52

For anyone who follows this site, you may have noticed that Area 51 has been churning out progressively weaker site proposals and more anemic site launches, about a third of which fail in private beta.

There is nothing inherently wrong with most of these subjects; it's just that we've largely tapped the audience of Area 51 who no longer provide enough momentum to build these sites FOR you. When a proposal takes 1-2 years to accrete enough passers-by to finally launch, it's not unusual to see less than half the committed users showing up at all.

I need to create a lifeline for Area 51 if we want to continue providing this service for avid communities with the means to build these sites. It would be nice if our site-creation process allowed us to experiment with more impromptu site ideas, but the current system is not sustainable in the old format.

Going forward, we generally expect proposal authors to have the means to see their site ideas through. If you are accustomed to throwing up proposal after proposal hoping somebody ELSE will build your site, Area 51 is not the place to START your community search.

Got a Community Needing Q&A? — Stack Exchange Can Help

If you have access to a healthy, enthusiastic community needing a Stack Exchange site, Area 51 is where you propose it.

Effective April 27th, 2018, we are implementing the following thresholds to focus on communities with the means to see this process through to a healthy site launch:

  • [UNCHANGED] Proposal authors have three (3) days to complete the minimum submission requirements (5 followers, 5 example questions)
  • A proposal has four (4) months to complete the Definition phase, and an additional four (4) months to complete Commitment
  • Abandoned proposals exhibiting NO activity for seven (7) consecutive days are subject to closure

If you have an engaged community ready to build a site, these milestones should not be difficult to complete. If we find that four months is not enough time for avid communities to organize their efforts, we'll make adjustments.

These changes are intended to allow Area 51 to focus on more-productive efforts and to stop luring unsuspecting users into activities with insufficient support to see them through. It's so easy to spot these futile efforts. If you’re organizing a community ready to build your site, then these changes should not present any problem at all.


Note: This is a general announcement. Please bring any specific discussions or comments about your site back to your proposal page.

23
  • 5
    Are these changes intended to be permanent, or are they just a temporary test to collect data? Apr 21, 2018 at 3:34
  • 1
    @damryfbfnetsi There are no plans to remove these changes, but if different numbers or different behaviors would work better, features are always subject to improvements. Apr 21, 2018 at 3:42
  • 19
    Perhaps another thing causing this issue is guidance. We tend to prompt users to just "go to Area 51" if they want a new site. We need to better let other users know that they should only propose a site if they already have a community interested in a site. Apr 21, 2018 at 3:49
  • 8
    Also, you might get a lot of complaints from users who were just a tad bit short of completing the requirements (e.g. just 1-2 questions or commitments short of it). Depending on how many such complaints we get, I'd recommend building in a short exception filter (e.g. if the proposal has finished 75% of a particular stage, it will be granted an extra month to finish it, but no exceptions after that month is finished). Apr 21, 2018 at 3:54
  • 1
    @damryfbfnetsi We will be requesting that proposal authors provide links or descriptions to the target community explicitly to help assure they understand the community prerequisite. (Incidentally, we're also adding a link to our site listings to have them verify if their subject might already be on topic elsewhere). Apr 21, 2018 at 4:03
  • So are you going to also do something about users who repeatedly create proposals that get zero-to-minimal additional audience involvement and content? It seems that a major issue for A51 is that most proposals aren't worth looking at, let alone joining, so people are giving up on expecting a decent proposal to appear at all.
    – Nij
    Apr 21, 2018 at 6:34
  • 3
    Is there any reason you obstruct reading by writing numbers twice (2×)? If you want to emphasise them, just use boldface.
    – Wrzlprmft
    Apr 21, 2018 at 8:25
  • 6
    The repetition is for clarity, not emphasis. The difference between 10 and 100 days is a heck of a lot.
    – Nij
    Apr 22, 2018 at 0:12
  • 4
    There can be nothing unclear about 10 either. Until you actually meant to write 100. Then it's extremely unclear. The odds of writing both seven and 7 when you meant seventy and 70 are insignificant; the odds of making just one mistake or the other are much higher.
    – Nij
    Apr 25, 2018 at 9:15
  • 22
    This is incredibly unfair. After 6 months trying to get automotive people to follow an autonomous vehicle proposal, without warning, it's been shut down. In fact Robert was still recommending changes and closing questions he didn't like on our proposal until a few days ago. How in any world is this fair? It's like playing a game of football and the referee changes the rules five minutes before the end so the other team can win. Ridiculous! Apr 28, 2018 at 22:35
  • 37
    It would be helpful to have answers or suggestions to the inevitable question: If I can't start building the community at Area 51, where and how could I create a community that I could bring here? This is a non-trivial question, as there typically is no sufficient pre-existing online community for the subject. Apr 29, 2018 at 9:28
  • 1
  • 2
    Kind of related, on MSE: Can we set better expectations for private beta? Apr 30, 2018 at 13:24
  • 3
    Elaboration on my comment on meta: Where should I start building the community for a new Stack Exchange site? Apr 30, 2018 at 15:17
  • 8
    Is the Area51 FAQ being updated accordingly?
    – Pandya
    May 3, 2018 at 5:35

