We currently have 29 questions tagged with faq. The faq tag is meant to be a category of up-to-date community consensuses on various aspects of the site, but having 29 of them indicates that perhaps some of them aren't so "frequently asked". Additionally, many of these are quite outdated - the oldest dates back to 2011, and while it's still somewhat correct it's been entirely superceded by a more up-to-date policy.
I'll list a few here that either are outdated, have been superceded by newer policy, or have other issues that simply make them not a good fit for the FAQ list:
- General rules for custom languages and libraries summarises various posts about rules around custom languages as of 2016, almost all of which have since been superceded. As I understand it, the general consensus is that, similar to command-line flags, languages using external libraries are considered separate to the original language, and almost any library/language is allowed as long as it's implemented with the sole exception of creating a language specifically for the purpose of the challenge (a loophole first created in 2014 and, as far as I know, never enforced since)
- How long should I wait before accepting an answer? is almost entirely irrelevant now that answers are almost never accepted. Note that when this post was promoted to FAQ three years ago, it was considered dubious.
- What details should always be given for a code golf task? and How does this site work? are almost entirely contained within the more recent Welcome to Code Golf and Coding Challenges Stack Exchange!
- Template For Challenges is a template for writing a challenge in one specific way. All the information it provides about challenge writing is present in the asking questions section of Welcome To CGCC, and the general structure it describes can be gleaned from looking at almost any code-golf challenge on the site, which has the advantage of demonstrating other subtleties in the way people write challenges.
- What makes winning criteria "objective"? is an exact duplicate of the more recent What are the winning criteria I can use for my challenge?. Furthermore, all the context of that question has now been resolved - we have a list of allowed winning criterion tags, and people can ask on meta if they think a new one is necessary.
- Let's decide what kind of non-challenge questions we want once and for all and its predecessor Where should non-challenge questions go? are discussions from 2014 about what non-challenges are allowed on this site. They're both quite outdated - for example, we have a well-received question about golfing language design, and a more recent consensus around how tips questions don't need to have objective winning criteria.
- Why do we have objective winning criteria? explains the philosophy behind the site having objective winning criteria. However, We've long since established that objective winning criteria are (almost always) required on this site, and while the top answer is a great explanation of how objective winning criteria drive competition and creativity, none of this is particularly relevant to the functionality of the site. It might be worth linking to this in the winning criterion list though.
- Should submissions be allowed to exit with an error? is a policy from 2017 about how to handle errors. While it's mostly accurate to the present day, we have a more recent policy on functions throwing errors, and with that change the rules around throwing exceptions can be simplified to "as long as it still creates valid output, it's allowed".
- Useful External Resources is, despite the name, not particularly useful, currently containing three bytecounters, the no-longer-functional TNBDE, a no-longer-functional graduation userscript, and a somewhat-useful SBCS keyboard tool. While it could become a list of useful tools, as it stands it is quite outdated. This was also added to the FAQ three years ago, and at the time caird mentioned that it could do with some updating.
- New users' guides to golfing rules in specific languages is ... complicated. While it is a somewhat useful resource for golfing in specific languages, the vast majority of the answers' content is covered more generally by standard I/O rules and Interpretation of Truthy/Falsey (that one has its own problems which I'll get to shortly). Additionally, this question currently has answers for 15 languages - updating it to have rules for the several hundred languages commonly used on this site is simply infeasible.
- Default for Code Golf: Program, Function or Snippet? is... If the entirety of the answer to a FAQ is one sentence, I don't think it needs to be its own FAQ post. This is also mostly covered by standard I/O rules.
- How can I incorporate good-looking mathematical exposition into my question/answer? is a post from way back when LaTeX had just been enabled for the site. Although it provides a decent explanation of how to use LaTeX, it's entirely covered by the asking questions section of Welcome to CGCC and the /editing-help page.
- Things to consider when creating a challenge is somewhat useful, but most of it is no longer relevant nor necessary for today's challenges. In the accepted answer, out of 18 points, points 2, 3, 4, 5, 6, 7, 9, 10, 13, 16, 17 and 18 have been rendered irrelevant by (mostly standard I/O) consensus.
- Interpretation of Truthy/Falsey is our consensus on the interpretation of "truthy" and "falsy" values. However, the accepted definition of "truthy" and "falsy" ("can you put it in an if statement") is quite dated and doesn't really work a lot of languages that don't have a generic "conditional" concept. (For example, most dialects of assembly have
jump if 0andjump if nonzerocommands, but not a generic "conditional"). Furthermore, it's been largely superceded by Default policy for output in decision problems, which also allows using two consistent values as "truthy"/"falsy" for output in decision-problems, which make up almost all cases where this actually is used. How to rectify this is outside the scope of this challenge, but take this as a reminder that even the oldest consensuses have issues.
This isn't necessarily a comprehensive list, but it's most of the ones that I've taken issue with. Feel free to edit in more.
While we're here, there are currently three questions marked with the faq-proposed tag, the staging ground for potential faq posts:
- What's the policy on interpreter bugs? is closed as a duplicate of Let's allow newer languages/versions for older challenges. On the other hand, it is the canonical source of "the implementation defines the spec", but as before one-sentence answers probably shouldn't be their own FAQs.
- Let's think of a creative name for our chatroom is from 2014, when we decided on the name "The Nineteenth Byte" for the main site chatroom. I assume it's proposed as an explanation for where the name came from, but I don't think it's significant enough to warrant being FAQ.
- Where to find an online testing environment for specific programming languages?, a compilation of various online interpreters, is actually a fairly useful resource and maybe does deserve faq.
These will likely need to be handled on a case-by-case basis - some being removed from the FAQ list altogether, some being edited, some being replaced and/or merged into other posts. Please answer with what you think should happen to specific posts.
