We've got a class of questions related to game identification that we seem to be rather inconsistent on.

We've gotten several questions where the asker wants us to identify something, and believes the original source is from a game, but has nothing but their memory to back up that assertion. This...sorta meets our artifact criteria, and we seem to be more accepting of visual artifacts, but we're still all over the place. I find that rather problematic, especially with how we're not even clear on what we think a concrete artifact is. I got a lot more disagreement than I was expecting there, which leads me to believe we're taking a harder stance than I was assuming.

So this is an attempt to see if we might be able to clarify a small part of our game identification exception. We seem to be playing magic word syndrome with them as well; just add, "I think this came from a game, but I can't remember which", and the only way to know is to prove it. Answerability makes for an extremely poor metric, but that seems to be our current stance; we can prove or disprove it, and we keep it either way.

Our old game-id text said:

Questions asking for help identifying a game, based on a description, feature list, or any other criteria are off-topic; this blog post might help. One exception is identifying games based on an actual piece of the game, i.e. screenshots or audio clips.

We didn't even follow that, really.

Our new close reason states:

Game identification questions that rely solely on memory are off topic here. If you find a game in a video, advertisement, news article, movie and so on, and you have a picture, video/audio file, or other medium to point to, we can answer that.

These questions don't even meet this criteria; all we have is a good faith belief from the asker that thinks their source is a game, and nothing but their memory to go off of. Isn't the whole reason we don't allow identification from memory the fact that it's pretty darn fallible? How does a non-game artifact, along with memory, somehow produce an acceptable question that meets our game-id exception?

Some examples:

So what do you think? Are these questions valuable? Do they produce good answers? Or should we disallow them without concrete gaming artifacts?

  • 2
    Thanks for putting this on Meta, I think we've both been patrolling these kinds of questions for a while and we've butted heads over this in the past. Writing up an answer now, but before I put it together I'm honestly curious: what questions do we have on the site that you consider non-borderline cases of good artifacts? The way I'm reading your argument now and in the past, it feels like literally every game ID question requires good faith that the artifact isn't just a clever mockup or custom pixel art. – GGMG Jul 9 at 1:33
  • 1
    I don't see how we can create any sort of criteria for "concrete gaming artifacts" without also excluding questions like What game is this character on Nintendo's eShop from?, which didn't have any controversy despite being a non-screenshot and, as the answer discovered, a reskin. – Thunderforge Jul 9 at 1:41
  • 1
    @GGMG The only ones I'd consider as, "good" game-id questions are the videos that show someone playing a game. If all you can get is a pic, that's also fine, because that was the whole point behind our exception. We've strayed far beyond what it was intended to be used for, and while they are popular, I don't believe they're very good questions in the first place. – Frank Jul 9 at 12:16
  • 1
    @Frank So in your opinion, and contrary to the game id text, we can never have a good question based around sound ID. – GGMG Jul 9 at 14:03
  • 1
    @GGMG We make a nod to it, and I do believe there can be good ones. But single sounds are not it, not by a mile. – Frank Jul 9 at 14:06

TLDR: A good question doesn't require an answer to prove that it is on topic.


A good question doesn't require an answer to prove that it is on topic. So, any game id question should state where the artifact was found, and why the author believes it to be from a game, or it is off topic by the same reasoning that questions requiring an answer from the game developer are off topic even if such as answer exists somewhere online. What @GGMG said about allowing an answer that states, "This isn't from a game." doesn't really work, because proving a negative can be impossible.

A good question is something like this. It clearly states where it was found and that it is clearly a game. Other examples of good game id questions would be screen shots or gifs of a video of someone playing a game with a link to the video as reference (please don't just link the video. Videos tend to link rot especially fast) or a game poster with characters in it asking for the identification of exactly one of the characters in the poster.