11 Answers 11

52

I think it's interesting that Robert's post above focus so much on communities - the idea that a site proposal should be made with an existing community in mind is new to me, and none of the site launches I've been part of really had a true community before the site launched.

Fundamentally this is a problem because nothing in Area 51 is set up to foster community development. On launched sites communities develop through the Q&A process. But Area 51 is deliberately not a place where questions actually get answered. And you can't really use the Area 51 Discussion Zone to build a community either.

I know development resources for Area 51 are extremely limited, but I do have one idea which probably wouldn't take too long to implement: what about creating a chat room for each proposal, and then linking to that prominently on the proposal page? And when you create a proposal, it should be recommended to add one or two news feeds to the chat room related to the proposal topic.

4
  • 7
    That is a great idea. We set an external chatroom in slack (actually 2 days before the shutdown) but it would be so much easier if there were something in area51. I upvoted your post, the idea to show who is doing the downvoting would also be interesting. If communities already existed, why would they migrate to SE? Case in point LabVIEW. May 2, 2018 at 9:44
  • 2
    Agreed. I was promoting for a Cloud Computing site and had no problem getting people signing-up (most of them from StackOverflow), but they had no concept of the Area 51 interface and how they should submit and up-vote questions. It is a very unfriendly site and doesn't foster participation let alone community-building. May 9, 2018 at 1:08
  • 1
    I like a lot the idea of chat rooms, I was about to suggest the same! Where "linking prominently" means "veeeery prominently"; I think a chat room is essential for the building process.
    – user124831
    May 12, 2018 at 9:54
  • 2
    you had a good point about the community, but chat room is not the tool that will make it. Perhaps letting people actually ask and answer questions, like on an actual site, would help much more. Perhaps the whole staging process here is too heavyweight and conceptual and not enough WYSIWYG, not instant and simple enough.
    – Tomas
    May 23, 2018 at 7:18
37

Going through the proposal process quickly appears to be a good signal that a proposed site is viable (because failed proposals evaporate, I don't have statistics on how many failed in private beta after making the 4/4 deadline, but I don't recall there being very many). However, the opposite situation of taking a while to go through the process doesn't appear to be reliable.

If you apply the current criteria retroactively to the last five years of proposals, only about a quarter of current sites (in beta or graduated) would be around. Specifically, we'd have the following graduated sites:

  • Bitcoin
  • Arduino
  • Emacs
  • Sofware Recommendations
  • English Language Learners
  • Magento
  • Network Engineering

and the following beta sites:

  • Stellar
  • iota
  • Augur
  • Bioinformatics
  • Sitecore
  • Monero
  • Retrocomputing
  • elementary OS
  • Open Source
  • Vi and Vim
  • Craft CMS

In short, what you're doing here is quickly eliminating non-viable proposals at the cost of rejecting about three-quarters of all viable proposals (sites such as Aviation, Biology, Chemistry, and The Workplace would all have failed under the new system). Is this the desired outcome?

