Whilst answering questions, is it Ok to use functions/bits of code where the output will be different for some computers?
For example I was thinking about adding a new time type to PYKE where the input could be in the form natively used by the computer (see here)
%c - Locale’s appropriate date and time representation.
Tue Aug 16 21:30:00 1988 (en_US); Di 16 Aug 21:30:00 1988 (de_DE)
This means that the code would be valid on some running computers but not others.
A more extreme variation of this could be a variant of the as of now standard loophole, Using MetaGolfScript but using something like the MAC address as the identifier. This could potentially allow any program (pyth for example) under 7 bytes long to be a zero byte answer in MAC-lang, which executes pyth code with your MAC address as input.
Should this type of input be allowed by default?
(Note should not be an answer for Default for Code Golf: Input/Output methods because this form of input isn't a stream/form of input)
-
\$\begingroup\$ I would say "Yes," given the answers here -- meta.codegolf.stackexchange.com/q/5851/42963 \$\endgroup\$AdmBorkBork– AdmBorkBork2016年05月02日 14:58:18 +00:00Commented May 2, 2016 at 14:58
-
\$\begingroup\$ Very much related \$\endgroup\$Peter Taylor– Peter Taylor2016年05月02日 16:10:39 +00:00Commented May 2, 2016 at 16:10
1 Answer 1
The question splits into two parts. The first part asks about locales: this issue has already been discussed in an earlier question, and there seems to be a qualified majority consensus that it's ok for answers to be locale-dependent as long as that is clearly documented.
The second part I would describe as asking about answers which take the piss. If someone did create a language which uses something like the MAC address to outsource part of their source code, I would consider that it violates the Outsourcing the real answer standard loophole, as well as the spirit of the Using MetaGolfScript standard loophole, and so I would downvote and either flag as low quality or vote to delete.