  • Is this different from what I said? – Wrigglenite Jul 11 at 14:58
  • @DanmakuGrazer, not so much different, just a different wording with some additional thoughts and some examples of what make good game id questions – Dragonrage Jul 11 at 15:07

Aside from finding the answer, we have no way of knowing whether or not an artifact is truly from a game

Consider these two t-shirts (on sale at Spreadshirtmedia.com):

Brotherhood of Nod t-shirt Mobile Suit Gundam Thunderbolt Zeon t-shirt

One of these symbols is from a video game. The other is not. Can you tell which one is which? If you don't immediately recognize the symbols, then it's impossible to distinguish.*

Similarly, one of these screenshots is from a video game cutscene. The other is from a movie. Which is which?**

Banner Saga 2 Cutscene Quest for Camelot

Unlike Science Fiction & Fantasy Stack Exchange, we can't tell from the user's question whether or not it is on-topic for the site. As you described in the question, this causes all sorts of problems (I think the Ingress one is especially problematic, since it was closed until it was discovered that it was from a video game and then reopened).

We should trust the question asker's belief that the artifact is from a game until proven otherwise

All of the above questions met the criteria of identification questions by providing "screenshots, audio, or other tangible media" (a symbol is debatable, but I think it's similar enough to a character's appearance outside of a screenshot, which was established as being on-topic).

It is my opinion that a question should only be closed if it is proven to be off-topic (I suppose you could say that the question is "innocent until proven guilty"). In other words, we should assume that the question asker is correct when they say it is from a video game. Now we could close a question for being "too broad" because they haven't provided enough information, but otherwise, we should definitely trust the question asker.

Having any sort of criteria for "concrete gaming artifacts" is a really slippery slope and one that has too much gray area to rule on. I believe the best policy is to trust the question asker until proven otherwise.

* The first symbol is for the Brotherhood of Nod faction from the video game series Command & Conquer. The second symbol is for the Zeon faction from the anime/manga Mobile Suit Gundam Thunderbolt.

** The first screenshot is from the video game The Banner Saga 2. The second screenshot is from the animated film Quest for Camelot.

  • 2
    Having to know the answer to close a question makes for incredibly poor policy. It's one of the reasons lore is so contentious, and why we close every single dev intent question, whether or not it's answerable. A question has to stand on it's own, without an answer, to know whether to close it or not. Anything else just continues our current inconsistency. – Frank Jul 9 at 11:35
  • @Frank Absolutely agree. My personal preference is actually to just ban identification questions altogether, but given that the community seems to like those, trusting the question asker seems to be the only practical solution. – Thunderforge Jul 9 at 13:31
  • 1
    We'd be on the exact same page, then. But if we have to keep them, I believe restricting them to the core exception we made in the first place sounds reasonable. See it in a movie, TV show, or something? That we can help with. Burn the rest. – Frank Jul 9 at 13:33

Aside from finding the answer, we have no way of knowing whether or not an artifact is truly from a game

And since a question's on-topicness is not influenced by its answer, the asker should reasonably prove that what they want to identify is in fact from a video game. To me, this means fan art, random symbols, any sort of artifact not found in game-related media are all off-topic.

Really, Thunderforge's answer says it all but reaches the opposite conclusion.

  • How do you distinguish between “game related media” and fan media? Would this also mean we can’t have questions identifying what game a character in a movie is playing? – Thunderforge Jul 9 at 13:28
  • A screenshot of a game shown in a movie is as game-related as it gets. What I meant by fan art is cases like this deleted question, which asked about this image. If a piece of fan art is titled "TOP 100 MANLIEST VIDEO GAME CHARACTERS", that's on-topic to me. I'm all for a case-by-case analysis of questions, the asker should present a reasonable argument as to why they believe they're looking for a game. – Wrigglenite Jul 9 at 13:48
  • So our highest upvoted game identification question would be off-topic under your criteria because it's not a screenshot from a game, nor do we have anything aside from the OP's word that it's from one? – Thunderforge Jul 9 at 17:27
  • 1
    Yes, definitely. I'm not sure why that would be an exception. – Wrigglenite Jul 9 at 17:34
  • @DanmakuGrazer, if it is asking for identification of all 100 of those video game characters, I would vote to close that in a heart beat. its way too broad and just leads to a partial answer war. if they want one of those video game characters identified, thats fine by me. – Dragonrage Jul 10 at 15:45
  • @Dragonrage Yeah, it'd be off-topic as too broad. It was just an example to show what I meant. – Wrigglenite Jul 10 at 15:59

Voting questions as off-topic because it's been decided the artifact isn't from a game is the wrong way to go about it. How about we just allow "Sorry, this isn't from a game" to be a valid answer?