5
  • 11
    That's nice work, but unfortunately your analysis is based on an incorrect assumption. Applying the new criteria to older proposals assumes that Area 51 participation was always as it is today. Nobody's claiming the new criteria should have always been this way, but the regular audience of Area 51 is tapped — repeat proposals fill the listings with thinning support. We're looking for strong communities who might need a site, but just hoping that repeating these passive efforts will eventually (barely) get a site through is creating increasingly weaker and failed sites at an alarming rate. May 1, 2018 at 3:44
  • 1
    Great point, chemistry is a good site. Interesting to see 5 cryptocurrencies and probably eosio should be on that list. May 1, 2018 at 6:51
  • 6
    @RobertCartaino, restricting it to proposals from the past two years improves the ratio, but not by much: Stellar, Iota, and Augur made it under the 4/4 criteria, while Quantum Computing and Constructed Languages didn't. (And if you factor in the rather spectacular margin of error from the fact that only five proposals made it to public beta, you can't say anything about how the success rate has changed over time.)
    – Mark
    May 1, 2018 at 20:27
  • 1
    That a site like Augur would still pass the new rules shows the new rules won't guarantee a successful site. Some successful sites would've taken 5-6 months, some that barely seem to be alive now might have gone through commitment in a month. Ultimately a line does have to be drawn somewhere though. May 2, 2018 at 3:48
  • 1
    @curiousdannii totally agree. As a person formerly involved in cryptocurrencies, it's pretty obvious that having an SE site for a crytocurrency will add to the value of the currency. Some investors will support a proposal to increase the value of their investment May 3, 2018 at 0:37
26

In short, this undermines anything other than technology sites.

And perhaps that's the point, given the structure of how Stack Overflow currently makes money.

The site offers a great opportunity to sciences. The collaborative structure, the protocol keeping questions and answers on topic and useful, and the efforts to circumvent elitism all foster interaction helpful to the scientific fields and encourages interaction with the broader community in a way few places do.

But heavy-handed choices like these reflect a larger lack of passion and commitment SE has towards such topics. Indeed, most of the science communities (and anything non-technology) wouldn't be around under these new rules. Science sites indeed may not always bring in the traffic levels sites on topics like technology do... but their importance to the lives of everyone is great.

If you really wanted to address problems, you'd invite broader input rather than making choices like these abruptly.

Honestly, I'd venture that traffic on Area51 as a whole is down, not just the supporters to new communities. (Of course it's tough to check that given that requests like this one were not seen important) If so, maybe that's a sign that fewer and fewer users are even aware of sites or proposals that might be of great interest to them, as they're usually only found via the long communities list. It doesn't help that Area51 may not be very intuitive to newer users (it certainly wasn't to me) or that the Hot questions list has settings that hinder smaller sites most (such as the time criteria).

It's ironic that the site's Blog post the same week was on how StackExchange can be a more welcoming place.

You speak of how you don't want to lure unsuspecting users into wasting their time? Perhaps don't change rules without adequate warning so people haven't wasted months of efforts trying to help sites that fell just short in the previous run.

You make it unnecessarily difficult to search details because you delete old proposals, but would think quite a few current sites took multiple attempts before getting started. But this change doesn't appear to do anything to stop silly proposals or make it clear such proposals won't work to unsuspecting users, it only seems to quietly shut the door harder on the proposals that have had a bit of energy to them in the past.

If the goal is improvement, then there's plenty of room for it. Perhaps a social media sharing bar on proposals like the one on Beta sites, perhaps clearer presentation of required milestones for people getting involved in proposals, perhaps expanding the featured site to show up with the Hot Questions list not just on the SE main page, but within communities/questions as well.

In the end, StackExchange's very success hasn't come from users "who have the means to see their site ideas through", but instead from its format that encourages anyone with questions to reach out to the knowledge of others. If communities centered around folks who bring in their own wide networks of connections, our traffic wouldn't be so Google-heavy.

When all is said and done, the message here really seems to really be: don't even bother. So be it.

1
  • 5
    I totally agree, I don't think there's any chance for anything except tech sites to exist any more. But, I think the deeper question is whether or not there's a chance for anything except tech sites to succeed any more. For instance, I was really in to gardening.se the year it launched, but then I switched to Christianity.SE (after the Fall, no pun intended) and never went back. Most of the original non-technical sites were constructed based on programmers' secondary passions, these are apparently waning. May 7, 2018 at 17:48
23

Translated to English, this proposal is saying "StackExchange doesn't want to participate in building new communities". The community needs to be ready and Area51 doesn't want to be the tool for people to build a new community.

