I'm a bit irritated how you handle things in the game developer stack overflow.

While it's is common practice on stackoverflow.com to share your personal knowledge and way to solve it, while here you are restricting discussions and different opinions/ways to solve a problem.
Maybe this is the reason why this board is rather small?

Giving two examples:

1) You shut down this thread:
https://gamedev.stackexchange.com/questions/71550/how-can-i-do-infinite-ground-for-running-game
while the first comment in the post is clearly the answer to the OP's question. I don't understand why you are closing his thread and not posting the first comment as answer instead.

2) Asking for best practices or options for basic workflow
https://gamedev.stackexchange.com/questions/71554/basic-workflows-for-game-development-udk-cryengine-unity-export-format?noredirect=1#comment129376_71554

-) I'm asking for rethinking about which questions are really about opinion on a unique author's subject (e.g. if one likes a color) and which questions just may have multiple ways to solve it.

-) I also don't think that there are question which have too many possibilties to solve so that you need to but them on hold. Else nearly every question at stackoverflow should be put on hold because there are 1000 ways to code a solution for a problem.

I experience the stackoverflow system as followed: Someone asks a question and users answer it with their own solution. Then the community upvotes the best/simplest/shiniest one. Additionally this helps the questioner to valuate the solutions.

migrated from meta.stackoverflow.com Mar 8 '14 at 17:15

This question came from our discussion, support, and feature requests site for professional and enthusiast programmers.

  • 7
    While it's is common practice on stackoverflow.com to share your personal knowledge and way to solve it we must be visiting two different Stack Overflows then. On the one I visit, questions inviting discussions and different opinions are shut down with extreme prejudice. – Pekka 웃 Mar 8 '14 at 17:16
  • Why? Isn't this the way it always works? Someone has a problem and one posts his solution in form of code or something which is the way he would solve this issue. There are of course 100 other ways and then another user posts his way. And the users upvote the easiest way. I could really start listing nearly infinite threads now. – Coretek Mar 8 '14 at 17:18
  • Just to note, the "you" in this question is about me. – MichaelHouse Mar 8 '14 at 17:21
  • 2
    It doesn't matter in my opinion which moderator did it because I believe it was done in best interest following your game dev guidelines. So I seek adjustment of the guidelines. Or learning to interpret them how the mods @ stackoverflow do it (because I think they wrote them and gave them their intention). – Coretek Mar 8 '14 at 17:24
  • What, specifically, are you asking for? You don't want opinion based questions to be closed? The only question I see in this post is asking if I think I'm the reason we're a "small" stackexchange. – MichaelHouse Mar 8 '14 at 17:25
  • No this is not an insult to you it's a maybe challenging question but with the aspect in mind of improving the game dev section. Nearly every answer to every question will have "opinion" or "one's way" in it anyway. Providing this way helps the questioner. – Coretek Mar 8 '14 at 17:27
  • OK... "What, specifically, are you asking for? You don't want opinion based questions to be closed?" Please edit the question to state what you're wanting. – MichaelHouse Mar 8 '14 at 17:29
  • Maybe it would be too much if questions like "Do you like this model to be more red or more yellow" are allowed. But I don't see in any way how these two question mentioned in my first post (especially the one about infinite levels) are purely opinion based posts. – Coretek Mar 8 '14 at 17:31
  • If a question has a provably correct solution then of course the opinion aspects are fine, but these suggestions would harm an already very damaged place. Those type of questions are already a problem on the site, if a question is specifically calling for opinion well it's time for you to do your own research and form your own opinion. A flame war of opinionated programmers all juiced up on theory helps nobody. – MickLH Mar 8 '14 at 17:32
  • The other question you're pointing out was closed as "too broad", not primarily opinion based. – MichaelHouse Mar 8 '14 at 17:32
  • --Edited (Misread): I see now that "too broad". But thats exactly the point. I extended my question @ OP. At stackoverflow there are nearly no questions which are "too broad" because there are 1000 ways to solve them programatically. At least that is my experience and I liked it that way. (I asked a lot of "broad" questions there) – Coretek Mar 8 '14 at 17:37
  • 3
    I'm not active here so I can't offer an opinion on what Gamedev.SE should or should not do, but those questions would have been closed on Stack Overflow, too, for the same reasons they were closed here. – Pekka 웃 Mar 8 '14 at 17:56
  • 2
    Very difficult to compare questions, saying this one is still open, why isn't this other one. You should instead focus on how you can improve your question. – MichaelHouse Mar 8 '14 at 18:01
  • 2
    Funny. It feels to me like we're talking about two different Stack Overflows. The one I know closes questions like those you show with extreme prejudice - at least in the tags I tend to frequent. – Pekka 웃 Mar 8 '14 at 18:03
  • 2
    How recent is your experience on Stack Overflow? The way the SO community handles broad and discussion-y questions has become radically strict over the past 2-3 years, so far that many think it's too strict – Pekka 웃 Mar 8 '14 at 18:11

