Time to Exit Beta?

Question

So, we still haven't quite reached the mark on questions per day. But we're killing all our other stats. It's hard to find a question here that doesn't immediately have a ton of answers and votes. We also get a very healthy amount of traffic. It's clearly NOT the case that people aren't finding us, or that they aren't engaged when they get here. It might just be the case that the writing process itself doesn't continually generate as many novel questions as (say) the newest programming language.

I know other sites have exited Beta without perfect stats. Philosophy had a very similar profile to ours when it made a successful pitch to exit beta. They had a devoted and engaged pool of users, like we do, and only a slightly better questions-per-day stat than ours. Is it time for us to make that push?

Please include in your answers any suggestions for moving this forward (if you agree) or any knowledge you have of the process. If you disagree, please explain why in your answer.

Answer 1

Thank you for pointing out the precedent from Philosophy.

We will probably never have 10 questions per day, for reasons already mentioned in other answers, and that's ok, because:

• We have sufficient new questions coming in, and a large body of existing questions that still get new answers.

• We have lots of high-quality content.

• We have many engaged users who ask, answer, help new users, and moderate the site.

• We have enough users at the various reputation levels to support the higher privilege thresholds that come with graduation.

• We have enough active users to support a moderator election.

• We have stood the test of time. We are one of the oldest sites still in beta and we haven't seen quality, activity, or the user base decline.

Update 2019-08-01: We are no longer a beta site! (Along with 28 other solid sites that don't happen to have 10 QPD.)

Answer 2

I think it is very definitely the case that writing does not generate the same volume of questions as programming. Programming is very analytic. It breaks problems down into smaller and smaller problems. But communication, by its nature, is synthetic. It is about how you put it all together. Fewer, larger, questions go with the territory. But if the audience is there, then that seems like a good reason to graduate us.

It is a mistake to apply the same thresholds and metrics to different applications of a model -- this is true across every field. Even if the model works for the new field, it is highly unlikely that it will produce the same metrics in every dimension. A good model can apply to many problems, but not if it is a straightjacket.

Answer 3

Giving the overwhelming show of support for this proposal, the moderators have posted a request to graduate. It's in the hands of the larger StackExchange organization now. :)

Answer 4

I think if we have four out of five Excellent, we should make the case to graduate. This is a solidly established site with plenty of active users and some great questions and answers. We have healthy Meta discussions about our scope and not many Immediately Closed questions cluttering the front page. Let's go for it.

Answer 5

I agree with Mark and Lauren.

Here are some reasons I think justify leaving Beta without a large volume of questions:

The coding and OS generate more questions because programmers (and network administrators and other techs) have very specific tiny detailed questions, bugs, and/or error messages to address, and they are indeed slightly different by the thousands, not duplicates that can get the same answer.

IMO this is just not true of "writing". In "Writing" this would be the equivalent of critique, plotting, or What to Write. Many Stack Exchange coding answers will both critique their OP's code and provide What to Write, cut-and-paste syntactically correct chunks of code that do the work requested. If the OP asks "where is my bug in this code" the answer is usually "write it like this". I know several programmers that have used dozens of these code snippets in their projects.

The rules of what is on topic here naturally restrict the questions that can be asked and reduce the volume of questions. I also think that guidance in writing is just generally broader; a site that did not provide specific answers to specific problems in coding, and only provided guidance in good coding style, would probably have even fewer questions than Writing!

I also think the more generalized type of Writing questions we answer will necessarily have a larger footprint than the specific questions about coding or grammar, and this broader footprint leads to more duplication and overlap of questions, and perhaps fewer questions if potential queries are short-stopped by a basic search.

Worldbuilding, for example, has a lot of questions because is like coding, the problems are specific. Map critiques and logic critiques (eg. on the rules of magic) are welcome, the equivalent of "What to Write" is not prohibited; e.g. "I need a flat plain in which a forest borders a desert, what natural phenomenon can cause that?" Or, "How can I make spiders go extinct?"

I am not advocating a change of on-topic rules in this post, just saying there is reason to accept we are a mature site without meeting some magic threshold of question rates. I should think that, like writing itself, the rate of production of interesting material is much less important than the number of fans!

Answer 6

A couple of years ago, we did some research on what makes a site self-sustaining. Overall, we concluded:

1. If communities that can keep their site free of spam and other nonsense, there's no reason we should consider shutting them down.
2. Sites that consistently field 10 questions a day will likely continue to have solid participation for years or decades to come.

Our long term goal has been to make sure that sites have the features they need when they need them. Initially, we assumed graduation was a binary state: either a site is large enough to host an election, justify a custom design, be ready for elevated privilege levels, and etc. But we've slowly come to the realization that not all sites develop the same way that Stack Overflow and other early entrants to the network have. In particular, some of my favorite sites don't get tons of questions as a result of their subject matter. So we are working on breaking up with graduation.

