|Hi there! I have a few items:
(1) I was having some problems with the C++ tester on the 500 pointer in the C++ contest (weird results that looked like segfaults). When I tried to go to practice room 1 today to investigate this more fully I was getting the result "Test Error" even for a simple function that just returned the input string.
(2) History for a coder in the status window from the contest applet shows only 1 line of the test input/resulting output, meaning that it is hard for big inputs to figure out what went wrong, before results are finalized and the more comprehensive history is posted on the website.
(3) For large test case examples (such as today's 1000-pointer) it is not reasonable to ask people to type them in by hand (hard to get exactly correct, especially with variable-width font, and quite time-consuming). Some mechanism for automatic entry should be provided.
(4) The excess in rating system volatility needs to be addressed. Is there an ETA on ratings patch?
(5) Finally, I was curious as to what QA procedures for problems have been put in place, since there still seem to be inconsistencies in problem statements/ bugs in official solutions which are arising during the contests.
Thanks - Dave