8
\$\begingroup\$

Language tags are really popular on GD.SE (3 out of the top 10 tags: c++, c#, java) and yet I've found them almost always useless. Just look at the questions list on the front page and you'll find plenty of examples. Most questions tagged with a language tend to be one of the following:

  • The problem isn't language specific (e.g. "How do I implement A*?" )
  • The problem is more specific than the language (e.g. "Alpha blending in pygame?" )

There are a few reasons why people use these tags; perhaps they prefer answers to be presented using that language. In very rare occasions there are gamedev questions which are language specific, for example a language feature that solves a gamedev problem in an interesting way, but I've seen maybe one or two questions like this. In any case I don't feel the popularity of these tags are justified.

Should we discourage the use of these tags, or start removing these tags from questions?

\$\endgroup\$
  • 3
    \$\begingroup\$ They are the easiest way to trigger syntax highlighting, much more accessible than manual language hinting. \$\endgroup\$ – Kelly Thomas Oct 1 '13 at 10:33
14
\$\begingroup\$

Tags are for searching and filtering.

While there is some validity in the assertion that many questions are fundamentally algorithmic in nature and thus language-agnostic, in practice many users will want help tweaking a particular implementation in their preferred language.

Similarly, many answerers will prefer to demonstrate algorithms through example code, or pseudo-code, in the style of a preferred (or topical) language.

Consequently, there is utility for both askers and answers in being able to filter or search posts using language tags, and I don't think it would be adding any value or improving clarity to discourage their use or actively remove them from the site.

Redundancy in tags (such as and ) only causes problems when you're trying to tag a question and hit the tag limit. At that point, presumably the user can decide which tag is more important and drop the other one. I haven't seen too many questions tagged to the limit or heard too many (read: any) complaints that we need more tag slots, so I don't think this is hurting anything.

\$\endgroup\$
4
\$\begingroup\$

Here on GD, the tag isn't a way to search for questions on the language itself. "What does the keyword 'mutable' mean?" Yeah, no. Go to SE. Those questions aren't here, no-one is going to be searching for them.

Instead, I see the tag as a hint for solutions -- and libraries. If someone flags a question using C#, the answer might probably come from .NET or a .NET-friendly library; they might be using XNA, and those people will come along, too. And a solution that refers to a python library would of course be irrelevant, even if it did solve the problem as asked. I'd hate to hear "if you were using python, you wouldn't have this problem," because it's irrelevant to the asker.

I think the language is part of the context. "How do I implement A* in C++?" is a very different question than "What is the A* algorithm?", although one could infer either if the question was just "How do I implement A*?".

Yeah, maybe it's often spurious, but I don't think it's detrimental to allow it.

\$\endgroup\$

You must log in to answer this question.

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