It has been mentioned in chat that we could, maybe, re-consider including some of brainstorming / discussion questions to the scope, as they are richer than font-identification or troubleshooting ones.

This recent question is a new chance to talk about it (well, unless it gets closed before we discuss it).

As per our FAQ, "Brainstorming or idea gathering" questions are not allowed. This has been discussed in 2011 and agreed upon again in 2012 (please read the answers for the second link).

How do we feel about "How do I represent..." questions? Should we allow them, as long as they follow a certain format and are not just about idea gathering? "What do you think of this design?” would be, for example, a No. "How can I visually represent connectivity?" would be a Maybe.

2 Answers 2


I see "how do I represent" questions as beneficial and more in line with "best practice" type of questions. But it is a very fine line between that and "do this for me."

Conceptual problems are design problems. Not everyone perceives a conceptual question as anything other than brainstorming. So, you get flag. GD.SE is a bit different than other SE sites in this manner. Design is not black and white, right or wrong, when it comes to conceptual issues. I don't think every question regarding a conceptual problem is brainstorming. However, each question has to be taken independently to make that call. There's no blanket guideline which would fit every question in my opinion.

I think, and this is just my perception, it depends greatly on how localized the question may be. If the question is asking about how to represent something which everyone will have to represent at some point or another, there is benefit in the answer. If the question is an off-the-cuff "how should I" then there's no benefit to others.

For example, these closed questions: How can I symbolize credit? and New generation of Add New icon that has no "+" (Plus) sign are merely trying to solve a conceptual problem. Yes it's brainstorming in a way. But it's brainstorming that will help many, many users.

Then again, these open questions: Alternative symbols to represent Valentine's Day? and How to diagrammatically represent High Value Added are clearly just asking others to do work regarding a specific, localized, project.

In short... I don't think there can be a rigid set of guidelines when it comes to the conceptual questions. Let the community decide what should be closed or not.


Scott's definitely right that there's a very fine line between very good and very bad. With the new 'On hold' thing we should be better placed to push potentially good questions the right side of the line.

Trying to turn this into something practical on where to draw the line, I think there's enough in the existing (new) close reasons:

  1. "unclear what you’re asking — Please clarify your specific problem..."
    • all questions need clear criteria, it needs to be "what would work to solve X type of design problem" with enough of an explanation of the problem that there are clear criteria.
  2. "too broad — There are either too many possible answers..."
    • it needs to be "what would work to solve X type of design problem", where it's enough of a real problem with difficult criteria that there are a limited number of things that would work (not something where someone could fill a sketchbook with plausible ideas then pick the one that best suits the client - that's creative work to just get on with, not a creative problem).
  3. "primarily opinion-based — ...almost entirely based on opinions, rather than facts, references, or specific expertise"
    • it needs to be "what would work to solve X type of design problem" not "what would be nice within X criteria". The deciding factor has to be something in the problem that an outside expert can judge - else the deciding factor will be what best suits the client, which no-one but the asker can know.
  4. " a request for free work... specific to one project".
    • It needs to be "what would work to solve X type of design problem" - not "gimme something for this one brief for this one client you've never met". (but avoid the trap many people on many SE sites fall into of having an attitude like "I've never needed to do that, therefore, only you will ever need to do that")

Using these criteria on the questions Scott linked to:

  • "New generation of Add New icon that has no "+" (Plus) sign". In its current form, it passes 2, 3 and 4, and is on the line for 1 - lots of people are asking "Why would you not be able to use a + sign?" and no answer is given. If they'd answered with something like, "Because it's for an app/site/report where '+' already has a fixed meaning (like 'positively charged' or 'mathematical addition') and using it to also mean 'new' would create confusion", we've got a perfect "What would work to solve X type of design problem?" question.

  • How can I symbolize credit? - passes all of them, could be improved on 1 (context always helps) but it's clear that credit could need to be symbolised in all sorts of contexts and it's a genuine challenge to genuinely represent something abstract like this.

  • Alternative symbols to represent Valentine's Day? - needs work on 1 as what works and what doesn't depends on why it needs to be symbolised. Whether or not it passes 2, 3, 4 will depend on that. For example, it could be for shop signs, calendars, etc - an equivalent of a christmas tree to show the 'Christmas' section - which would be fine. Or it could be some very specific project, or "I wanna make a valentine's card and I can't decide what would be cool as a central image", which wouldn't.
  • How to diagrammatically represent High Value Added - this one is currently falling down on 1 and 4 - there's a vague reference to "...because that wouldn't fit very well in the particular presentation format", which a) isn't enough to go on and b) sounds too much like "gimme something my client will like". I could imagine a good question on this topic, however: something like "Lots of industrial product categories are easy to represent with icons for use in things like reports, dashboards, B2B advertising, etc, but what about 'High Value Added' [explanation]? It's a challenge to represent visually in something like an icon because the products could look like anything: the defining feature is something abstract and economic"

It would be helpful to have some examples of questions we definitely don't want to test this fully.

Something to think about for these type of questions - thenounproject.com. It's:

  • full of iconographic representations of things, many of which are the kind of uncommon difficult-to-visualise things these questions would cover
  • a good first place to check for existing answers; an answer could be "X Noun Project user has something that works for this [image] [link]. It works because [...]."
  • if you find it easier to sketch out something than describe it, you could if you wanted make your answer even more useful by sticking it on the Noun Project as a usable, downloadable vector icon (and you could even sell it through that site it if you think it's worth it and there might be enough demand...).
  • +1 for the thenounproject.com.
    – Kromster
    Commented Dec 20, 2013 at 8:21

You must log in to answer this question.

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