6
\$\begingroup\$

So, recently this question (on which I have the only answer) has been bumped and as a result I have noticed several upvotes coming in on my answer.

Today I received some downvotes, and as a result became aware that the question has been marked as off topic. The reason given is that it has "no objective winning criterion" when in fact it does have one:

The winner will be the person who first posts a program which accomplishes all of the following

As such it is a "fastest gun in the west challenge." I looked to see what the consensus was on Meta, and the most relevant post Are pure programming puzzles on topic? shows divided opinions on whether such challenges are off topic - in fact there is a slight leaning towards them being on topic, though not the best choice of winning criterion.

Clearly the OP of the Schlafi challenge recognised that significant time would be required to answer this question, and thought that fastest gun in the west was the best fit.

The current situation, however, is that the "winner" is already decided and this is claimed to discourage further entries. But the action taken today (closing the question) does nothing to address this issue as it now completely prevents further entries! This is especially true as the OP of the question is not particularly active on PPCG.

So, I have two questions:

  1. Are those pure programming puzzles which take a significant amount of time to answer on topic? It's clear that the OP thought completing the challenge at all was sufficient achievement to award the acceptance (and there are a number of similar questions around.)

  2. What should we do with this question now?

a) Leaving it closed achieves nothing.

b) Should it be reopened without editing?

c) Should it be reopened with different winning criteria, what should they be and is the OP the only person who can choose? I would welcome an answer more mathematically elegant than my own (and assume that's the kind of answer the OP would like to see) but that wouldn't likely be achieved by setting code-golf as the winning criterion. So what else? In order to encourage mathematically elegant answers, dare I suggest popularity contest? (ducks and runs for cover...)

As an aside, the behaviour of punishing the answerer with downvotes for an issue with the question seems somewhat irrational to me, but it's not a big deal, and did alert me to the fact that the question had been closed.

EDIT

Per Peter's answer, it seems I misinterpreted the meaning of "pure programming puzzles". I did search the whole of meta for "fastest gun", read quite a bit (rather quickly) and didn't find anything else relevant, so wasn't sure where to look for whether fastest gun is an on/off topic winning criterion for questions such as this one (though of course I understand why it is not a good winning criterion.) Peter's link is indeed more relevant.

\$\endgroup\$
  • 5
    \$\begingroup\$ Leaving it closed does not achieve nothing. It prevents people from thinking that the challenge is on-topic. If we do wish to prevent such challenges in the future we ought to close this one. \$\endgroup\$ – Sriotchilism O'Zaic Feb 5 '18 at 21:22
  • 1
    \$\begingroup\$ @HeebyJeebyMan sidestepping the issue of whether the winning criterion used makes this question on or off topic, very few people will be looking to an old question with little traffic as an example of a good or bad winning criterion. The better thing would have been if someone had raised the issue when this question was posted, but nobody picked it up at the time. It's true that an OP more familiar with this site would likely have picked a different winning criterion. \$\endgroup\$ – Level River St Feb 5 '18 at 21:32
  • \$\begingroup\$ Our perspectives change now that there is a solution. When first asked, I did not think anyone would come up with a solution. I think fastest gun in the west combined with a bounty provided a better incentive to at least get an answer than code-golf. In fact, I am not sure code-golf and a bounty are compatible, since code-golf should always go to the lowest number regardless of when it happens whereas a bounty gets awarded and cannot be moved. Now that FGITW is awarded it seems our only recourse is to reask the question as code-golf and essentially have 2 questions. \$\endgroup\$ – Tony Ruth Feb 6 '18 at 21:22
  • \$\begingroup\$ @TonyRuth Thanks for dropping in to comment, I honestly didn't expect to hear from you, I literally just edited the original question to codegolf. I don't see the need to reask it. It means this one would still closed and the other one would be an identical task. Our test for duplicate questions is that if an existing answer answer to an old question is a valid answer to a new question, then the new question is a duplicate. As that would clearly be the case, I dont think reasking would be deemed possible. \$\endgroup\$ – Level River St Feb 6 '18 at 21:31
  • 1
    \$\begingroup\$ @TonyRuth as the comment you've left on the question itself indicates you're OK with editing it to Code Golf, maybe we should leave my edit to Code Golf. It means that if someone posts an answer shorter than mine, you should unaccept my answer and accept the other: I lose 15 rep and the other gains 15 rep. It keeps the question alive. Bounties are separate. You or I or anyone else is free to offer a bounty on any question at any time and under any criteria (such as a particular language they want to see or whatever.). \$\endgroup\$ – Level River St Feb 6 '18 at 21:37
4
\$\begingroup\$

