Project Euler Forum

elendiastarman wrote:About the only problem with that idea is that it would take far too much time for the staff here to look through all of the solutions, and they might be in alien languages, so to speak. I think these objections have already been made earlier anyway...

EDIT: Here are 3 posts related to this:

viewtopic.php?f=5&t=1599&st=0&sk=t&sd=a#p15689
The argument here is made against a whitelist, but it could be applied to your question.

viewtopic.php?f=5&t=1599&st=0&sk=t&sd=a&start=15#p15728
Your idea has already been proposed.

viewtopic.php?f=5&t=1599&st=0&sk=t&sd=a&start=30#p15924
Here was the response to it.

However, there is nothing else anywhere else in this topic specifically against your idea.

Hey,

I am not saying: the staff should check every problem submitted. It could be a tiny fraction of the solutions submitted; it should just serve as backup in case someone is going up the scoreboard too high. In the forum for 255, hk has been 'suspicious' of a few members, I believe, rightly so - then if users were forced to submit their solutions he could just glance at them and leave it at that. And there's no harm done if simple members not necessarily staff members can have a look at solutions too - most 100%ers don't even post their codes, unlike me, I'm sure they have nothing to be ashamed of

Another way to avoid cheating would be to give a random input to the user for which the answer needs to be computed.
For example, in most of the project euler problems, the inputs are of the order of 10^12. This gives flexibility to vary the inputs in the range 10^12+\- 10^11
You can have the answers for hundreds of thousands of inputs stored in your database. The user checks in if his answer matches the required one.
Alternatively , you can select a subset of inputs from the database(say 1000 of them) and ask the user to compute the answer for each one of them. This sub-set can vary from time to time and hence would be extremely hard to get into the forum unless you have the correct answer for all the inputs.

So you're saying that the user could potentially be required to submit 1000 answers?

yes

The sheer audacity of this astounds me. 1000 solutions submitted at a rate of 1 per second is 16 minutes and 40 seconds of typing time. You would have to require that the submitter make his program output a list of those 1000 solutions. Given that some of the problems actually do require a minute to run, you could potentially force a solver to leave his/her program running for 16 hours and 40 minutes. Checking would probably commence much faster, but you would still have the problems of checking 1000 solutions from thousands of solvers...which I'm sure PE is in no way equipped to handle. Plus, some problems require a specific set-up that simply cannot be expanded to include 1000+ solutions... How would you go about fixing these problems?

1000 is an upper limit. Even if you were to choose a set of 10 random inputs, it should be good enough. You ask the user to submit a text file which the system will check against the prescribed input. Once the user downloads the required input, the (input,user) pair can be stored in memory that can last for some time(say 10 minutes -- this would incorporate the 1 minute rule --solve 1 input/minute) after which the user would be required to download another set of input which he has to solve in a 10 minute duration. In some problems, solving 1 input in 1 minute also implies solving 1000 inputs in 1 minute(especially if the algorithm is sequential).
A more relaxed version would be to keep the random input file common for all users for a fixed amount of time(say a few hours).
But I am not sure how you will fix the bandwidth problem. One way would be to set a daily limit for each user(say 15).

It is our hope that this restriction more naturally reflects the increasing difficulty of the problems remaining and no genuine problem solver should ever experience the restriction.

Missed the target on that one I'm afraid... This "genuine problem solver" is sitting on 14 solutions here that I can't submit. And more by tomorrow afternoon I'm sure, when apparently, I'll get permission to submit some more...

Any chance that you could institute a system where people opt out of being displayed in the site rankings or whatever, and by doing this, they can remove the restriction on submissions? I'm here for the challenge of the problems, and I'd be just as happy to be anonymous and unseen; all this restriction does is forces me to track separately which problems I've solved and what my solutions are, and by the time that I get access to the forum posts to look at other folks' solution methods, the problem will no longer be fresh in my mind.

Now as my reward for getting to Level 2, now I'm on a two-day timeout.

I think that the system counted my answer to Problem 278 against my allocation of 25 submissions... I thought that new problems (for the first week or something) weren't supposed to count against you?

That was an OLD waittime system which had much more severe restraints.
And perhaps you can try to adhere to the one minute rule for solving problems. This seriously slows down your solving rate for several problems.

Apparently, it used to be the case that you could always submit an answer to a problem less than a week old?

I can't submit for the latest Problem 279, and it's only about 4 days old. It tells me I need to wait 2 days, 9 hours, etc...

Shouldn't recent problems be immune to the wait period?

I guess it's my own fault for cranking through a whole boatload of older problems in the last day...

The new wait time system introduced 20th July 2009 doesn't make any difference between old and new problems.
This system replaces the system introduced 16 th July 2009. In that olders system a difference was made between recent and older problems.

I guess what I'm saying is, shouldn't there be an exception for the new problems?

Apparently the system is supposed to stop mass submissions by other than genuine problem solvers. Nobody can ever be doing mass submissions to problems less than a week old, right?

If the system is known to you you can allways reserve a slot for a new problem can't you?
So I don't see any reason to make exceptions.

I could if I kept track of all of my submissions with timestamps. The current system gives no warning telling you how many submissions you have left before you get put into "time out" purgatory... The system frustrates me enough as is; I'm not about to add extra work to deal with it. Solving problems is fun; "busy work" to deal with the time out system isn't.

Ouch... Turns out it was a really bad idea to work my way through the problems backwards. Means I got into the "easy" ones (average 20 minutes per problem) while on level 5 --- one day of having fun got me locked out until after Easter

Especially annoying with two new problems coming out and "fast solver" scoring coming up.

I usually solve some problems while offline, then at the next time I'm online, I submit all the answers. However, with that wait time system, I can't submit them all. Maybe lower the wait time is a good idea, or use other system besides wait time will work?

umisef wrote:Ouch... Turns out it was a really bad idea to work my way through the problems backwards. Means I got into the "easy" ones (average 20 minutes per problem) while on level 5 --- one day of having fun got me locked out until after Easter

Especially annoying with two new problems coming out and "fast solver" scoring coming up.

Maybe it would be possible to revise the system to always accept attempts when a problem has less than 20 solvers? (The fastest solver list currently goes up to 20, in case the number changes later and this post would no longer make sense.)

I've never run into problems with the wait time system because I started at the beginning and went slowly, but I know I'd be annoyed in the situation umisef described.

The tool tip claims:
"You are limited to entering no more than twenty-five correct solutions within a period of time equal to your level in days, which is currently 1 days for you. This restriction should naturally reflect the increasing difficulty of the problems remaining."

However, I have only submitted 24 solutions (being a new user) so I should be able to do one more. This look like a bug.

Please forgive me the little joke, but the first problem you solved was Problem 0: registering.
Not a very hard problem I must admit.
Explanation of the "bug":
At the moment you register the time of registering is saved. This counts as your first problem.
At the moment you solve your 24th problem the time of solving that one is noted.
At the moment you solve your 49th problem the time of solving that one is noted.
At the moment you solve your 74th problem that time is noted etc.
This is done so because the levels end at 49,99,149 etc.

Today, similarly to rplunkett:

[Problems page] Problems Solved: 24 out of 320
[Profile page] Solved: 24/320

And yet "You are limited to entering no more than twenty-five correct solutions within a period of time equal to your level in days [...]".

But you say I've entered only twenty-four correct solutions in total! I have not reached the limit as stated.

Sorry, whatever the jovial explanation of this bug, this sort of off-by-one error on a mathematical website is a really bad look.