# Tour of PPCG: Voting Phase

As described in my last meta post, the purpose of this post is to nominate and vote on posts that deserve more visibility.

The criterion for voting and nomination include 4 aspects: Quality, difficulty, acceptability, and visibility.

1. Quality: The post should be well written, and on-topic under our current guidelines. Furthermore, the challenge should avoid practices that we are trying to move away from (such as arbitrary language restrictions, difficult I/O models, or pop-cons that don't need to be pop-cons)

2. Difficulty: Avoid the extremes: Extremely difficult challenges may cause users to give up, and too many easy challenges makes completing the Tour less of an accomplishment.

3. Acceptability of new answers: Some problematic challenges include:

1. Challenges that are closed to new answers (most and )
2. Challenges that don't work with a large amount of answers (most )
3. Challenges with little room for new, distinct answers (such as "Hello, World!")
4. Visibility: We are actually looking for lesser-known questions here. While our 2014 challenge may still be unseen by some of our newest users, we're hoping to introduce challenges to as many users as possible. The age of a challenge matters as well, but not as much as the popularity of the challenge.

Remember, all of the above are relative criterion. The question doesn't have to be the ideal in all of the aspects, but a terrible question won't make the top 25.

Please include the tags of the question in your post so that we are able to easily choose a wide variety of questions

• I don't have a suggestion, but since the name clashes somewhat with the help tour, perhaps someone else could think of something catchy and meaningful? – trichoplax May 9 '16 at 20:18
• Name suggestion: Community Picks – ArtOfCode May 9 '16 at 20:25
• Name suggestion: Most Significant Bits. – AdmBorkBork May 9 '16 at 20:27
• "The age of a challenge matters as well": are older challenges more or less desirable? – Zgarb May 9 '16 at 20:28
• @Zgarb More desirable, but only on the basis that they are lesser-known. Many of our older questions are off-topic, and shouldn't be considered. – Nathan Merrill May 9 '16 at 20:47
• Name suggestion (if we only end up with golfing challenges): The Golf Course. – Martin Ender May 9 '16 at 20:52

## Squarefinder - Locating regular tetragons

Pros:

• Less than 600 views as of writing this.
• Only 3 answers so far, but all use totally different approaches.
• A good spec with pictures, easy to understand.

Cons:

• Somewhat difficult.
• Oooh, I really like this one. +1 for difficulty. – Nathan Merrill May 10 '16 at 14:16
• I would have to agree, this question provided me lots of hours of fun and I am sure there are better answers out there. – MickyT May 10 '16 at 19:20

## ASCII Art of the Day #1 - Double Knot

Pros:

• It has a solid spec
• It's part of a series, so users that enjoy the challenge will be able to find more like it

Cons:

• It's newerish (Posted almost exactly a year ago)
• It doesn't have a large scope of possible answers

Trivia: The top answer outgolfs Dennis

• the top answer outgolfs Dennis totally important, can't discourage newer users. :P – Rɪᴋᴇʀ May 9 '16 at 21:09

## Lattice Points inside a circle

Pros:

• Is old. This beauty is from 2011
• Despite its age, actually has a pretty solid spec
• Very few answers

Cons:

• I'm unsure how many techniques can be used to calculate, so answers may start to overlap

## Calculate practical numbers

Pros:

• Short and easy-to-understand spec.
• Only 4 answers, top answer is "not really golfed".
• Over 2 years old.

Cons:

• The time limit makes it more difficult.
• I think a question where the answers show that scoring is usually in "bytes" not "chars" would be better. – mbomb007 May 11 '16 at 19:55
• @mbomb007, that would rule out 99% of old questions, which is rather missing the point. – Peter Taylor May 11 '16 at 19:58
• @PeterTaylor Then perhaps the tour/help page should include a description of the usual proper answer format, so that new users don't pick up the bad habits of the past. – mbomb007 May 11 '16 at 19:59
• I think we can safely edit the scoring to be bytes as long as it doesn't change the score for any of the answers. – Nathan Merrill May 13 '16 at 17:57

## Generate Skolem sequences

Pros:

• It's from June 2013, so almost 3 years old, and has less than 700 views.
• Not too hard, but not trivial either, and there are probably many different approaches.
• The spec is short and simple (ignoring the weird background story).
• I do like the problem, but I'm not a huge fan of the spec. Maybe we clean it up? – Nathan Merrill May 11 '16 at 19:02
• @NathanMerrill I edited the spec to better fit our standards. – Zgarb May 13 '16 at 17:52
• Ooh, this is much better. – Nathan Merrill May 13 '16 at 17:55

