On 11 October 2012 16:52, Kerim Aydin <ke...@u.washington.edu> wrote: > > > On Thu, 11 Oct 2012, Jonathan Rouillard wrote: >> On Thu, Oct 11, 2012 at 6:20 AM, Pavitra <celestialcognit...@gmail.com> >> wrote: >> > On 10/11/2012 03:50 AM, Arkady English wrote: >> >> Pardon my ignorance, but what is this? Or is it playing code golf in >> >> an obscure language, because I'm up for that! >> > >> > That's what it is. >> >> Oh god. If so, I'm definitely in. =) > > That's what it is, all right. :) > > On Thu, 11 Oct 2012, Benjamin Schultz wrote: >> How would BF golf differ from BF joust, which you ran some years back? > > Actually, I came up with BF Joust *after* we (can't remember who ran it?) > held a BF golf contest here. That BF golf ruleset at that time had a few > bugs in it; here's my quickie proto: > > 1. To begin a round, the Contestmaster will post a programming task, to > be completed in BF. E will *also* announce Par, which is the length of > a program e wrote to do the task. [E is encouraged not to optimize eir > solution; but this ensures the task is confirmed reasonably doable in > reasonable time]. > > 2. The first player to submit an Par-or-under working solution (hereafter > the Leader) gets P points plus 1 point for every character under Par e is. > No time limit for submission of the first solution, though if long time > goes with no submission, the Contestmaster can end the round. > > 3. The Contestmaster will announce the length of the Leader's solution. > This begins the 1-week shoot-off. > > 4. At the end of the week, the anyone (including the Leader) who submits > a shorter solution than the Leader's gets (for eir shortest submitted > solution) 1 point for each character under the Leader's initial shot. > The player with the lowest score gets a bonus of Q points. All solutions > are published. > > 5. The first person to N points wins the contest. > > At issue is that the the values of P, Q, and N can have some impact of > strategy. E.g. Do you submit an unoptimized under-par solution early to > get the bonus, knowing this means many people will get points from more > optimized answers? Or hold off? > > -G.
Does "length of a program" need to be defined? (In characters, I guess?) In addition, you may want to include provisions for the contest master to check that entries actually work.