The two questions you mention are indeed both poorly suited for the Stack Exchange platform. The problems they have are rather different, though, and some of them may be fixable.


The first question was closed as "too broad", but IMO, a more accurate closing reason would've been "unclear what you're asking". Basically, condensing the question and its answers and comments into a few lines, they essentially went like this:

Q: "I want to make a running game with infinite ground. How could I do that?"

A: "One way is to split the world into several parts, and keep creating new parts in front of the player and removing old ones behind them."

C: "I tried that, but the player falls through the floor! Why does that happen?"

Basically, it's a kind of a chameleon question — it started out as a conceptual question about maintaining the illusion of an infinite world, and then morphed halfway through into a "there's a bug in my code" question (with no actual code included, which makes finding the bug extra difficult).

Questions like these cannot really be meaningfully answered, since the actual question being asked is a broad, nebulous and moving target. To try to give a useful answer, one would have to guess which level of abstraction, from general game mechanics to specific coding syntax, the OP is actually having problems with — and it may well turn out that the answer is "all of them", in which case the question is indeed too broad for Stack Exchange.

The general way to fix such questions, where possible, is to engage in comment dialogue with the asker to try and clarify what they already know and what they're specifically having trouble with, and then edit the question to focus on that one, specific, answerable issue.

In some cases, it may also be possible to split such questions into multiple parts, one for each of the component problems. For example, this example question is broad enough that it could be split into (at least) three distinct questions with distinct levels of abstraction:

  • "How can I create the illusion of infinite ground in a running game?" (abstract, platform-independent),

  • "How to implement infinite ground in AndEngine?" (concrete, platform-specific), and

  • "Why does the player fall through my infinite terrain?" (very specific; should include a minimal example for testing).


As for your second example question, the main problem is simply that it's phrased as asking for the "best" way to do something, even though, underneath, it's really a "how to" question.

To see the difference, compare these two hypothetical cooking questions:

Bad question: "What's the best way to cook an egg?"

This will invite a countless list of subjective answers, with one answered preferring their eggs hard-boiled, while another one likes theirs fried sunny-side up. (Cue lengthy comment debate on the relative merits of sunny-side up vs. over easy.)

Good question: "How do I cook an egg? I tried putting it in the microwave, but it exploded!"

or:

Good question: "How do I cook an egg? I know I can just boil it for ten minutes, but I never get a nice runny yolk that way."

These questions clearly specify what the asker wants, and how far they've got on their own (i.e. in development terms, which level of abstraction they're interested in / having problems with). While the topic is still very broad, it's clear that the specific question can be adequately answered in a reasonable space — at a bare minimum, "immerse the eggs in boiling water for three minutes" is technically an adequate answer.