Since the question of mine that was nominated here was received so poorly, allow me to nominate one of my own:

# Build a Flood Paint AI

Pros:

• It's relatively old (from 2014).
• It's only got about 3,000 views so far, even with 27 upvotes.
• It has a very granular scoring criterion, with 7-digit scores sometimes varying by as little as a few hundred.
• The optimal solution is still unknown. (In fact, this particular problem is NP-hard.)
• It's relatively easy, but not trivial, to write a solution for and prove that it works.
• There's lots of different, creative approaches that can be taken to solve this problem.
• No loophole abuse; because the challenge is algorithmic in nature, there's not really a flaw in the spec that can be exploited to produce a trivial good-scoring answer.
• No esolang abuse; because the challenge is algorithmic in nature, golfing languages do not have a natural advantage in producing better-scoring answers than long-form languages.

Cons:

• It requires a large input file, which is mirrored on a third-party website.

• There are a bunch of existing solutions (although the last time anyone actually posted a new answer was May 2014, two years ago).

• It has one of my famous "last-place reference solutions", which is something I do with a lot of my code-challenge questions that was recently declared as frowned upon.

The answer for this problem in particular, however, requires a bit of proof to show that it's actually valid, so I think it can stay there. It also acts as a bare-minimum guideline — if your solution is not at least as good as this one, it shouldn't be submitted, because there's something wrong with it.

• Also, this seems to be the only non-code-golf question being presented. Nobody has any other code challenges? – Joe Z. May 14 '16 at 6:39
• Nope, other scoring methods are allowed. However, there are a couple of other scoring methods (fastest-code, cops-and-robbers, king-of-the-hill) that don't work too well, which is why the majority of the answers are code-golf – Nathan Merrill May 14 '16 at 12:03
• That said, I'd recommend splitting up the two problems into two different posts (so we can talk/vote on them seperately) – Nathan Merrill May 14 '16 at 12:05
• Alright, done. meta.codegolf.stackexchange.com/a/9182/7110 – Joe Z. May 14 '16 at 16:24

## Reverse Indentation

Pros:

• Written in 2014 (barely)
• Has a humorous, interesting spec
• Is easy to understand, but has lots of approaches
• As of this writing, only has 10 answers
• Teaches the importance of Java scripting

Cons:

## Lego gear ratios

Pros:

• Simple, concrete and fun spec.
• Old, from September 2012.
• There seems to be room for more answers.

Cons:

• The I/O formats are strict (but not too annoying).

Here's also another test-battery question for your consideration:

# Build a minimum-clue Sudoku unsolver

Pros:

• It's only got about 600 views so far, even with 11 upvotes.
• It's only got 3 answers right now, so the solution space is largely unexplored as of now.
• It has a very granular scoring criterion, with 7-digit scores sometimes varying by as little as a few hundred.
• The optimal solution is still unknown.
• There's lots of different, creative approaches that can be taken to solve this problem.
• No loophole abuse; because the challenge is algorithmic in nature, there's not really a flaw in the spec that can be exploited to produce a trivial good-scoring answer.
• No esolang abuse; because the challenge is algorithmic in nature, golfing languages do not have a natural advantage in producing better-scoring answers than long-form languages.

Cons:

• It requires a large input file, which is mirrored on a third-party website.

• It's a bit harder than the Flood Paint problem to prove that a solution will work.

• It has a last-place reference solution just like the Flood Paint one. This one isn't quite as novel though, so I can remove it if desired.

• I don't like "with shorter code breaking a tie" in a code challenge, but +1 anyway it's a really interesting challenge. – trichoplax May 15 '16 at 21:18
• @trichoplax The "shorter code breaking a tie" is a catch-all in the rare case that two solutions have exactly the same score, which has happened with the Flood Paint question. – Joe Z. May 22 '16 at 23:30
• I understand that - I just much prefer "first posted" as a tie breaker, so there is a single objective rather than also golfing. There's more in this meta answer (plus the opportunity to vote...). – trichoplax May 22 '16 at 23:42

## Word Search Puzzle

Pros:

1. Common problem that is quickly understood
2. Doesn't require a dictionary
3. Low chance of solution overlap
4. The accepted answer has a great explanation

Cons:

1. May be a bit too easy
• If you want a slightly harder version, I've got a similar question from 2011: codegolf.stackexchange.com/questions/3383/wordsearch-solver – Gareth May 31 '16 at 10:08
• (With apologies if it's a bit egotistical to propose one of your own questions for this :-) – Gareth May 31 '16 at 10:09
• @Gareth post it, and let us vote on it :) – Nathan Merrill May 31 '16 at 12:50

