Problem 2959. Divide by 4
Recent deluge of trivial problems is really getting out of control. Any thoughts people?
Yes, cody is becoming less interesting.
I doubt mathworks are interested in curating cody, and giving that authority to some people may be too much as there it's always a judgment call as to what is trivial, but they could at least enforce a minimum number of different test cases. That would force posters to put some minimal effort in their problem at least. They should also enforce writing a test for the test suite that actually passes it.
Maybe there should be a way to flag problems. At the very least, a few people could have the privilege of reverting problems to drafts.
Yes. I agree that triviality is a subjective term. But you can easily classify this one as repetitive. And I definitely agree that there should be some flagging system.
Interestingly this surge is associated with a number of posters solving the exact same five problems (1, 2, 3, 167, and 189) and then creating their own (e.g. users 6047402, 6037968, 6042063, 6038651, 6034379, and 6033534), is this some new form of spam?
and all of them from @uab.edu, perhaps this is homework instead! (we should really have a talk with their TA :)
It seems that Ned Gulley is still active, though not enough to keep up with the recent spam influx. Is he the only Matlab moderator? He's helped clean up a lot of test suites before.
Also, (I hope Ned or someone else from Matlab is listening), I think that new sets of problem groups (challenges) would help spice things up. There aren't that many, to start with, and I think that having a bunch of well-written easy- to moderate-difficulty problems would help draw more people in or get others active again.
In my opinion, the best way to prevent this kind of spamming is to set a condition that a problem creator must solve all the Cody Challenge problems (or at least half of those). One who has solved these problems would definitely appreciate the effort people have put here in Cody.
I definitely agree lots of questions exsits like this one
I'm listening. Thanks for this discussion. I have been in contact with the prof at UAB who has his students post these problems. This particular flurry happens in October and February when the course is taught. I agree this is not an appropriate use of Cody. We'll be dealing with this by making a minimum Player Score before you can submit a problem. It'll be impossible to just show up and make a problem, having never played before. We'll have to see if that resolves the issue, but I think it's a good step.
Incidentally, I would delete this problem, but now it has this useful commentary on it.
Players need a minimum score of 100 to create a problem. Joshua Newcomb has a score of 130.
Ned, thanks for setting up the threshold. Simultaneously, this should stop newbies threating Cody like Answers and some spammers as well. According to the last comment by Rifat, maybe threshold of 10 solved problems would be better? It seems to be the same like 100 points, but there are badges points too. In some circumstances one can have over 100 points by submitting only one solution, like and comment: solution(10) + solver(10) + speed demon(50) + leader(20) + promoter(10) + commenter(10) sums to 110. I agree with suggestions about flagging problems and reverting to drafts - maybe in the form of warning, giving authors about 72 hours to fix problems, after that time, problem would be reverted to draft? Speaking less seriously, I would give acces to create problems to those who solved problem 9 without reading comments and asking questions :-)
@Aditya. Technically, his score excluding points for creating problems would be 60. (2x15 points for two problems - which are duplicates, 20 points for Creator badge and 20 points for Quiz Master badge would be gone)
Definitely "spam" when in addition to being trivial a valid solution is presented to the person trying to 'solve' the Problem!
I agree that a threshold is a good idea, but perhaps twenty problems would be a better level.
I am starting to feel that maybe even more than 20 problems would be a suitable threshold. Like ~50 from the Cody Challenge (i.e. about half of them), in line with rifat's suggestion from Feb 2015. (I think solving _all_ of the Cody Challenge is a bit too onerous as a threshold.) It is relevant to specify the Cody Challenge due to the large number of 'fun' or 'trivial' or 'spam' problems present in the Community category.
A further proposal I have to address this issue is to ALLOW people (who've reached the threshold) to create whatever 'fun' / 'trivial' / 'spam' problems they want, but to NOT award any points to players for solving those problems! I.e. create a new category with a name like "Not for credit" that a group of responsible people (staff and/or players) can assign problems to. That's much less harsh than deleting problems, but they'll be easier to avoid if shifted out of "Community", and there'll also be less motivation for creating them. It would take some work on the MathWorks side to adjust the point scoring to recognise this: if that's really too much effort, then at least a new Category named "Trivial" (or suchlike) could still be created.
Solution CommentsShow comments
Problem Recent Solvers1008
Vectorize the digits of an Integer
Make a run-length companion vector
More from this Author1
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!Start Hunting!