Sure. Although I don’t really feel entitled to a strong opinion on long challenges as I generally avoid them now. Nonetheless, I’ll (mostly) address each point individually.
Benefits
Not really much to say here, you have good points. However, I think something like the DSA learning series is probably better for learning the types of data structures and algorithms - you get a series of hints, problems specifically tailored for certain topics, and can ask for help if necessary. For long challenges, right at the end, you get an explosion of pent-up questions from participants and probably many of them end up ignored or unaddressed. I think the best thing about long challenges for beginners, though, is the motivation of rating, as it usually keeps people interested. I also think it’s the worst thing. More on that later.
Demerits
Easier to increase rating in long challenges
Inevitably this is true, however, this is akin to the so-called rating inflation on Codeforces, where the actual effect is the rating span being wider due to having way more participants. I think the real issue with rating and long challenges is that they’re so different from short contests like cook-off and lunchtime, and it’s hard to aggregate all three into a single rating. That’s kind of why the individual contest ratings were, and still are, a nice idea.
Some good coders, once start the contest, are bound to invest their 10 days otherwise their previously gained rating would decrease.
Totally agree on this. I’m not sure how many people know how hard it is to maintain red - you basically have to place top 10 or 15 every single contest on short contests, maybe it’s a bit more generous on long contests. And I’m not too keen on spending 10 days stressing about hard problems only to inevitably lose my fake red. Short contests are much less of a time investment and are thus easier to get into.
Major factor is the telegram groups, whatsapp groups, where solutions, test cases, the actual logic of the problem are shared.
Yeah, this sucks. It’s part of the problem with having rated long contests - they motivate cheaters along with the rest of us. But to be honest, it doesn’t affect high-rated competitors, and for people with lower rating, it kind of balances out the rating “inflation” you see :P. Of course I don’t condone cheating, but that’s kind of an effect of it. Since we can’t really stop it, we might as well look for the silver linings.
Ridiculous to compare the rating of two coders, the one who has gained it through long challenges and the other who has gained it through short contests.
Yep
(It is one of the reasons why Codechef rating is not given value in the community)
Although I disagree with this point. Long challenges are hard, especially in Div. 1, and I have respect for anyone that can legitimately solve, in any time frame, the insane problems that are often posed.
Making the Long Challenge unrated, would highly reduce the number of participants.
Probably, but those who are really interested in learning would stay. Although there isn’t much of a reason to completely eliminate rating from the equation - I see that from your suggestions below.
Codechef is a great site, with a great UI and beginner friendly.
Totally. This community is a lot less toxic/stupid than that on Codeforces, for example. Still some, but less, and the admins are mostly great at least on this forum!
Continuing the way it is going, Codechef will surely lose its importance in the near future. Deserving candidates feel sad when a non-deserving person cheats and gains rating in the long challenge.
To be honest, it definitely has already. But this community is growing and thriving so far, and it doesn’t seem like it’ll get too much worse over time.
Mere use of plagiarism checkers won’t help as it is very easy to change one’s code, the variable names, the syntax and all.
karangreat vibes
Possible solutions
Keep the long challenge rated, but don’t count Long challenge rating in calculating the overall rating . The codechef profile displays graphs separately for Long challenges and short contests.
Possibly a good idea. Or maybe we can also emphasize the separate ratings (like somehow, on the site/discuss profile) more.
Introducing a new short contest , making a total of 3 short contests per month. May be naming them as breakfast, lunch, dinner.
Sounds like a good idea, but…
It is not that much of a difficult task as there are many good problem setters out there.
The issue with this reasoning is that CodeChef has scheduled contests - even if there isn’t a good batch of problems for the month, there’s still a requirement for contests, and this can lead to inconsistent and low quality. Introducing more contests probably won’t help this and could further hurt CodeChef’s reputation.
My thoughts
First of all, I really like the format of this post. Thank you for actually providing reasoning instead of a jumbled mess of broken English and (often unreasonable) suggestions.
As for my own thoughts on the format, well… I don’t have much experience with it. All I’ve done is a couple a year ago and placing 4th in April Long (div. 2 of course) because I was severely underrated at the time. But I think the rampant cheating, which seems to be the biggest issue and is delegitimizing ratings, can actually serve as a motivation. Don’t complain about cheaters, work hard and be better than them!
Also, I think there’s nothing that can be done about cowards like me who avoid contests to save rating, and keeping it rated is a good idea because it adds to the motivation above and will keep people interested. Alright, I’m repeating myself. So,
Thanks!
no u
