For example, could I invent a language that contained only two characters (for example 1, and 0) and use it to answer a question (assuming it fulfilled all the necessary site criteria), and then score it as 1 byte per 8 characters? Or would this still require 1 byte per character?

  • \$\begingroup\$ Well, you could trivially create a derived language which has 256 characters, each being blocks of eight of the original ones. \$\endgroup\$ – Adám Jun 7 '17 at 0:01
  • \$\begingroup\$ It would be a very limited language but yes. It's the same as someone submitting an answer as binary but using 1 bit rather than 1 byte per command \$\endgroup\$ – caird coinheringaahing Jun 7 '17 at 0:01
  • \$\begingroup\$ @Adám yes, the only advantage of such a language would be the fact that one could also use shorter commands. \$\endgroup\$ – Gryphon Jun 7 '17 at 0:02
  • \$\begingroup\$ @Adám, I think there is a slight difference, given that I'm actually asking whether this is viable on normal code-golf questions, and that question asks whether the user could posit this while asking a question, but feel free to VTC on this as a dupe. \$\endgroup\$ – Gryphon Jun 7 '17 at 0:05

If there's an interpreter that supports this, yes

On PPCG a language is defined by its interpreter. If there's already an interpreter which can read the condensed format (8 characters stored in one byte) and properly execute the code, by all means use it. If there's not, feel free to create one and use that.

Please note that if the interpreter is newer than the challenge, it is not allowed to compete unless the challenge specifically allows it. But you can still answer challenges with that interpreter, as long as you mark the answer as "non-competing".

  • \$\begingroup\$ Thanks @ETHproductions. Darn, now I need to code an interpreter for my language before using it. I was hoping I could just define a bunch of commands and stick them on GitHub or somewhere. \$\endgroup\$ – Gryphon Jun 7 '17 at 0:22
  • \$\begingroup\$ @Gryphon An interpreter is required, see this question about a language that could never be implemented. \$\endgroup\$ – Draco18s no longer trusts SE Jun 7 '17 at 13:49
  • \$\begingroup\$ I realize that now, @Draco18s. I just wish one wasn't. \$\endgroup\$ – Gryphon Jun 7 '17 at 16:46
  • \$\begingroup\$ @Gryphon Understandable, but sadly the truth: someone needs to be able to verify the answer, afterall. \$\endgroup\$ – Draco18s no longer trusts SE Jun 7 '17 at 16:47
  • \$\begingroup\$ True, but for simple programs, a list of commands should be enough to allow for quick human verification. \$\endgroup\$ – Gryphon Jun 7 '17 at 16:49

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