The one thing these good questions don't do is ask for the single "best" way to do something. Somewhat paradoxically, by not doing so, they actually encourage comprehensive answers that describe several good solutions and compare their respective advantages — like, say:

"There are many ways to cook eggs, but, as you've noticed, microwaving is generally not one of them (although it can be done). One of the simplest ways is to immerse the eggs in boiling water for a short time; depending on the immersion time, this can produce either a fully congealed ("hard-boiled") or a partially runny egg. [...] Eggs can also be fried on a hot pan. Again, there are several ways to do this: [...] Other methods of preparing eggs include shirring, poaching, scrambling, etc. [...] Eggs can also be used as an ingredient in many dishes, such as..."


Addendum: Prompted by your comments, I tried to see if I could edit your question into something that I would personally find useful and answerable. Here's what I ended up with.

However, while doing so, I ended up repeatedly running against the fundamental issue that your question doesn't seem to include "a practical, answerable question based on an actual problem that you face". Specifically, based on the impression I got while editing your question, I strongly suspect that you asked it before even trying to just make even the simplest 3D model (say, a yellow sphere) in any modeling program (say, Blender) and importing it into any game engine (say, Unity).

At this point, my main advice to you regarding this question is:

  • If you have tried that, please describe what you did (briefly — no need to go into details) in your question, so that potential answerers have some idea of what you're familiar with and what needs to be explained.

  • If you haven't tried that, go do it now! Then come back and, if you still feel that there must be a better way, clarify your question based on what you found easy and what you found difficult.

  • I totally agree with you on the first part. This could all be explained in the comments, and the user could post a follow up question which is surely something others are also interested in. Or one could post it like YOU DID there. Answering the question and hinting to potential problems. Hell this is what I see on stackoverflow all the time. Look at BalusC's post. Maybe he is one of the best posters out there but this is how it should be done imo. He is not afraid to share his knowledge because he seems to have infinite deep knowledge. No jealousy from his side. – Coretek Mar 8 '14 at 22:44
  • And therefore I don't see a reason to close the thread about infinite levels. Maybe edit/format his question by a mod or himself but not locking it. I found the topic interesting too, reading the various ways how it could be done. – Coretek Mar 8 '14 at 22:45
  • To your second part I completly disagree. Because I already have a way in mind shows that I did research but not sure if it's a correct way. And I'm asking it because I didn't find my answer elsewhere. Yes there may be different best workflows for everyone put what is wrong in telling your workflow and show the PROS and CONS about it. You can learn from other answers too and the community will upvote the best anyway. Again this is something wich happens on stackoverflow all the time. People post their coding solutions or hints and the easiest/best one is upvoted. – Coretek Mar 8 '14 at 22:48
  • What is wrong with comprehensive answers? What is wrong with having content on your site? I really don't get it why there shouldn't be different opinions with their pros and cons. And about comprehensive answers: On Stackoverflow some hot questions are 1-liners with code asking how this code produces a specific outcome and other people wrote half site long in depth explainations. How is this bad. – Coretek Mar 8 '14 at 22:51
  • Additionally I altered my question from best practice to: what other options are there. I bet such questions are more interesting for the most of the internet population than questions like "how do I get the second left pixel in uberhax0rdes1ng0r-tool to be half black?" – Coretek Mar 8 '14 at 22:53
  • @Coretek: Comprehensive answers are good. We want comprehensive answers. Unfortunately, they're also exactly the kind of answers that you generally won't get if you ask for the "best" way to do something. So, you know, don't do that. Instead, the trick is to ask for any way to do what you want to do (and, preferably, include an example of a way that you tried and found less than optimal, so that answerers have some context for proposing alternatives). You'll find that, this way, people will still naturally suggest the solutions they like best, but they'll include other solutions too. – Ilmari Karonen Mar 8 '14 at 23:10
  • Excuse me, this is exactly what I did now in my thread? I asked if my way is feasible and what other ways are possible. On a top level, so the answers wouldn't be too long too. – Coretek Mar 8 '14 at 23:12
  • 1
    And you've got one reopen vote since you edited it already. (IMO, you could still improve your question by shortening it and narrowing its focus -- currently, it's kind of hard to tell at a glance what you're actually asking. In particular, your "Q1" and "Q2" parts should really be separate questions. Oh, and fix your question title; "Basic Workflows for Game Development" says nothing about your actual question; "How to make a model 'game ready'", from your lead paragraph, would be a much better title.) – Ilmari Karonen Mar 8 '14 at 23:19
  • Ok I get it that I can improve it in detail but your suggestions seem to make mountains out of molehills. – Coretek Mar 8 '14 at 23:22
  • 1
    I'm making molehills out of molehills. You want your question answered, right? If you do, it's a good idea to phrase it so that others who see it won't just go "too messy; didn't read". I can't claim to speak for everybody, but I'll typically look at a question for maybe three seconds before deciding either "this looks interesting, let me read on" or "nah, not interesting / can't make sense of this, I'll pass". Currently, your question still fails that test for me; it took me way more than three or even thirty seconds to even figure out what it's really asking about. – Ilmari Karonen Mar 8 '14 at 23:31
  • Thats absolutly fine and I don't have a problem with that. What I argue against is putting a question on hold and therefore restricting others from answering too, even if they want. I did have some thoughts of how to phrase my question - the question as you read it now was my first attempt, I put in the best practice afterwards (still before posting) because I thought this would lead to less possible ways but more ways how people are engaging it and therefor get different perspectives. – Coretek Mar 8 '14 at 23:34

