In looking at the stats for this site on Area51, it looks like we have a lot of areas classified as "Needs Work." I think this site fills a gap in the Stack Exchange offering, and I do hope it continues beyond beta. We are closing in on 200 questions, so we are definitely seeing activity.

What are the hard and fast rules around site graduation? Will we make it in our current state? Do we need to improve beyond "Needs Work" in all areas?

  • 2
    We'll probably survive. In the meantime, if you have a website, (relevant!) back-linking would be nice.
    – wizzwizz4 Mod
    Jun 7, 2016 at 16:42
  • 6
  • 3
    I think we'll make it. The last site I was involved with from launch went in a similar way. All those stats start as "Needs Work", it takes time to build it up and get things really pumping.
    – Matt Lacey Mod
    Jun 14, 2016 at 5:16
  • 2
    Compare Amateur Radio: niche subject, almost a thousand days in beta, Area51 stats that don't look too great at a glance, yet the site is still alive. (I'm obviously biased, because I held a pro-tem diamond there for a good while. Yes, left in good standing.)
    – user
    Jun 25, 2016 at 18:28

7 Answers 7


As I already said elsewhere, I think we should definitely try to market the site to a broader audience. I suspect a number of people think that this site is just for "experts".

(In fact, I originally thought that I'd never be able to answer any questions here because I'm younger than these technologies, and have never used many of them.)

It's not just for experts because you don't need firsthand experience with retro computers to answer questions here. Most of the time, an answer can be formulated purely out of digging through pages of Google. (That's what I've been doing at least, and it's been working so far.)

Personally, I will try to ask more questions, maybe about programming. At this point in time, we shouldn't be worried about asking anything "too easy" as long as it's an on-topic, non-duplicate question. It also seems like "flooding the site" isn't a necessarily a bad thing either (again, as long as they're quality questions), since we could use more questions. I may end up self-answering some, too.

As for answers, most questions have a great answer, but I doubt that any are perfect and all-encompassing. Even if current answers are pretty complete, there is merit to adding another answer that covers different details. It just makes the knowledge on that page a little more complete.


I've been monitoring the stats on Area51 and I have a few thoughts. Our biggest problem is definitely the number of users / traffic levels. I think that is just a function of retro-computing being a niche subject but certainly its something we can actively work on by trying to get the word out more.

The other problem is the low answer ratio. A lot of questions seem to get really well researched, thorough answers right off the bat thanks to the number of highly knowledgeable people we have on here so they often end up "done in one". I don't really see that as a negative though, its just a bit unusual for an SE site (I think).

  • 4
    I have seen a few instances where comments are posted that would just about count as answers in their own right - I think people may feel constrained to match the answer quality (and in a sense, a reasonable number of weak answers will lower the bar, so allowing for a 'hidden gem' answer to sneak in). Jun 7, 2016 at 17:27
  • 2
    Good point! I think when we see an "almost answer" as a comment we should point it out and try to encourage the user to flesh it out and post it as an actual answer. That might help a bit.
    – mnem
    Jun 7, 2016 at 17:31
  • 2
    It will help with both the answer ratio, and potentially give some users a higher rep too (which builds engagement if nothing else). Jun 7, 2016 at 17:33
  • 1
    Users asking questions should also be aware of/reminded about marking accepted answers too soon. From what I have read in the past,1-2 days after asking at a minimum. Because there is a tendency not to answer a question when an answer has already been accepted. Depending on the topic, I might not even read such questions.
    – user3169
    Jun 8, 2016 at 16:48
  • Questioners shouldn't accept an answer until they've reasonably verified that its correct, right? Or is that a whole other kettle of fish?
    – mnem
    Jun 8, 2016 at 18:16
  • 1
    I think that getting the word out can be a "two birds, one stone" deal. I was doing the CV queue on SO and I saw something that would have been better suited here. So I recommended they try here next time, and they said they would.
    – Laurel
    Jun 10, 2016 at 20:17

To add to the existing answers: I have been involved with a few other beta site creations - although not as deeply involved as this one.

It is normal in my experience for there to be a flurry of activity in the first week or two, then to see a dip as things settle. I would expect activity to pick up.

As has already been said, it is quality that drives activity that drives graduation. It is true that not every beta gets to graduate and some remain in beta for a long time. I can't see The Powers That Be shutting us down without a good run.

We have already overtaken 5 other sites for traffic and 29 for questions per day - These are statistics available to all

We have a good core, we have good quality, if we keep building it, they will come.


I've been trying to share some of the our questions on reddit/twitter where appropriate. I've even managed to get 1000 uniques to Why use static RAM addresses instead of the stack? I think as more questions are asked and shared, we will draw an audience that comes for one question, but stays to ask and answer other questions about retrocomputing.


Good time to ask the question, some early effort to raise the profile of the site should help in the longer term. Looking at the stats in more detail, the quality of the site looks good, it is the activity which is a little on the low side.

At only half-way through the minimum beta period, we seem to have nearly enough members with around 1k rep (assuming they will reach 2k given another 45 days) - and this is presumably good taking into account the low-ish volume of both questions and answers.

We have 58 avid users and 879 in total - I don't see any historical data, but it does feel that this has flat lined a little - I think this needs to more than double before we can meet the graduation metrics (and realistically the other measures will be easy if this happens)

I think it does take a while for visibility of a site like this to spread, and even longer for it to 'stick' with visitors.


Adoption will take time to hit critical mass --- if that happens at all. Realize that retrocomputing is at once both a broadly scoped topic and a niche. Also, SE sites are Q&A rather than social in nature, and retrocomputing is often a social hobby.

Suppose, for instance, you're interested in a 1970's era Hypothetical500 system. Odds are you're already in contact with other Hypothetical500 users through some other forum, perhaps one that specializes in that machine. If you have a question, you know your best chance of an answer will be there, rather than here, and the format is less strict, allowing you to also catch up with people you met at last year's Vintage Computer Festival. And you won't have to browse through 200 Amiga messages to find one relevant to the Hypothetical500.

SE sites are great, and I love StackOverflow in particular, but given what is already available online I can understand the challenges faced by the Retrocomputing site.


With the benefit of hindsight, I think I can say that the site has managed more than just fine. Here’s a comparison of Area 51 statistics from around when the question was posted versus those same statistics as of now. Other than the question rate, all the metrics have significantly improved (and the high answer rate was sustained).

I would even dispute that the low question rate is troubling; I think it’s preferable to being flooded with low-quality questions, like some other Q&A sites on the network. Still, I think even the question rate threshold can be met eventually, just by maintaining steady growth. (Unless we try some alternative approaches.)


You must log in to answer this question.

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