But this is not the way StackExchange flourished. It has successfully build a communities of people around various topics.

I think Area51 should do the opposite and facilitate people getting together in a community, helping them to get together and launch the site together. Provide them space to connect. Currently, it is perhaps too heavyweight and conceptual for this purpose, especially for non-SE users, not enough WYSIWYG, not enough instant. Perhaps letting people actually ask and answer questions, like on an actual site, would help much more to build community instantly. In a natural and user friendly way. It seems that Area51 has a lot to improve in this regard, but instead of facing it, it's saying "we are not the place to build a community". I know, it's easier, you get rid of the bulk of lame proposals, the same way you get rid of lame questions, but this is throwing baby with the bathwater. It's the easy way, but not the highest potential way.

21

This is a reasonable policy to apply to Area51 (although it was done rather abruptly).

But Stack Exchange should provide some other process or mechanism to create beta sites that have a good probability of turning into avid and active SE communities even though at the outset they lack a coherent community of at least 200 people to advance them through this more demanding Area51 process.

There are a number of obvious topics of broad interest that exist now as successful SE sites that I would not expect to have formed under this new policy. Some of them – especially the earlier ones – did not form through Area51 at all. Others did, even though it turns out that the core "community" of users that sustains them is far less than 200. (Law.SE is one I know intimately that is a good example of the latter.)

1
  • 17
    This. The new process will only promote already established communities - which will only likely be technology-related communities. I've tried for a few years with Urban Planning - and while the bulk of our users have recommitted year after year, we could never get pass through the main stages (this year we had a really good shot at getting to Commitment - but we just got shut down). Area 51 is already unwelcoming (interface wise), so getting non-SE users to come and commit to a site that is barely near fruition is already a difficult goal in itself.
    – Zizouz212
    May 18, 2018 at 4:40
17

There is some problem with the new approach: I have seen some sites proposed under the new rules advancing to the commitment phase. In the commitment phase they fall short on the criterion "100 committers with 200+ rep on any other site"—so there is still the need of an already existing community of experienced stackexchange users. A completely new community has no chance to succeed.

5
  • 1
    That's simply not true. Many committers gain their reputation after they commit, particularly in new communities who actively rally for support — How to help a NEW community get through Commitment Sep 20, 2018 at 14:09
  • 5
    That advice is well hidden indeed! What about adding it to the FAQ?
    – user138118
    Sep 20, 2018 at 14:20
  • 1
    jknappen, Already in progress (draft-in-progress for the new FAQ). I hope to integrate an improved version of that guidance DIRECTLY into each phase of the UI. Sep 20, 2018 at 14:29
  • 6
    @RobertCartaino ... and it is true yet. Right now a proposal for Mainframes SE was closed with 335 committers, but only 58 experience Stack users.
    – user138118
    Oct 26, 2018 at 8:55
  • If you are looking for an active community of mainframe users, there is always IBM Developer Answers. 105160 questions at the time of this post... Nov 12, 2018 at 6:05
16

Hopefully these changes will help, but I am skeptical. The real issue is addressed in https://stackoverflow.blog/2018/04/26/stack-overflow-isnt-very-welcoming-its-time-for-that-to-change/?cb=1.

Most new proposals mean gaining non-SE users. For many, the benefits of SE's QA model do not outweigh the costs of learning a new platform (particularly for older experts) nor the emotional costs of SE's condescending user base.

After all, someone can spend hours on a question or answer only to have it downvoted away without anyone bothering to take the time to explain why. This doesn't promote learning. It promotes frustration and rage-quitting.

After all, people spend their free time here. Who wants to be made to feel bad for their efforts in their free time?

2
14

Hmm, just an observation, but it turns out my hypothetical scenario in the comments actually ended up being true.

This proposal was closed for being just three commitments short of the required 200. Does this mean that the proposal is non-viable? Also, how did you choose the number four months, instead of, say, three or six?

Just asking.

1
  • 2
    In practice it is necessary to set the limits somewhere. Negotiating every single case separately is unfeasible and easily leads to unfairness. It is unfortunate that some end up just below the threshold, but such is life. But I do agree that the exact numerical limits should be reconsidered. May 8, 2018 at 22:05
10

From Closing a Proposal After 4 Months in Definition or Commitment:

Do you have a community ready to build your site?

