Evaluation

Hi, I want to know if speed is important for the solution, If a method can solve "any" maze but is rather slow can it lose points?
Is the User Interface important? What kind of interface do you want? Do we "upload" mazes or a textarea where mazes can be "pasted" is better?

Well I can tell you that if a program can solve more mazes than another, then the one that can solve more will win regardless of speed. The first and foremost factor is traversing the maze. What good is speed if you don't get the correct result?

I leave the user interface totally up to the programmers.

If it isn't specified, it's up to the programmer.

I can tell you, if I have two programs that perform identically (speed, accuracy, resource usage, etc), then the prettier one will win. I doubt, however, that we will be in a situation such as that.

Get your code working all well as it possibly can according to the specs given. THEN worry about the fluff

Hi all,

I have a question for the webmaster (Matt, I suppose): how do I find out if my code for the contest was submitted successfully?
And how do I find out how many coders have entered the contest?

If your code was submitted to contest@codealkers.com I will send back a confirmation that I received it. I believe I remember sending you a confirmation.

"And how do I find out how many coders have entered the contest?"

There is no way.

Hi there!



The confirmation did not get back to me! Please try to send me another.

Hi Matt,
I disagree with what you wrote. Please note, that when coding the script you might optimise different factors, such as: speed, use of resources, clearness of source code, look of UI, etc. Often it is not possible to optimise all those factors in same time - you usually need to chose which would you like to optimise and which one not. Only with such definition of clear evaluation criterias it make sense to take part in contest othwerwise you risk, that you will write the script optimising one factor and then you will prize other program, which optimied some other factors. It just like taking part in the run where the finish line is not defined.

In your original reposnse there is a sentence that programms that produce exact solution are always better. Please imagine you have two programms: one produce the shortest path with length 170 in 1 hour. The second one produces path of length 173 in 5 seconds. Which one is better is not this simple answer.

Finally, I do believe you should rethink evaluation criterias, also for the future contests.

I go back to my original post:

"What good is speed if you don't get the correct result?"

Think of it this way. A doctor gives me a liver transplant and it only took him 2 hours. He botched the job, but he got it done in two hours. My buddy got a liver transplant and it took his doctor 12 hours, but he got it right. I would rather have it done right than fast.

That's the problem with our society today, everyone wants everything done too fast. Let's worry about correct results before speed.

Matt,

Maybe I have not made it clear before, but my response was rather about the absolute need for clear evaluation criterias rather then what is more important: time or accuaracy. From your response I understand that accuracy have absolute privilage. It is OK and I like it. However I would suggest you to include clear evaluation criterias into terms of contest.

Aside, sometime it is more important to have answer fast. If I want out of the maze it may make a difference wheather I will wait a while till programm tell me what is the shortest path, or I will have to go a little longer way, but I will save a time on calculation and in effect will go out in shorter time.