## Solve an 0h n0 board

### Pros

• Low view count (~450)
• Only one answer
• Fairly easy to understand (At least in my opinion, correct me if I'm wrong)

### Cons

• Quite difficult (I haven't even figured out where to start solving it :)
• Fairly new (~2.5 months old)

# Tiling Given Vertex Configuration

• Has several levels of complexity (beginners are able to start easy)
• Very few answers

## The Floating Horde

Pros:

• Great theme
• Part of a series, introducing users to other questions
• 2 years old

Cons:

• It was pretty popular back in the day
• The I/O requirements are needlessly restrictive from a modern view. Not sure what you can do about that. – xnor May 13 '16 at 3:19
• I'm not sure what you are considering "restrictive". It allows input via STDIN, arguments, or from a file. That pretty much covers most of the bases. – Nathan Merrill May 13 '16 at 17:58
• I meant that the list has to be input and output as a space-separate string. At least, the existing are doing that processing. – xnor May 16 '16 at 10:50

# Self-Shortening Prime Tester

Pros:

• Straightforward challenge
• Introduces users to quines

Cons:

• The "sub challenge" (prime checking) is pretty simple, so we may end up with duplicate entries

# Decimal Multiplication of Strings

Pros:

• It has a solid spec

Cons:

• It's newerish (Posted a couple months ago)
• It doesn't have too much of a large scope of possible answers. Not that bad though.

# Lossy ASCII art compression

Quality: Unambiguous spec, script to verify and score outputs, objective validity criterion and scoring.

Difficulty: Challenging enough to be interesting, but naive approaches are possible because the required compression quality is a lower limit, not a strict target, allowing a range of solutions gradually getting nearer to optimal. Optimal is likely unattainable, so the challenge is open ended. This has a nice balance of easy to enter and difficult to win.

Acceptability of new answers: Does not require challenge author's involvement to score new solutions. No upper limit on number of solutions. A wide variety of approaches is likely to be possible.

Visibility: Posted July 2015. No answers yet. 12 upvotes and 597 views. An intriguing but accessible challenge that seems to have flown under the radar.

• It is well defined, but it took me several reads to comprehend what "lossy" meant for this challenge. Also, does the verification script really have to be in CJam? – Nathan Merrill May 16 '16 at 3:51
• I agree. CJam isn't a good choice for a verification script. – mbomb007 May 18 '16 at 16:32
• Would this be better accepted if someone wrote a verification script in python or as a Stack Snippet? – trichoplax May 22 '16 at 15:34

## Print every character your program doesn't have

Pros:

• It's from 2013, so it's fairly old in terms of challenges.
• The newest solution was posted over two years ago.
• It's one of our highest voted questions.
• It's a reasonably simple challenge that leads to interesting solutions.

Cons:

• It has a lot of solutions in many different languages already, so the potential for duplicate solutions is high.
• Though it isn't as popular now, it was extremely popular when it was posted, so it may not be a great fit for the "lesser-known" criteria.
• There's a lot of abuse of loopholes that later became standard forbidden loopholes in the existing answers.
• While there are a lot of submissions already, its quite easy to come up with alternate solutions due to the nature of the challenge. – Nathan Merrill May 9 '16 at 8:39
• It's rather unfortunate, though, that the top voted answer is nowhere near competitive and that the second top voted answer is a standard loophole. – Doorknob May 9 '16 at 11:13
• I don't conside that question one of PPCG's best. I think it's actually think it's one of our lower quality questions, and raising it up may introduce a return of poorly formatted questions. It allowed what was already a known loophole and had answers that were voted up despite it. I had to take a short hiatus from here after that, but I'm glad to see how far this community has come with its rules and standards. – miles May 10 '16 at 1:05
• @miles The challenge would be much improved if solutions which violate loopholes were removed, even if they predate the loophole being banned. The spec itself is fine. But, alas, I am in the minority in that debate. – user45941 May 13 '16 at 0:17
• I come across this thread to find that the one question on this list written by me is also the lowest voted candidate. Are my puzzles that bad? :( – Joe Z. May 14 '16 at 6:08
• @JoeZ. I like the challenge. I think the downvotes are because of the rampant loophole abuse in the solutions (which was not your fault; it's because the standard loopholes weren't banned back then, and people don't want to retroactively apply those bans to existing answers for some reason). – user45941 May 14 '16 at 6:21