Retrogaming seems to be acceptable as a topic for retrocomputing, and this seems to admit arcade games that involve some form of computational element.

But we're also seeing questions about pinball machines and coins-on-a-string, the latter giving rise to answers involving washing machines and pay phones. Neither of those have anything to do with retrocomputing.

Is this the way we want the forum to go?

Updated, 2021-01-30

Now there's a question about colour of the lights on a pinball machine. The retrocomputing content is zero.

  • 1
  • As for payphones, I think questions about phreaking should be considered on-topic, both under the ‘history of computing’ umbrella, and also because the computing aspects can be quite relevant in such questions – both of the phone itself and of the telephone network to which it’s connected. Dec 27, 2020 at 12:47
  • 3
    Agree on phone networks, but not on how to defeat the mechanical coin-handlers in phone boxes. Dec 27, 2020 at 14:01
  • Many of these examples seem to drift into the realm of electromechanical automation (e.g. youtube.com/watch?v=NmGaXEmfTIo). I don't think that in itself is a cause for rejection, though, as we obviously welcome questions on the (electro)mechanical computers of Babbage and Zuse.
    – Kaz
    Jan 5, 2021 at 12:06

1 Answer 1


If we’re to ban such questions, it would be good to formulate an actual criterion, other than the Potter Stewart test, that distinguishes them from allowed questions.

For example, I don’t think would be fair to ban questions about certain devices merely because they are based on mechanical principles rather than electronic logic: this would cover history questions about the Analytical Engine, or electromechanical machines like the Enigma and IBM’s early punch card systems, which I believe should be accepted.

So I would like to propose two rule-of-thumb criteria:

  • If the question would make just as much sense if the device were made of Magic Space Mice, the question is probably not about computing;
  • If it’s just a ‘small’ finite-state machine, so the only ‘algorithms’ it’s capable of running are O(1), it’s probably not a computing device.

If a question fails both criteria, I think it should be considered off-topic. I’m not sure if those are sharp enough to capture what ‘computing’ means, so feel free to poke holes in them. The former is meant to capture the principle that questions should focus on ‘the computing aspects’, while the latter is meant to capture the intuition that the device is ‘morally’ Turing-complete (even if technically all computers ever built are finite-state machines anyway).

Given these two tests, I think the pinball question fails both, while the coin-on-a-string question fails the complexity test, though not the Magic Space Mice test. Then again, the latter question need not necessarily quality as ‘retro’; vending machines that accept coins are still in use, so mechanical coin receivers can be considered ‘current’ technology.

  • I think your rules of thumb are fine, though I don't think the Potter-Stewart test should necessarily be thrown out the window; I use some variant of it all the time at work ("I don't know what bad code is, but I know it when I see it"). Dec 28, 2020 at 0:18
  • 1
    @another-dave the problem with the "I know it when I see it" attitude is that different people will have different ideas of arbitration. The point of this meta is to discuss those boundaries to get some kind of consistency.
    – nabulator
    Jan 13, 2021 at 3:38
  • 1
    I think human interpretation beats algorithmic exactness in such matters. Jan 13, 2021 at 3:57
  • It’s not particularly exact anyway. But even if the line is blurry, it’s advantageous to have one, so at least there will be some questions that are undoubtedly on one side of it or the other. Feb 3, 2021 at 14:23

You must log in to answer this question.

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