Sticky

Post general comments here

by Alex on Apr 28, 2015 - 11:48:46 pm UTC
  • (3/0)
This is the place to post general comments, such as pressing bugs or broader questions.

Comments (Search)

hi what place is this like the question place or sumthing??

This is like a guestbook.

You get to ask questions, share your opinion about something, give feedback to others.

Basically anything for the general public. If you have problem-specific question, you can post them on the problem page.

that's a good idea

It doesn't work - at least for me. I got a "Fatal Error" =D
And I don't think user colours are necessary, we already have ranks.

It should work now.
Do you think it's worth it to implement problem ratings?
(It won't be much work for me, I just have to copypaste the comment rating code)

I could just take the "General" section out, and just move it to the PEG forums.

Maybe we need to split these up into multiple pages, it's getting a bit annoying to scroll down - speaking of which, why not just put the newest comments at the top?

But either way, perhaps we should make this more like a forum and split it into threads? Though this is probably a lot of work, so it doesn't matter too much.

Yes or no?

I'm not sure if I agree with the "Problem Type" category.

Once you see the problem type, you can't help but narrow your mindset into thinking about it in terms of an end result. For example, if someone tells you a problem is DP, you will immediately try to break it into logical subproblems, even if you can't quite prove the optimal substructure.

As another example, if someone tells you the problem requires data structures, you might make a preemptive judgement to optimize it rather than see through the process why the solution requires a better data structure (as opposed to a better algorithm).

Another example... someone tells you the problem is greedy, so you go through the formula of sorting and stepping through locally optimal states, without fully proving why the global optimum is the sum of greedy choices.

I can think of other examples, and all of them limit the learning experience of the participant. You can

(Post got cut off by character limit)

You can ignore the category, but even if someone accidentally glances at it, they can't help but start thinking about what they should do rather than why they should do it. And not only is that sometimes the hardest part of formulating the solution, but revealing that aspect destroys at least half of the problem solving process.

So my suggestion is that you only categorize by vague problem types which can be *immediately* inferred from the problem statement.

Here is an example list of things that I think (in general) could be immediately inferred:
- Math
- Geometry
- Algebra
- Optimization
- Combinatorics
- Graph Theory (only in some circumstances)
- Logic
- Strings
- Puzzle Solving
- Searching
- Basic Programming

etc...

But things like "Dynamic Programming", "Greedy" and "Data Structures" are definitely bad indicators in my books.

CCC programs are categorized for the purposes of classwork: when a topic is taught, this makes it easier to find problems for people to test their programs on.

But you're right, I think I'll declassify real contest problems and such.

Haha, "declassify" contest problems... "de-categorize" might have been a better word

It's so laggy when you try to view all the problems.
I just press view 50 or view 100 instead because it's way faster (and it shows all anyways)

what're you talking about there's no lag at all

No lag anymore

204 problems currently available, lol.

I think we should all stfu and let Hanson do his thing =)


A search -->button<-- would be nice on the problems page, because if you just want to change the problem sorting order you have to go to the search box and press enter. It's a bit of a pain. Or maybe just make it so submits the form automatically once the sort order is changed or soemthing.

That's strange - it's already supposed to update automatically when you change an option. Do you have javascript disabled?

The checkmark idea is on my to-do list, I will work on that sometime. :)