort of SO dev <-> Facebook dev meet-up in the future, to help familiarize their engineering team with how participating on Stack Overflow can help developers gather feedback on their work.

All of us agreed that a big part of the problem is the large number of low-quality or straight-up off-topic questions posted to . Some of this is unavoidable - as traffic increases, so does noise; fortunately, Stack Overflow provides everyone with the tools to help clean this up. Some of this has been due to the ease of use of SO in contrast with Facebook's own support forums (combined the occasional lack of clarity on Facebook's end as to what actually belongs here). They're going to work on providing better support options for questions that don't involve programming.

My gut feeling is that Facebook is in a situation today somewhat similar to Microsoft in the early to mid '90s: a popular, fast-growing platform, with documentation and support for developers lagging behind. It took a while, but Microsoft caught up - by making developers a priority, even as their income came from sales of Windows and Office. Facebook needs to do the same.

Finally, thanks for holding our feet to the fire on this. Those of you who are active in the [facebook] tags are the heart of this community on Stack Overflow - if you're feeling slighted, overwhelmed or powerless to effectively moderate your own community, fixing that is a huge priority for us.

  • 1
    only time will tell... but this looks like a step in the right direction. fingers crossed.
    – Lix
    Commented Apr 6, 2012 at 0:03
  • 1
    @Off: as Lix notes, only time will tell. They wouldn't be the first company to put devs on the back burner; our goal is to let them know that they do so at their own peril.
    – Shog9
    Commented Apr 6, 2012 at 14:38
  • 4
    This part is very disturbing: "fortunately, Stack Overflow provides everyone with the tools to help clean this up." Why are we cleaning up their stuff?
    – casperOne
    Commented Apr 8, 2012 at 0:03
  • 3
    Same reason we've been cleaning up Microsoft's stuff since the end of the private beta, @Casper. And Google's, and Apple's, and Cisco's and Zend's and... The price of popularity is an endless stream of crap.
    – Shog9
    Commented Apr 8, 2012 at 0:10
  • 1
    @Shog that doesn't give them much incentive to help address the problem, does it? That said, I don't see the reason we indulge them any special attention.
    – casperOne
    Commented Apr 8, 2012 at 0:31
  • 1
    @Shog oh, and let's not forget that those other companies you mentioned don't have a dedicated sub-site; for better or worse, we climbed into bed with each other, we both have to do the work to keep it made. If we didn't have a sub-site then it would be just another tag and this wouldn't be an issue.
    – casperOne
    Commented Apr 8, 2012 at 2:33
  • Perhaps I'm misunderstanding you, @casper, but it sounds like you're suggesting we forgo moderation on [facebook*]... This... seems like a bad idea. Perhaps you could elaborate?
    – Shog9
    Commented Apr 8, 2012 at 4:45
  • 13
    @Shog9 No, we forgo the facebook.so subdomain; no special treatment, no branded logo, nothing; those questions get moderated the same as everything else, for better or for worse.
    – casperOne
    Commented Apr 8, 2012 at 4:51

Until someone from Stack Exchange Inc. or Facebook says something here is what I was able to pull up.

The following graph's data was used from Facebook Developer Love Blog posts.

You can see the data here


Graph of Facebook Activity

Missing data needed

  • Votes
  • Views
  • Anonymous feedback

With respect to the users themselves I don't how best to quantify other than looking at their acitivty tab.

Igy - https://stackoverflow.com/users/21062/igy?tab=activity
Cat Lee - https://stackoverflow.com/users/821917/cat-lee?tab=activity
Jeff Sherlock - https://stackoverflow.com/users/850494/jeff-sherlock?tab=activity
Colm Doyle - https://stackoverflow.com/users/508382/colm-doyle?tab=activity
Dhiren Patel - https://stackoverflow.com/users/855298/dhiren-patel?tab=activity
Jeff Bowen - https://stackoverflow.com/users/426839/jeff-bowen?tab=activity
Douglas Purdy - https://stackoverflow.com/users/845837/douglas-purdy?tab=activity
Matthew Johnston - https://stackoverflow.com/users/850947/matthew-johnston?tab=activity
dschultz - https://stackoverflow.com/users/850730/dschultz?tab=activity
squinlan - https://stackoverflow.com/users/318666/squinlan?tab=activity
Alexandre Couturon - https://stackoverflow.com/users/851643/alexandre-couturon?tab=activity
Paul Tarjan - https://stackoverflow.com/users/90025/paul-tarjan?tab=activity
c-abernathy - https://stackoverflow.com/users/850495/c-abernathy?tab=activity
Simon Cross - https://stackoverflow.com/users/851727/simon-cross?tab=activity
Jonathan Dean - https://stackoverflow.com/users/850570/jonathan-dean?tab=activity
Charles Dowd - https://stackoverflow.com/users/850573/charles-dowd?tab=activity
Scott MacVicar - https://stackoverflow.com/users/945990/scott-macvicar?tab=activity
J Starr - https://stackoverflow.com/users/469456/j-starr?tab=activity
Constantin Koumouzelis - https://stackoverflow.com/users/851194/constantin-koumouzelis?tab=activity
Gareth Morris - https://stackoverflow.com/users/850493/gareth-morris?tab=activity
Vikas Gupta - https://stackoverflow.com/users/910258/vikas-gupta?tab=activity
Alex Himel - https://stackoverflow.com/users/851019/alex-himel?tab=activity
Daniel Torvisco - https://stackoverflow.com/users/860170/daniel-torvisco?tab=activity