I think it is off topic and should be closed until further notice

I recommend you take a look a Martin Ender's answer to Why do we have objective winning criteria? Here Martin outlines what our objective winning criterion requirement accomplishes. I'm not going to talk about the whole answer but I think it is definitely worth a read. But I am going to highlight a few sections that I think are important. In this section Martin discusses what a challenge with an objective winning criterion feels like in comparison to challenges without any criterion.

Here is how a challenge works that has an objective scoring: you come up with a solution that is valid, and may or may not do reasonably well. However, you have a pretty good idea how you can or cannot improve that score, and if you can, you do. At some point you'll think "yeah that looks reasonably good, I'll go with this". But then you'll see the other participants' scores and notice that you're only a few "points" behind (whatever those are in the specific challenge). And suddenly you've got a very specific goal: come up with something to beat the other score. That's a tangible goal and a motivation to try new things, and it eliminates the problem of thinking "yeah I think this is as good as it can be" before that's really the case.

Challenges that operate on a fastest guns in the west approach while objective, don't do any of these things. Once you have an answer that works (or you at least think probably works) you post it. Fastest guns in the west does not allow you to improve your score, it only gets worse as time goes on. There is no drive to make your answer better than the bare minimum.

Fastest guns in the west may be a objective winning criteria, but it doesn't do what we want from an objective winning criteria. So if what Martin says we want is really what we want from a question, I think it ought to be considered off topic.


As for it's closure I think off-topic questions should be closed, that way they can either be fixed or replaced before any more answers are added.

\$\endgroup\$
3
\$\begingroup\$

As such it is a "fastest gun in the west challenge." I looked to see what the consensus was on Meta, and the most relevant post Are pure programming puzzles on topic? shows divided opinions on whether such challenges are off topic - in fact there is a slight leaning towards them being on topic, though not the best choice of winning criterion.

...

  1. Are those pure programming puzzles which take a significant amount of time to answer on topic?

This question is predicated on the challenge being a "pure programming puzzle", but it isn't.

The linked question is not about whether FGitW questions are on topic, but whether puzzles are on topic. It quotes the tag wiki for :

A programming puzzle includes a goal, a partially completed program, and rules outlining how the program can be modified. The program is specifically designed to make achieving the goal difficult. An answer to a programming puzzle takes the program and modifies it only in ways specified in the rules, so that the goal is achieved.

The Schläfi challenge is clearly not in this category, and so to use answers to that question to argue that the challenge is on-topic requires decontextualising them. The more relevant meta-question is The Tag Categorisation Project:

Winning Criteria/Question Type

Every question should have at least one of these tags.

...

The Schläfi challenge does not have any of those tags, and nor could any of them be applied.

TL;DR: the answer to question 1 is that this question isn't a pure programming puzzle, and it is off-topic.

\$\endgroup\$
3
\$\begingroup\$

I have edited the winning criterion to Code Golf.

I left a comment for the OP but have had no response.

As no other ideas have come to light, I have revised the winning criterion to code golf. I believe the question is now on topic and can be reopened.

If people disagree with my actions it can be rolled back.

I will refrain from saying any more in this answer, so that it can be voted on in terms of my actions, rather than my philosophy. I will edit my other answer with some more philosophical comments.

\$\endgroup\$
  • 1
    \$\begingroup\$ OP has responded and seems happy with the suggestion. Can we get this open now? \$\endgroup\$ – Level River St Feb 6 '18 at 22:58
0
\$\begingroup\$

Now that one of the close voters has posted, I give my own thoughts as poster of this meta question.