Gamedev.SE has a tough time being a good SE site. It's true. However, your ideas (they weren't very clear) for solving them don't address those problems. In my opinion, your ideas for solving them represent lingering problems on SO.

When a person posts on SO "How do I do X in javascript", and they receive a slew of answers complete with 4-part jsfiddles, that's...dubious. Yes, they got help, and yes, the multiple solutions can be helpful to people. But it sets an unfortunate tone for the site. People expect to be able to do no research on their own, simply post their problem, and receive a complete tutorial instantly. That isn't really the spirit of the SO; it's something of an emergent behavior.

Maybe that's ok for javascript. There's usually a quick solution that is based on utilizing some feature of the DOM or HTML5 or CSS, and the questioner simply didn't know what they didn't know. They couldn't really google a term to research, and they didn't want to read the full spec, so they expressed it the best they could, as a raw how-to question.

But Game development questions (and frequently, a lot of SO questions as well) cannot be answered like that. You can't post "how do I make an MMO" and get an answer. There is no right answer.

There is no equivalent to jsfiddle for game development. When people post those questions here, they expect someone to tell them the right answer so that they can avoid studying and building their knowledge the hard way. That is unacceptable.

GameDev.SE can answer (some) game design questions. It can sometimes list graphical techniques. It can flesh out algorithm implementations. It cannot manufacture tutorials, nor can it teach a person to develop games.