Igy is by far the most active of the bunch, but any more data such as number of consecutive days at the site will be up to the SO team to release.

Looking at the Graph alone, there is not much going on except a steady decline despite the December holiday period.

What's interesting to note is that previously Facebook Developer Blog posts were formatted as

received a reply from a community moderator or a Facebook employee

Now there are formatted as

XX replied, XX% reply rate

There is now no indication what percentage of this reply section is indeed Facebook Developer Relations Team. But, in the end it's all speculation off of incomplete data. Though at the current stance I highly believe there needs to be a stricter policy/guidelines set in place for this community.

I tried creating a separate proposal as well a Facebook Chat Room (now frozen/deleted?) in the hopes of getting the attention to an area solely surrounding Facebook but unfortunately Area51 is a different beast to understand; the proposal was closed as a duplicate with the reasoning

This proposal would tend to drain audience from an existing Stack Exchange site.

So how does it drain from the proposal while at the same time shows to be overbearingly low quality, low views and low votes? Hmmm, not sure (see first sentence).

  • I notice that the "questions asked" stat drops sharply right after the moderator elections. I bet they don't count closed / deleted questions in that listing. Greg Hewgill's charts show a sharp increase in closed questions overall as a result of the election, probably due to the activities of the new moderators: hewgill.com/~greg/stackoverflow/stack_overflow/stats.html Commented Feb 7, 2012 at 18:58
  • The reply and answer rates seem consistent but 50% reply/answer rate looks horrible when compared to other SE sites.
    – phwd
    Commented Feb 7, 2012 at 19:02
  • I don't think the methodology for how the blog post stats are created has changed, so I don't understand the dropoff in number questions asked - that data seems to support that the reply / answer rate is roughly the same as it has been since August - one thing to note is that (compared with our/my expectation) few of the answers get accepted as they were asked by first-time users who never come back - hence the inclusion of the 'reply rate' stat too
    – Igy
    Commented Feb 8, 2012 at 10:43
  • 1
    @Igy - As I suggested, perhaps closed and / or deleted questions are not included as part of the "questions asked" statistic. The sharp drop exactly coincides with the appointment of new moderators after the election here, and the number of closed questions on the site spiked up after that due to their activity. I know at least two of them have been doing extensive cleanup of the Facebook tags, so perhaps a large percentage of questions are now being closed and / or deleted as a result of their efforts. The fact that only ~30% of the question traffic survived this is troubling. Commented Feb 8, 2012 at 15:32
  • 2
    @Igy yes the stats around December were a bit unusual though this is the data that is represented in the Developer Love Blog. Data is provided to show something; either support is good or support is not. The raw data is in the Google docs but I am pretty sure the reply/answer rates are around 50%. This is bad there is no other way to say it. There are 11,320 unanswered questions in fb.so. That accounts for a 56% answered ratio. If Facebook.SO was its own site on stackexchange.com/sites it would be at the bottom of the pile. Most sites have around a 80% answer ratio.
    – phwd
    Commented Feb 8, 2012 at 17:13
  • 1
    @Igy I am hoping the point of the partnership was to avoid the barren lands of the Developer forums so whether it's a Facebook employee being given mod rights or dedicated users flagging questions every day to curb the influx as well as the tumbleweeds something must be done
    – phwd
    Commented Feb 8, 2012 at 17:15

The primary problem here seems to be the quality of questions being asked by new users. I believe this may be due to how easy it is to set up an account here when coming from Facebook. As I understand it, if you have a Facebook account, it's a one-step process to create a matching account on Stack Overflow.

