-2
\$\begingroup\$

We have never had a criteron about time making code. Do we need one? If we do, how do we calculate the time? If not, why not?

\$\endgroup\$
  • \$\begingroup\$ The answer to "We've never had X. Do we need one?" is no, with possible exceptions where X is a rule banning some boring loophole. \$\endgroup\$ – Peter Taylor May 5 '17 at 11:13
5
\$\begingroup\$

Probably Not

This is rather difficult to accurately measure in our format, and makes the Sandbox practically unusable. Any simple challenge may be done in minutes, making the winners basically whoever sees it first.

\$\endgroup\$
  • \$\begingroup\$ Hmm. Good point, I was about to apply it on a very hard question such as: 'Find the result for a chemical reaction' though. I did expect there would be some flaws on the timer. \$\endgroup\$ – Matthew Roh May 5 '17 at 7:08
  • \$\begingroup\$ Rather Difficult ones might be an exception, yet still advantage is given to those who see it first. \$\endgroup\$ – ATaco May 5 '17 at 7:09
  • \$\begingroup\$ I think you misunderstood something. It's time based on the time starting making code, not time based on answer post. \$\endgroup\$ – Matthew Roh May 5 '17 at 7:10
  • 1
    \$\begingroup\$ Regardless, it's far too easy to lie about that. \$\endgroup\$ – ATaco May 5 '17 at 7:14
  • \$\begingroup\$ Ah, yes. I feared about that. That was why I posted. \$\endgroup\$ – Matthew Roh May 5 '17 at 7:15
4
\$\begingroup\$

We already have this criterion…

…in a possibly disguised way. The problem with "write this code as quickly as possible" is that you can't verify that people didn't start writing in advance, except by starting everyone at the same time. But if you start everyone at the same time, there's then no way to improve after the first answer is posted. The solution is to have a bunch of miniature challenges, posted at arbitrary times, and have people compete to be the first to solve each individual challenge.

And of course, I just described (cops setting the challenges, robbers solving them). General challenge structures for which being the first to write the code are interesting should likely be run in that format, as it already achieves most of what you're looking for in a way that fits much better onto PPCG.

\$\endgroup\$

You must log in to answer this question.

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