  • I understand your point but there is also no right answer to how to fix a specific javascript problem as there are many solutions for one problem also. If the answers are too broad ppl ask in the comments to specify. And also on the other side: if no one wants or has an answer for a question (which never happend in my sections of SO) then there is also not any harm done in adding a comment that the question is hard to answer and leaving the question open to submerge. – Coretek Mar 8 '14 at 18:21
  • Another point is, that I find far less broad documentation on game design than I find broad information (like differences and so on) for different coding languages. So I would especially here appreciate the opportunity to ask the questions I have here (and googled before - generally speaking: if there are solutions in links, provide the links, simple as that, or mark threads as duplicates). – Coretek Mar 8 '14 at 18:23
  • QUOTE: "There is no equivalent to jsfiddle for game development. When people post those questions here, they expect someone to tell them the right answer so that they can avoid studying and building their knowledge the hard way. That is unacceptable." And this is how I get the feeling. You don't want to help users here because you are afraid they get where you are in less time and devaluate YOU. Why not close this site at all? This is elitism. + it slows community building down - at least. The one that you help now, may be the one to provide an answer for you in the future. – Coretek Mar 8 '14 at 18:27
  • The problem is not that we don't want to help them. The problem is a lack of effort to help themselves. If the question defines what they have tried, what research they have done, and a narrow scope of problem, then it is answerable. If it's asking for a tutorial, it is not. If it's a code dump with no explanation, that is also a lack of effort. A good SE question should define the problem AND what attempts have already been made to solve it. That's what's missing from a too-broad question. And asking "best method" is similar; the poster should try it and see what happens. – Seth Battin Mar 8 '14 at 18:34
  • + One last thought: If you don't want to answer a question why not leave it open? Maybe other ppl want to help. – Coretek Mar 8 '14 at 18:34
  • 5
    We don't leave them open because we don't want crappy questions on the site. We DO want to build those search results, so that the greater internet can google things in the future. But lowering our standards won't accomplish that. – Seth Battin Mar 8 '14 at 18:41
  • Funny that I got "those search results" with one broad question leading to opinion based answers and still people thank me for this question. stackoverflow.com/questions/14136731/… But I see where your comment is going, just like your original post. You don't seem to be that creative to let your work speak alone - you prefer to guard the basics to hinder other peoples creativity. Then again I bet you didn't learn all by reading docs for yourself but had a teacher or ppl to answer your questions. This is sort of greed: demand without contribution. – Coretek Mar 8 '14 at 22:32
  • + to your person: meta.gamedev.stackexchange.com/questions/1545/… Why are you still on this stackexchange? You clearly seem not one person to help or care about others. Are you a mod here? That would truly be a shame. You should really carefully listen to Josh Petrie's answer. – Coretek Mar 8 '14 at 23:25

The goal of the site is basically to help all users get from point A to point B.

point A is where a certain specific (in this case) game development issue they've been working hard on isn't solved. They then may read a book about it or search the web and perhaps still feel they could benefit from someones advice. That is because they are stuck regardless of how much effort they are putting into it, it's not happening.

point B is where they have all the tools and information to resolve it properly. This could happen because they searched and found the answer on the site or they asked a new question that may help others in the future.

Getting people from point A to point B is what makes this site great and what makes it a success in my opinion.

Users in point A describe why they're somewhat stuck and what have they tried so far:

Connecting Circular Rooms

There are two distinct problem (with the questions) that you're possibly ignoring:

  1. Opinion based questions are fine when there is none-subjective meter to test them by. For instance questions that ask how to get something specific done (in code perhaps) are fine because you can run the answer in an interpreter (or compile it) and see if it works. Other questions where the opinions can be strongly based up by fact, work too. The facts are the meter where the compiler was the meter in the former. This is why we don't call them opinion based questions. On the other hand there are questions based where answers will nearly entirely be based on taste, personality and other factors, such as which computer should I program my games on? Which language is better for games? Which engine should I use? This are not helpful because the user is not in Point A, she is not stuck with a problem she can't resolve. She often did not even get actually started. The magical path to greatness is not something we oft can provide in a two paragraph answer to a person we haven't met in a way that any reader would gain something useful by reading it also. Therefore it does fit in a Q&A. It fits in a laid back discussion forum.
  2. Questions about a specific game mechanic or how to make something are also fine as long as that something can be reasonably explained in an answer that fits into a screen (pictures don't count). If someone asks question that require an entire book to answer or a few years of rigorous learning and training regimen than we can't take it seriously. It needs to be either broken down to some smaller scope questions or thrown in the recycle been. Any answer attempting to treat the question seriously would be very shallow and actually not take the users from point A to point B. Mainly because the users are not truly in point A, the possibly haven't gotten started yet or they are not asking the right question.

You must log in to answer this question.

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