I hate to suggest making something more difficult for new users, but what if we removed this fast-track process for creating an account based on your Facebook one? Right now, it's easier to ask a question as a new user when coming from Facebook than from anywhere else, and I believe that this is allowing for more drive-by users to ask terrible and off topic questions. It only takes a little effort to create a new account through other means, and someone who really wants a question answered will still overcome this minor obstacle.

I have noticed a significant drop in the number of terrible questions coming in since we started requiring an account to ask questions. Perhaps raising the bar slightly for new users coming from Facebook would similarly reduce the number of off topic questions being asked in the related tags.

  • 1
    Your last paragraph contains an interesting claim. It certainly seems logical. I wonder if there are any data which back that up. Commented Feb 7, 2012 at 18:37
  • 1
    I seem to recall seeing a (since deleted and I can't find a link) massive rant on MSO about how hard it was to create an account from one facebook user.
    – Flexo
    Commented Feb 7, 2012 at 19:17
  • 1
    @AndrewBarber Facebook.SO was released in late August and 1 month later users are required to log in. Looking at the close rate there's seems to be little change. However if we could see the Delete rate we might see a different story. Commented Feb 7, 2012 at 19:52

Is it time to re-evaluate the Facebook-Stack Overflow partnership?

Well surely that's up to the Stack Exchange overlords to do, not us as the community.

Regardless of whether there's a business relationship or not between SE and Facebook, we as the SO community should be carrying out our activities in a "business as usual" manner by "taking out the trash" and hopefully educating new users on how best to participate in our sites.

Just because Facebook don't throw much "developer love" at indecipherable "I haz cnt my like buton to wrk?" questions doesn't mean they don't care. In fact they cared enough to move developer support to Stack Overflow because we as a community are awesome and for two main reasons

  • Expertise in the subject matter
  • We don't take any crap and we don't allow bad questions and question vampires to drain our energies.

As a moderator I see some truly awful posts that surface there but our community votes and flags them to oblivion and this is good. This is why Stack Overflow is successful, because it shuts down the vampires and the folks who never give back, the Facebook subsite/tag isn't really any different from the Android or iPhone tags. Trust me.

Just because facebook.stackoverflow.com has the backing of Facebook doesn't really change how our community moderates. Crap is crap, get rid of it.

  • 1
    Could the partnership itself qualify as "crap" we should "get rid of" )?
    – chown
    Commented Feb 7, 2012 at 18:49
  • 5
    But, Joel says that they are just "taking care of it on our behalf", so, one might say that the community can help dictate the nature of the relationship, since they're doing it on, you know, our behalf.
    – casperOne
    Commented Feb 7, 2012 at 20:59
  • @casperOne - how the site is moderated, how we maintain a high signal to noise ratio, how we ensure we uphold the FAQ's I guess you're right. How they want to run their business I somehow think that falls outside the scope of what the community can dictate.
    – Kev
    Commented Feb 7, 2012 at 22:13
  • @Kev <insert snarkiness nere> Sorry for the confusion, but it was totally meant as snarky. I don't nor do I expect to ever have a say in the business strictly because I use the site or am a moderator.
    – casperOne
    Commented Feb 7, 2012 at 22:34
  • I think this is kind of an edge case. Of course SE, Inc. are free to do business with whomever they please; however, if the arrangement affects the day-to-day business of answering questions in that more askers of bad questions are driven there, the community does have the right to protest, and the management better listen. IMO.
    – Pekka
    Commented Feb 8, 2012 at 19:27

I'm new here, but, in regards to the original question, I think that the answer is "yes" but only because this experiment is deeply flawed. Just admitting that "FB is like Microsoft in the 90's" or any other company shows that your experiment isn't very generalizable.

Facebook reneged, but that doesn't mean others will. Hell, even if 10 companies had failed, just means you haven't tried varying other options. Even if whatever super-set of features you offer (like admin control of displayed topics in the sub-site) turn out to be a really bad idea, it's trivial to shut down a sub-site because it's just a filter tag + minor UI tweaks.

Sub-sites would stop the snowflaking of SO into a multitude of sub-quality mini-SE sites and the number of failed experiments in Area 51. If a sub-site turns out to be large enough or a company wants more control of the experience (and is willing to pay SO HQ for it) then push it into Area 51 and (eventually) an independent SE site.

Finally, this would address the scaling problem SE has in regards to new expert users. Rep inflation distorts the consensus approach and it sometimes feels like you are getting piled on by well-intentioned but mis-informed users.


You must log in to answer this question.

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