For some kolmo questions, I use the golfing algorithms created for the challenge Golfing Strings in Fourier.
Is this okay to do? Does this constitute as cheating?
Some examples:
\$\begingroup\$
\$\endgroup\$
-
\$\begingroup\$ Related \$\endgroup\$ – user45941 Sep 8 '16 at 20:55
\$\begingroup\$
\$\endgroup\$
I don't think these answers are cheating, exactly. I don't believe they break any current site rules.
That being said, I'll probably downvote them when I see them. They're just not interesting or useful to the site in my opinion. The program that generates it might be interesting (and indeed, I upvoted issacg's program when he posted it), but I don't think that extends to the output itself.
Of course, this applies to the straight "copy/paste" answers. If all you did was copy the desired output into a script and paste the output as an answer, I don't want to reward that with rep. If you're manually golfing it before or after running the script and making tweaks to optimize it, that's a different story. If that's the case, say that in the post.
-
2
-
\$\begingroup\$ Maybe this is a case for community wiki? (Please tell me why this is wrong) \$\endgroup\$ – Blue Sep 8 '16 at 19:29
-
\$\begingroup\$ I'm not even sure I'd be interested in it as CW. It's just a copy/paste job that I could do myself. Anything interesting is going to be happening within the program, not in the output. \$\endgroup\$ – Geobits Sep 8 '16 at 19:30
-
1
-
\$\begingroup\$
\$\endgroup\$
If you can do that and get competitive solutions, it's a sign that the question should be closed as a duplicate of a standard kolmogorov-complexity question (probably the rickroll) because it isn't adding anything new to the site.
answered Sep 8 '16 at 13:36
-
\$\begingroup\$ Alternatively as a dupe of Hello, World since for that one automated metagolfers are most likely to give optimal results (whereas the rick roll actually contains some interesting structure that can be exploited). \$\endgroup\$ – Martin Ender Sep 8 '16 at 13:42
-
3\$\begingroup\$ Although of course, just because Fourier doesn't have any viable approach apart from using its automated metagolf results that doesn't necessarily mean that it won't be an interesting challenge with novel approaches in other languages. \$\endgroup\$ – Martin Ender Sep 8 '16 at 13:43
-
\$\begingroup\$ I was assuming that automated metagolfers would be able to exploit interesting structure - perhaps tunnel vision resulting from having written one which does. \$\endgroup\$ – Peter Taylor Sep 8 '16 at 13:50
-
\$\begingroup\$ I never got one which did that for the Starry metagolf challenge for example. \$\endgroup\$ – Martin Ender Sep 8 '16 at 13:56
-
\$\begingroup\$ Not always. I used this brain-flak metagolfer for my Hello, World! answer, and it still took a lot of work to golf it down. \$\endgroup\$ – James Sep 8 '16 at 16:34
