15
\$\begingroup\$

Previously:


We, as a community, still seem to be very divided on the concept of challenges ... I'm not speaking about PopCons in general, just those that are specifically related to graphical output or artistic flair.

I had thought that we had this sorted out a few months ago, and came to the conclusion that challenges need an objective validity criterion in order to be allowed on this site -- that's where the "Guidelines..." Meta post linked above has, but unfortunately there's nothing close to consensus on that post, and most answers don't actually answer the question -- but given the reaction to Helka's recent "Cracked Soil" challenge, it seems that's not the case.

However ... just take a peek at the below ...


Relevant challenges that were closed:

Relevant challenges that were allowed:

Relevant challenges that were re-done to not be a PopCon, after pushback from the community:


Given all of the above, and the fact that this is still a very divisive topic, and the fact that we as site do not like subjectiveness, the question I'm posting to Meta is this:

What, definitively, constitutes an objective validity criterion when formulating a subjective artistic merit graphical-output popularity-contest challenge?

What, specifically, needs to be included in the challenge for the challenge to be on topic?

\$\endgroup\$
13
  • 6
    \$\begingroup\$ The number of rules to post a challenge on this site is starting to get a bit ridiculous. "generate a unique 500×500 pixel image that looks as similar as possible to these 6 real life images of cracked soil" is a perfectly clear way of describing validity unless you don't act in good faith or have the mental flexibility of a computer. \$\endgroup\$
    – Fatalize
    Jun 17, 2016 at 13:55
  • 6
    \$\begingroup\$ @Fatalize It's perfectly clear, but nowhere near objective. \$\endgroup\$ Jun 17, 2016 at 14:10
  • 3
    \$\begingroup\$ @MartinEnder Popularity contests are subjective by nature, so asking for an objective way of judging the validity of a submissions makes no sense. \$\endgroup\$
    – Fatalize
    Jun 17, 2016 at 14:34
  • 8
    \$\begingroup\$ @Fatalize ...which is why a decent amount of people think that popularity contests should have no place in this community. \$\endgroup\$ Jun 17, 2016 at 14:35
  • 4
    \$\begingroup\$ @MartinEnder Then let's disallow them and stop asking questions that make no sense \$\endgroup\$
    – Fatalize
    Jun 17, 2016 at 14:37
  • 8
    \$\begingroup\$ @Fatalize ...aaand then there's the equally decent amount of people who think popularity contests are the best part of this community. \$\endgroup\$ Jun 17, 2016 at 14:41
  • 1
    \$\begingroup\$ @MartinEnder True, but this question still doesn't make any sense. Either people act in good faith and start to accept subjective criteria for subjective challenges, or nothing changes. \$\endgroup\$
    – Fatalize
    Jun 17, 2016 at 14:47
  • 1
    \$\begingroup\$ I'm not sure if it would be better to allow them or not, but where we're at right now isn't helping anyone. They're in this awkward limbo of sortof on-topic as long as they meet a bunch of poorly defined criteria. \$\endgroup\$
    – DJMcMayhem
    Jun 17, 2016 at 14:54
  • 2
    \$\begingroup\$ @MartinEnder, the second class of people must be really bored with PPCG. Popcons seemed to be dead until yesterday. \$\endgroup\$ Jun 17, 2016 at 16:13
  • 1
    \$\begingroup\$ @PeterTaylor I don't think they were dead so much as just mortally wounded by all of the hotly contested discussion surrounding them lately. \$\endgroup\$
    – Alex A.
    Jun 17, 2016 at 16:49
  • \$\begingroup\$ @MartinEnder You could make such challenges objective if you defined a way to calculate image similarity. But that doesn't seem to be the topic of this question. \$\endgroup\$
    – DanTheMan
    Jun 17, 2016 at 17:29
  • 2
    \$\begingroup\$ @DrGreenEggsandIronMan Well, the issue with allowing them is that allowing them can never mean "all popcons are on topic", because we also don't say "all code golfs are on topic". Some simply don't meet our general quality standards. The issue is that the suitability of popcons is as subjective as their judging criteria, so as long as they're allowed I think there will always be debate whether any given popcon is too broad or not. \$\endgroup\$ Jun 17, 2016 at 17:38
  • 4
    \$\begingroup\$ I think that sometimes pop-cons attract poor answers that get far too many upvotes - much more often than poor code-golf answers do. This is because of HNQs as well as being simply a pop-con. They're either poorly received or are recieved just well enough to hit HNQ and explode with upvotes from off-site users \$\endgroup\$
    – Blue
    Jun 17, 2016 at 21:01

3 Answers 3

9
\$\begingroup\$

Popcons have needed to be popular

Note: This is simply my perspective on what has been happening, not what should be happening

We have a challenge writer, Helka Homba, who is quite popular in our community. He writes great challenges: They are interesting and easy to read. His questions each get about 20 votes (rough guess), and lots of answers.