There's no need to retroactively come around to and close questions as off-topic because it's been argued that they're probably not from a game. That's a ton of work, a ton of arguing, a ton of rules lawyering, and (in my opinion, at least) is an abuse of the flagging system in a QA environment, and all we get from it is a confusing mess of rule enforcements, and even more confusion when the flags turn out to be wrong and the artifact was from a game we just didn't immediately know about.

That can come off as hostile and inconsistent, especially if a user took the time to go to the help page, read what's allowed, and then follow it to the letter. You give What game is this symbol from? as an example and say it "sorta worked", but it didn't work at all. This is the worst possible scenario for these questions using the current assumed close rules. The community badgered the asker, insulted their memory, argued for three days in the comments, closed an on-topic question, forced the asker to receive an answer in the comments, and then silently re-opened it.

Everything in the Stack Exchange system broke down there, and it broke down because we're all scared to just say what we already flagged: "I don't think this is from a game because..." Then instead of being a bad moderation call, which requires a lot of energy and time to reversing, it would have just been a bad answer, which the site has tools to deal with.

Because, here's the thing: no matter how stringent we make our requirements we will never completely remove false-positives. It's not going to happen. Let's say we go with an incredibly stringent reading of the existing rules, even stronger than yours: we need video proof of the game being played, in full view, with no obstructions.

There will be Devry University ads.

There will be Simpsons episodes.

There will be unreleased game footage.

There will be beta footage from existing games.

We will still be facing the same problems and shutting down questions as off topic for following, to the letter, whatever strict rules we create. And then we'll be right back here and people will be arguing to make the requirements even stricter to remove those false positives (Hold a controller up to the screen to prove the game is programmed as accepting input?)

So let's just use the QA platform as it was meant to be used. Answer the question if you feel strongly enough to flag it, let the community vote and decide if your reasoning that the artifact isn't from a game holds up.

  • 2
    Because we are gamers. We deal with games. Not random things that people think are from games. What value are we providing by allowing things that aren't even about games to exist at Arqade? It dilutes the signal to noise ratio. – Frank Jul 9 at 16:08
  • 4
    @Frank Kind of you to say, but we're absolutely not. We're stack exchange members, and nothing more. And we are using punitive flags in a way that they were not designed to be used and harming other members in the process who have followed the rules we've provided them. I feel very confident in saying that needs to stop. – GGMG Jul 9 at 16:15
  • If you're at Arqade, you're a gamer. That's our core expertise. That's our subject matter. That's what we deal with. As for flagging, old questions are not immune to today's rules. If rules change, we close and either lock or delete them. That's just basic curation. It's not punitive, it's not personal, it's just keeping the site clean. After 60 days, there's not even a rep impact; they keep whatever they gained. – Frank Jul 9 at 16:18
  • @Frank I mean, I'm not. You can say I am, but I'm not. Haven't played anything more than Rocksmith for the past three months. I also don't use Bitcoin. I'm a Stack Exchange member first and foremost and I want the sites to function as intended. Shutting down an answerable question and forcing it to be answered through un-filtered back-channels is not working as intended. It's not even "sorta works". – GGMG Jul 9 at 16:34
  • "As intended" is up to the community, and the whole reason we have Meta. Arguing that we should allow these because we're part of SE is going to be a bit of a non-starter. And, again, answerability is a poor metric for quality. It makes no difference whether a question can be answered or not. It only matters if it's a good question. – Frank Jul 9 at 16:56
  • 1
    @Frank Votes are intended to tell the difference between good or bad questions, not flags. So I say, let's answer the question and let good/bad questions be decided in-system. – GGMG Jul 9 at 17:21
  • 1
    Here's the thing; game identification questions are popular. They'll get an inordinate amount of votes, even if they're pure crap. Because people like games. That doesn't make them good. It's why we banned the vast majority of them. Letting votes decide is poor policy. – Frank Jul 9 at 17:26
  • @Frank Could you provide an example of a good Game ID question currently on the site? I'm upvoting the questions you think are "pure crap", so I don't know what you want out of this new policy. – GGMG Jul 9 at 17:51
  • What I want is consistency, and less community tug of war between who's currently on and defending the questions. It's clear you and I are opposed on fundamental grounds, so I don't see much point in continuing this. We disagree. We'll leave it at that. – Frank Jul 9 at 17:58

You must log in to answer this question.

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