Heeby Jeeby Man has made some good points, which apply very well to new questions. The point of an objective winning criterion is to encourage creativity. Therefore if no or a poor objective winning criterion is included on a new question, it may be commented on and temporarily closed while the issue is sorted out. The purpose of closing, except in cases where the question is not salvageable, is to improve the question so that it can be opened again in the near future. Hopefully this will lead to more interest and better answers

Silently closing an old question does not achieve any of this.

Old questions may have been written when what was on/off topic is different to our current standards. If the issue is a simple matter of the objective winning criterion, the question is salvageable. If the aim is to get more answers on the question, silently closing a question because it has a fastest gun in the west winning criterion achieves nothing. Rather than promote more answers, it prevents them. "Closing this question until further notice" effectively means closing it forever and not receiving any more answers.

What should be done

To me, leaving the question open and unedited was an acceptable course of action. I do not agree that leaving old low-traffic questions with imperfect winning criteria causes people to select inappropriate winning criteria for new questions, as they do not tend to review old low-traffic questions when posting. They are influenced by commenters and close/reopen voters in the very active phase immediately after posting their question. I think reopening the question as is would be preferable to leaving it closed indefinitely. Reopening is following the spirit of our laws encouraging creativity. Leaving it closed is followng the letter of our laws to an end which is completely distinct from their intention.

The OP is not very active on PPCG. I assume he would have liked to see other answers and would have appreciated a more mathematically elegant answer than my existing one. If we leave the question open Dennis (just might) come along with an answer a tenth of the size of mine that runs on an algorithm simulating tetrahedra in n-dimensional space that self-assemble into N-dimensional polytopes. But if we close it the chances of that happening are essentially reduced to zero. Popularity contest is not well liked and suffers strongly from the fastest gun in the west effect. The OP probably did not want to select Code Golf because either it would lead to unreadable code or because he thought that producing an answer at all was challenging enough and that the additional challenge of golfing would discourage people.

On balance and reflection, I would like to see this question edited and reopened as Code-Golf (as a default winning criterion as I do not realistically see a better option.) But I am reluctant to take that course of action unilaterally, especially as the poster of an existing answer. If others feel it is a good idea I can do edit or (preferably) somebody else or a moderator can do it.

EDIT

To answer Peter's comment on what I mean by "silently closing a question":

By "silently closing a question" I mean closing a question which is no longer near top of the "Newest Questions" list without thinking what other options are available and with out discussion.

Thus there are 2 types of what I would call "NON-silently closing" a question

The first type of NON-silent is closing a question when it is brand new, near the top of the Newest Questions list, and all eyes are on it. It frequently occurs that flaws are found in questions, and it is desirable to close the question temporarily to prevent answers while this is sorted out. Unless the question is low quality, this usually results in the question being reopened.

The second type of NON-silent closing, is to ask a meta question first, similar to what was done in "Good" Questions That are Incomplete. The question under discussion here was at least as complete as the ones mentioned in that meta post, and had about as many upvotes as one of them. I do think that when you find a high quality old question that has an issue, the order of preference should be

  1. FIX it (with discussion if necessary)

  2. leave it alone in case someone wishes to answer it

  3. Close it, but not before seeking advice on meta or chat.

If it wasn't for the fact that some of the close voters chose to downvote my answer, the closing would have gone unnoticed. And anyone wishing to answer later, rather than being "discouraged" by the fastest gun criterion would have been unable to answer without doing the legwork to get it opened (which is surely even more discouraging!) It also means the person trying to reopen a question in order to answer it would likely have to suggest or specify a winning criterion which is an awkward conflict of interest.

\$\endgroup\$
  • 1
    \$\begingroup\$ I'm not sure what you mean by "silently closing": what would non-silent closing look like? Besides, question closures are neatly listed, at least for people with 2000 rep, at codegolf.stackexchange.com/tools?tab=close \$\endgroup\$ – Peter Taylor Feb 6 '18 at 15:15
  • \$\begingroup\$ @PeterTaylor thanks, I wasn't aware of that link. \$\endgroup\$ – Level River St Feb 6 '18 at 22:57

You must log in to answer this question.

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