I talked with the moderators here on Writing and the consensus is the main concerns are with the "Beta" label (which makes Gmail seem rational by comparison) and the intentionally unembellished beta theme. (The images are missing from the blog right now, but you can still see them on the Wayback Machine.) Both of these issues will be addressed by the new customizable theme framework which is currently being tested on Meta Stack Overflow.

Now Writing will start off with the stock theme, but once that's in place, it'll be a lot easier to offer customization to sites. In the past, creating a site design has required a month or more of work from one of our professional designers. (One of my favorite designs, Worldbuilding took longer than that.) By building a customizable stock theme, we can reduce a lot of the inefficiencies in the process. As a result, we'll be able to provide design customizations to our "beta" sites. The current plan is to start with the oldest ones first.

With 173 unique sites, there's a ton of variation in what they might need. But I think we can build a system something like our privilege system to celebrate community milestones. Writing is an excellent example of a site that has been quietly building a solid collection of questions and answers over the years. That should be acknowledged somehow.

Answer 7

I think it would be a good idea to have this site graduate. Not only are there enough people to support a steady supply of answers to all the questions that may arise (99% answer rate means that basically all questions are answered), this site also has a lot of content that attracts people that are not regular users of SE. The following is an addition to what has already been said in other answers here.

Take a look around to see what I mean, for example with the search query views:20000

There are 57 questions with at least 20k views. A wild guess is that those views do not only come from the users of Writing.SE and not only from other sites of the network. This site attracts lots of people that simply search for things via Google and therefore this is an important site. Even if not so many people stay on this site and become users, that's still a lot of traffic that we are generating.

For comparison: WorldBuilding has the same amount of questions with such high view counts. We are on the same level as an established, graduated site when looking at this stat. Granted, the stat is chosen completely arbitrary and not included in any of the official metrics, but in my opinion that only goes to show that there is more to a site than the metrics that we can see on the Area51 proposal or on the right side in the little bar on Main.

When looking at even higher levels of views, such as 30k, we get even more results than WorldBuilding. At 50k WB only has 3 questions, while Writing has 19! Questions like What is the correct usage of “P.S.” in Emails? and How to write a polite reminder email? attract loads of traffic to the site and thereby to the network: ~600k and 2.4 million respectively! The highest question on WB is at ~68k.

I took WorldBuilding as a comparison mainly because that's the site I am most active on. StackOverflow as the by far biggest site on the network has 39 results with more than 2.4 million views, but a better comparison would probably be other Beta sites such as History, whose mostly viewed question is at ~900k or Philosophy, the graduated example from this Meta discussion which has two questions at ~125k and 130k views, while Writing has 5 questions that are at least at 125k.

The amount of people that find help on this site is an indication that this site is valuable and that we should graduate. There are viewer questions, but these fewer questions tend to reach a broader range of people looking for help than on some other sites.

---

Another thing to keep in mind is that we currently have an answer rate of 100%. Not only the 99% anymore that you can for example see in the beginning of this post. There are still a couple unanswered questions, but overall it's not even enough anymore for the statistics to declare this as 99%. That shows that you can find your writing related answers here - it might take some time, but in general every on-topic question will receive an answer at some point.

Answer 8

Yes, the time to exit beta has long passed; lonnnng passed, in my opinion.

I wanted to update everyone on what I've been working on because I've seen many discussions, proposals, and improvised solutions about how the Community Team might allocate various "graduation" features with or without a custom design (e.g. increased reputation threshold, elections, migrations, etc).

There hasn't really been a comprehensive solution regarding how this would work sustainably juggling ~173 sites, much less deciding which sites would be eligible and when/how it would get implemented. So efforts to handle each request one site at a time has been caught up in indecision, confusion, and back logs… which does not work. This request is almost a years old, arg.

I'm taking a different approach.

I've taken it upon myself to write up a comprehensive (i.e. sane <grin>) entire-site life cycle workflow which would include breaking apart everything once bundled up under "graduation", and enabling each feature as soon as a site can support it. It's roughly based on the premise I outlined here — But what happened to Graduation?.

It's hard to say yet how well it will be received in its first draft. Completion and full implementation would likely be at least a few months out, but I suspect this has a much better chance of getting this community what it needs rather than waiting for action on this aging, one-off request.

Answer 9

I think the site is lacking expertise; I do not mean people who answer questions, often with skill. I mean professionals who write for a living.
Frankly, I don't know those sorts would ever have the time.

I also think the question count per day is low.

I don't oppose, or really care, about the beta state changing; but there's a clear difference between the subjectiveness of this sub and the clear answers on the technical subs.

My understanding is it's just brownie points anyways. Solid meh from me. Seems to be doing enough as is.