We generally expect proposal authors to have access to a community to help build your site. Area 51 cannot find that community FOR you.

If so, then I think we should have a community wiki guideline for authors to find such community, and an example proposal to explain to that community why they should go to Area51.

-5

I think this is a good step forward. I am guilty writing a few proposals just because I think they are a good ideal, just to see that they are not. I think a lot of people get into a proposal, get passionate about it than sort of lose sight of if it is really something that is needed or will work.

I was disappointed to see a proposal I had been following closed when I came on today, but that is what it is. The particular proposal was on it's third maybe fourth try. Hard to say why it failed, might of been the example questions where to broad, or to focused, or it just didn't seem like an good way to approach the topic to people already involved in the topic someplace else.

I think what they are doing now will lead to better communities that will be able to get out of beta better. Having said that I would love to see proposals that are making good numbers of commitment's, but low numbers with committers that have 200 points or more go forward to beta. I think they will be able to work out the SE part of things. I don't think just because there are lot of experienced se users not interested in a topic, it indicates the topic is not viable.

-7

I really support this change. I've seen a pretty direct correlation between the length of time it takes for a proposal to meet the bars for private beta and whether it does well as a site in the long run. One of the sites I moderate, Arts & Crafts was a very slow burn, nearly running out of time in both phases. We made it through private beta but the site is still quite low-volume.

I attribute that to a lack of an enthusiastic base community for the site who were willing to create content. We have good answered stats and decent answer ratio but we get very few questions, which are an essential part of the process.

Considering this change, and the focus on bringing the community with you, I wonder if it doesn't make sense to lower the network reputation requirement (100+ users with 200+ reputation) for a successful commitment phase in some cases. If a group bands together to suggest a new site for the network and, in a short period of time, has gotten through the proposal phase and has 200+ users committed, does it make sense to roadblock them by requiring 100 of these users to have 200 rep on the network?

Perhaps a rebalancing or a sliding scale... for example

  • 100/200 200+ users
  • 75/300 200+ users
  • 50/500 200+ users

So, if a proposal has 100 committers who have 200+ rep, and they have 200 total committers, they're good to go, same as usual

If a proposal gets up to 300 committed users, they only need 75 users with 200+ reputation.

If a proposal gets up to 500 committed users, they only need 50 with 200+ reputation.

This would never remove the requirement - we can make 50 the bare minimum - but if they can really get 500 interested users and just happen to be a subject that very few existing users have interest in, they may just have what it takes.

It's worth noting that meeting these goals in short periods of time (<4 months) would be extremely rare. One of the few times I've seen it happen recently was with the eosio proposal, which would have made it to private beta about a month and a half sooner than it did if this was enacted. From the beginning, this seemed like an inevitable site because it had the support of someone willing to drum up committers. Whether the site succeeds or not is yet to be seen, as it hasn't started private beta but it will be interesting to see.

This seems like a good carrot to counter the painful stick of cutting the accrual time by two thirds.

Risks:

  • There's already a slight issue with users creating fake accounts just to get their proposal through A51. Y'all are good at catching this, so hopefully it wouldn't be an issue.

  • This is, admittedly somewhat complex to explain... so it could cause some confusion... but it's already somewhat confusing to people and the exact requirements are hidden behind a click button reveal.

3
  • 3
    Would it make sense to say that we want 50 experienced users plus 200 other users, where an experience one above the quota is worth two regular ones? That might be more transparent. The whole system of requiring prior SE experience could be reorganized, and it might be worth a separate discussion. Should users with more experience (2k+, 20k+, 200k+?) have more weight, for example? May 3, 2018 at 18:01
  • 2
    Looking at some of the 400 commits, barely having 100 SE experienced people site, what you're suggesting can backfire badly. Of course it depends on the individual people but lowering the amount of Q&A experienced people isn't my favorite idea.
    – Helmar
    May 7, 2018 at 18:41
  • Instead of "# people following this proposal", it could be "# reputation following this proposal" lol Apr 7, 2022 at 9:04

You must log in to answer this question.

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