His pop-con challenges are also popular. Whether it is by the virtue of his name being on the post, or because they are well-written, they become quite popular. Nearly all of the listed allowed challenges are written by him.

Now lets look at the disallowed challenges:

  • Make the pattern, What time is it, Warhol: Unpopular (received few upvotes and few downvotes)
  • Mona Lisa: This challenge was pretty popular, but was in the middle of a debate as to whether it should be allowed.
  • Tweetable Mathematical Art: Immensely popular, but because it was clearly off-topic under our previous guidelines, it got closed later on

When judging whether or not to close, we, as imperfect humans, often appeal to whether or not we like the challenge. Challenges that are enjoyable to read/compete in are going to have less close votes. While our official guidelines have asked for objectivity, we are more likely to think that a challenge is objective if we liked the challenge in the first place

Therefore, if we take a stance, we should apply those guidelines regardless of the popularity/enjoyability of the post.

(All of this said, my personal opinion is that the majority of the challenges listed above are on-topic for this site. To be more clear: I'm not saying that Helka's posts are off-topic; I think they should remain open. However, if we are making exceptions to the rule because of his name, we should stop)

\$\endgroup\$
2
  • \$\begingroup\$ Playing devil's advocate here: challenges with severe problems that lead to closure often don't get upvoted nearly as much as challenges without those problems. You may have the cause and effect reversed. \$\endgroup\$
    – user45941
    Jun 17, 2016 at 23:22
  • \$\begingroup\$ If that were true, we'd see lots more downvotes, but the questions posted all have positive scores \$\endgroup\$ Jun 18, 2016 at 0:43
5
\$\begingroup\$

Explicit distinction between objective and subjective rules

First of all, I do agree with Nathan Merrill's post. I'd like to add following point:

When writing down the rules for the challenge, the author should make an explicit distinction between

  • objective rules: Things that can be objectively be measured (by a "dumb" computer). These rules form/include the objective validity criterion
  • subjective rules: The goal of the challenge, which some submissions do achieve better than others. (The heart of the pop-con, the artistic merit that should be achieved)

And both those sets of rules should also be declared as the objective or subjective rules. This distinction should be clearly present in the challenge.

Example: As an example of this I'd like to refer to my patch-the-image challenge, as I wrote this challenge after some of those mentioned discussions, and I very carefully tried to implement the points that were established.

The objective rules were just that outside of the given area, the image must remain unchanged, and that the program may only read from the outside area.

The subjective rules said that the inside area should be filled, such that it is as unnoticeable as possible, basically as a list of voting suggestions.
There is no way how this could be judged objectively. Some of the answers did a relatively bad job (my personal opinion), some of the answers did a very good job at achieving this. This is where we need the judgment of the people who are reading the challenge, and what makes this a popularity-contest.

\$\endgroup\$
-6
\$\begingroup\$

They should still follow our rules

(No this is not about any specific pop-con, I'm just putting this here)

If you go to "Ask a Question" on main, you'll see two main rules that answers should follow:

  1. A clear specification of what constitutes a correct submission, so that it is possible to indisputably decide whether an entry is valid or not. Test cases are highly encouraged.
  2. An objective primary winning criterion, so that it is possible to indisputably decide which entry should win.

So pop-cons fulfill #2 but #1 is where new users go wrong. Here are some hypothetical questions (both are bad, but one is better):

Write a program to add A and B, most votes wins

Write a program to output a randomly drawn goat, most votes wins

Okay, so it's pretty clear which one is more on-topic, but here's my stance:

    Pop-cons are only a winning criteria, judge the question's on-topicness, regardless of the pop-con tag.

Some examples of on-topic pop-cons would be:

  • Make your language unusable – it could very well be a code-golf competition, but pop-con encourages creativity in the answers, "creativity" in the output isn't bad, but in many cases it can become a way for loopholes to open
  • Draw an Image as a Voronoi Map – Answers can be objectively specified whether they are correct, or invalid. Downvotes shouldn't be what make the specification.

But what's brought up is that there is a specific (objective) task that answers must do. Pop-con can be used for answers to creatively solve the problem, and have voting descide the best solution, but that's no excuse for not having an objective specification for the task.

In conclusion: the task to be done should be clear, an answer can be objectively measured whether or not it fullfills the question's criteria without requiring downvotes to enforce rules. Subjectivity is perfectly fine as long as there is still some sort of objective specification

\$\endgroup\$
1
  • 1
    \$\begingroup\$ Since nobody has explained their downvotes, mine is because I think the current rules are bad and we should change them. I agree with how you interpret and apply the rules as they are and I don't like the result. \$\endgroup\$
    – xnor
    Jun 23, 2016 at 8:43

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .