@Gareth If you dislike code-trolling, have you considered making a post on meta to get rid of it as a valid challenge form?
@Gareth Also you may wish to consider writing a useful and valid explanatory comment such as "I dislike code-trolling" when voting to close a question. An explanation of your personal tastes will be very useful in helping others learn how to tailor to your likes and dislikes when forming a challenge to post on the site.
@Gareth I do feel your pain, though. I also agree that things that I dislike are impossible for me to ignore. When I see them, I can not really focus on the other things I do like any more. They generally sap all joy out of everything else, and I also feel compelled to mention my disdain for them at every opportunity. Things I do not like have to go away. I can not really proceed to enjoy anything else until they are gone, and I definitely cannot overlook anything like that. I feel you.
Another great example is when somebody writes "have fun" at the end of the question. It's so noisy! Those two words at the end utterly distract me from the question at hand, and I can barely focus on finding an interesting answer to the challenge. Challenges that explicitly tell me to "have fun" are virtually impossible to enjoy. Know what I mean?
codegolf.stackexchange.com/users/737/… Oh weird, here's a whole list of rubbish code-golf. There's a couple great questions in there, but some how, even the most modest code-trolling "garbage" seems to garner more interested participants than that junk. I wonder why there's so much of that. Who keeps posting that stuff?
@PeterTaylor While you're right, I have to say, it's kind of too easy with Gareth. Everything I have ever seen him write, aside from some decent challenges here and there, is so utterly inconsistent and self-gratifying that it's virtually impossible to respond to it in a serious manner -- even serious responses end up looking like trolling. You may think it is trolling, and yet you will not find a falsehood in any of my above statements.
Two people came and downvoted every answer on codegolf.stackexchange.com/questions/25207/…. You guys should be ashamed. PCG is kind of a fail. Its close reasons are inconsistent garbage, its tags mislead people into believing something is acceptable when its not, and the only people that bother talking on meta are the people who think the entire site should be some sort of serious-business-only over-constrained anti-fun trash heap.
Such a wonderful idea for a site, squandered by pretentious users and worthless moderators. An eyesore of an SE site. Its low popularity should be a signal. There are other places where "programming challenges" are actually acceptable. I'll be there, along with everybody else outside of PCG's little circle-jerk club.
@JasonC I don't think that insulting people is the best way to get your point across. Calm down, step away from your computer for a while, and perhaps you should come back when you're a little more relaxed.
(I'm not sure what you "Edited to remove during grace period", but I'm pretty sure I don't want to know.)
Did JasonC just talk to himself for hours or did someone else remove all of the comments that he was responding to? Or did I ignore someone on accident?
Better question: did someone star those rants? Who and why? O_o
and @JasonC, now that I've followed those links and realized that you're downright insulting @Gareth and acting as if he's some troll, I'm debating whether or not to flag for mod attention.
Guys I figured out the formula for rep on Stack Board and Card Games. Just write loooooong answers with tons of buzz words and make sure you say "Don't forget that it's all just fun and games."
@Rusher well i did rolled it back because with the change it looks like it's all one single input when it is really several example of input (if i understood it correctly)
PPCG in three acts
Act 1: Happy Times!
2011 - Nov. 2013
Not much traffic on the site, but our small community is happy and thriving, and when a new question is posted it's almost always high-quality and enjoyable.
Act 2: The Dark Era
Dec. 2013 - Jan. 2014
Systematic invasion and sneak a...
There is no provision to unwiki things en masse. Some moderators use a script for that sort of thing, though you have to be careful with that too, since the site has rate-limiting measures and will block you if your exceed its limits.
@TheDoctor There are apparently two reopen flags on that question. I don't really want to burninate the question if people are in dispute about its suitability on the site.
I wanted to put a comment on the question: This site should not be a place to take the piss out of SO users. But I haven't to avoid inflaming the situation.
@ChrisJester-Young but ain't there some automatic action taken by the site if a post is flagged by many users for the same reason? (just asking out of curiosity as i have some vague memory of a post explaining that on meta.se)
I like this one the most though. The board games community is like a bunch of teddy bears, SO is just shallow because they have sooooo many people, but you guys are pretty deep.
Re: code-trolling - I know opinions are divided on them, but can we at least agree that any that come directly from a question on another SE site, and are therefore mocking other users, should be closed? It feels too much like a form of bullying to me.
@Gareth I agree that questions coming directly from SO seems a bit mean. However, if we don't want that, we should definitely change the wording of the [code-trolling] tag wiki:
> Suppose some Stack Overflow user asks a homework give-me-teh-codez question. The task is to give code that works, but is useless, severely frustrating the OP.
I need to convert shortened versions of month names to its longer representation (e.g "Dec" -> "December"). The months name is a String and I would rather not convert it to a date object first.
Is there an easy way to do this?
Edit:
My problem was associated with inserting "Dec" as a table name...
@PeterTaylor Opinions are divided on that, to say the least, and I've voiced my opinion on it in the past. What I'm saying is that if we are to keep the tag, but don't want obvious cut/paste questions from SO, we should change it. As it is, it seems to encourage exactly that, and nothing else.
@Rusher Criteria (from tag wiki): For code-trolling the winning criteria would normally be popularity-contest, but this winning criteria will not be implied, so you must manually add it, be it popularity-contest or something else.
@Rusher If that were the case, every tag except the winning criteria would disappear. Since it's been established here that all questions must have one of those, no other tags could be used alone.
So I shouldn't ever use the [ascii-art] [math] [string], etc, etc tags? They can't stand alone here, yet are some of the most popular tags on the site.
The specific wording of that rule kinda falls apart when you have a required subset of tags, though. All tags outside that subset would have to be burninated, and that's clearly not the intent.
Chris, how is it that you write everything in a way that is reasonable and understandable even when you are completely disagreeing with me? You are my favorite mod.
@Rusher Why, thank you. I don't want to be inflammatory, so when I disagree with a stance, I try to explain why in a way that allows people to come to a useful middle ground.
I found this post from Jeff Atwood and I understand what you mean now. "I’m less worried about the ragged edges of the system; it’s when we have three GRATUITOUSLY meta-tags hogging up space in the top 10, top 30, top 40 tags that I start to get worried.
So we don’t have to stamp out every iota of meta-tags, we just have to DISCOURAGE it, and make sure it doesn’t come to DOMINATE the top (n) tags list — as it did on Stack Overflow."
Let's invent tag-bowling: How many tags can you add to your question without becoming contradictory or being closed as too broad. do not take this proposal seriously
Omg... I've been working at this place for months now and been wondering why "output from debug" in visual studio would randomly print "true". I just found this buried like 5 levels deep Console.Write(true);
It serves no purpose, gets called multiple times on every page load, and has been bugging me for ages
@Rusher Like I said, deeper problems, lol. That sounds like a horrible design, but I honestly wouldn't be surprised to see it in some places I've worked.
@Rusher Remembers me of a case where we had to print white font color on white paper. Otherwise the printer driver would eat up empty pages and the client wanted to have empty paper on the printer.
Ok, I just asked the guy who wrote it (because I too can annotate), and he said that he wanted to convince himself that the function was making it all the way through without stopping to debug it over and over, and then he forgot to remove it.
@ChrisJester-Young So true. My last project, there was enough disagreement we had to do a formal review of logging frameworks before picking one. Did that so no-one would wind up insulted ...
@Rusher lucky (maybe?). We had an issue with logging framework only b/c the development team we brought on board all had their own "fav" without any strong reasoning behind that favorite. Chaos!
Since you are all programmers here, I know some of you must get the feeling that I get sometimes that if we worked together, we could build something "cool". But then you despair because you can't think of anything cool. That's why I strongly believe that we should all play Minecraft
As soon as Diablo 3 gets old, I'm gonna set up a server and invite everyone.
@Rusher, I don't see the logical connection between "We could build something 'cool' if we could think of something" and "We should all play Minecraft".
I had a bunch of colleagues all on the same server, playing actively -- we have yet to build something cool together. The closest is only cool to math-heads -- we built a system of straight-line tunnels at level 11 on multiples of 256.
@Geobits You are wise to do so, when I first started playing MC I spent all my spare time on it. And then some. Life conspired to pull me away and I've been avoiding getting back in since, focusing more on programming in my spare time.
@Rusher Apparently squeamish ossifrage, ProgrammerDan, ProgramFOX, Jan Dvorak, and Vereos are some of the people on the other side of the code-trolling opinion divide.
People who voted to reopen the code-trolling question troll
err, the recent meta-troll's code-trolling question
I disagreed with the reason it was closed, I'm not particularly in support of code-trolling but quite a few people answered and many of those answers where clever.
Is there any tag for challenges where the programs 'battle' against each other and the highest score wins, but it's not one-to-one like 2048 king on the hill?
problem is scoreable, so the winning solution can be objectively determined, but it doesn't appear covered by other tags -- seems like the definition of the code-challenge tag: A code challenge is a competition for creative ways to solve a programming puzzle with an objective winning criterion not covered by other scoring tags (e.g. code-golf).
@ProgramFOX If you look at meta.codegolf.stackexchange.com/a/1309/18487 starting around comment 7 or 8, we had a short discussion about the tag being used for multiplayer challenges
I think the one-on-one concept should be removed from that tag
King-of-the-Hill everywhere else either means "free-for-all with one guy on top" or "That cartoon where people drink a lot of beer"
@ProgramFOX it's almost a ... race of sorts. Every answer puts in its best time independently, and the winner is the person who satisfies the objective win criteria "more" -- e.g. shortest time, lowest/highest score based on a metric, etc.
@Rusher Good idea. Perhaps we could change it into: "King-of-the-hill indicates a game where the submissions battle against each other, either through a one-on-one game played in a round-robin all-against-all or through scoring."
The submissions for the new 2048 question would not directly battle against each other, e.g. they do not interact. They merely attempt the same feat, and their scores are compared. It's more like most of the events in the Olympics.
@Rusher Other side of divide would be the code-trolling and popularity-contest tag wikis (all objective criteria met), the people that voted to reopen the question (it is open again), everybody that answered the question (10 so far) and in some sense the mod who stated, "both [popularity-contest] and [code-trolling] are being tolerated".
If 10 answers are being objectively measured and then compared, that doesn't make it a KotH challenge. Otherwise, every code-golf question would also be HotH
@Rusher Okay, so if we assume that as a basis for definition, then King of the hill should retain language to the effect of "battle", and the new 2048 would be outside that tag, for sure.
@Howard Same goes for other KotH challenges, though. I can't prove Gareth's results in the BattleBot tournament, and there were several bots that took random actions.
E.g. the actually-not-that-interesting 2048 question (pick the best from stackoverflow.com/questions/22342854/…, port to JS) and about 95% of every over-specified code-golf?
@Howard Right. Perhaps the "more fair" approach for the OP would be to run each solver multiple times, and take an average of the outcomes for each solver.
@Howard Of course will not be deterministic, that's not achievable, but would be more "fair".
I have a KotH challenge sandboxed that is not deterministic, but I wrote at the end something like "You will be scored on your survival rate over three trials that each have 1000 iterations." That's still somewhat random, but statistically the best submission will always win
Since the 2048 programs are being run for an hour(and will presumably restart if the game ends), don't you effectively get an average from multiple runs already?
I don't know if anything is wrong with it (besides what's generally wrong with that question), but one effect it does have is that it means algorithms that, say, fail 99% of the time but get a massive high score otherwise beat out algorithms that, e.g., win the game 100% of the time although not necessarily with the highest score.
I dunno if that's good or bad, I'm just pointing it out as a quality of the criteria there.
That's true. I think one of the key points is that you can only move once every 250ms. That's a hard cap of 14.4k moves in an hour, which puts a damper on "try it randomly a thousand times and pick the highest score."
What he should do is perhaps seed the random number generator the same for every person's test run, and also do a hard count of 14.4k moves to avoid variations in total count due to timing inaccuracies.
@ChrisJester-Young KotH should be about programs battling eachother, "as far as you understand the concept"? You tell us what it's about as per the site's philosophies. That's precisely what I was talking about. It's a circus, enforcing "objective criteria" only when convenient, and wishy-washy subjective interpretations of rules otherwise.
@Rusher Rich B was an early troll on Stack Overflow who liked to skirt the very, very edge of the site's policies to make a point.
He was frequently suspended; last I heard, he received a one-year suspension for his "follow the letter of the law but break its spirit in every way" behaviour.
So considering my one specific question that I gave as an example.
That question does fit 100% into the concrete criteria described for both code-trolling and popularity-contest questions. It simply does; there's no debate there. If you go through those lists, it hits all the do's and avoids all the don't's.
That question had two "issues" that I can think of. 1. There are people who reasonably dislike popularity-contests / code-trolling. 2. Some people took offense to the link I made back to the original SO question.
#2 was an honest mistake. I gave the OP a good answer on SO, and he also seemed to have good spirits about the PCG post. But that is beside the point. I can understand removing that link. However, the real approach here should have been to post a comment on the question that said "please remove the link", or to simply edit the question. Not silently down vote or vote to close, that does nothing.
#1 Is a reason to go to meta or chat and continue / start a debate about the appropriateness of code-trolling / popularity-contest. It is also an (arguably) legitimate reason to down vote the question, I suppose. However, it is not a legitimate close reason, given the site's currently stated criteria for questions that belong here.
1. Clearly there are plenty of users who want to close the question, as much as there are users who want to reopen. So there is certainly debate about whether the question is a good fit for the site.
2. The guidelines in our community are more descriptive than prescriptive. Sure, perhaps reopening a dialogue about whether code trolling should be encouraged is a meta post someone should write. dmckee already said he wanted to, but lacked time to.
@JasonC I had the same problem the other day. A question was closed because it basically looked like a homework problem. Guess what I did? I started a meta post that didn't begin with "PCG is a Joke."
@Rusher That's fine, I will concede it was a bad title if you will agree to focus more on the meat of it.
@Chris Now hold on, though, I'm not quite done.
So, @Chris, the thing is: When multiple users ask you to close the question, your first responsibility as a minimal yet objective moderator, called in to bring order to chaos, is to say "While this question may end up not being a good fit, it currently is a good fit. Please stop stewing over the question, and let's have a discussion on changing our posted site policies on meta so this kind of argument doesn't happen again in the future."
@Chris Additionally, when you see that no constructive or explanatory comments have been left on the question, to me that seems to be a red flag that the close votes are more of an "I'm angry at these questions in general" -- a valid reason for a discussion, an invalid reason to cause confusion by closing questions that are within currently stated bounds.
@Chris As a moderator you want to stay minimal, that's SE's brilliance, but if you don't bring guidance to these kinds of debates, they fester and stew and end up leading to e.g. uncontrolled close reason abuse, angry meta posts, etc. You have to be minimal but there is still some moderation that has to be done.
@Chris And yet, it is a responsibility that every other moderator on every other SE site I've seen seems to handle quite well, but PCG is lacking entirely in that area. That is why I said PCG is acting differently than other SE sites.
@Chris Well the parliament can't get its sh*t together, so step in!
And by the way, my intention isn't to personally attack moderators, and it's also not to point fingers and say "you are a stinky moderator!" That's not what I mean at all.
This site has 3 moderators (so far). The fact that the other mods, and most of the other users, haven't jumped on your bandwagon tells me that you set the line (at which to take mod action) somewhat differently from the community in general.
My intention is that the site is fundamentally so lacking in clear direction, and the moderators need to set that direction. Nothing more. No massive involvement. Just point us firmly in the right direction. That's all. Until the mods here can do that, we're just wandering. If that makes sense?
It's good to have this discussion about where to draw the line, and that's why I appreciate your post. But I do not have to agree with your placement of that line.
The gold standard for whether a question is a good fit isn't tag wikis but the help pages, and in particular the per-site customisable one. Note in particular, "All questions on this site, whether a programming puzzle or a code golf, should have…a clear specification of what constitutes a correct submission."
A quick look at the deleted answers to your question casts doubt on your question in particular, and I am quite willing to argue that code-trolling in general fails that test.
@Chris The "most of the other users" is not a good criteria, for that you would have to also show that "most of the other users" actually pay attention on meta.
I think, given that my response to your post has more upvotes than your post itself, that the general "tone" of the community is that my view is more congruent with the view of the community than yours is.
@PeterTaylor The code-trolling wiki's take on the "objective-criteria" thing is that it should clearly be labelled as a "popularity-contest" - at least that's current policy. The "popularity-contest" object criteria is vote count.
@PeterTaylor Sorry to derail, but it looks like the "per-site customizable page" needs to be updated -- Help may be available on the Puzzle Lab chat or the meta Sandbox, and the use of these resources is encouraged -- both links in that line are no longer valid :(.
@Chris The fact that you cite vote counts on meta as evidence of "rightness" is disturbing, and for that I suggest you ask some of the mods on other sites, SO in particular, what their opinion on that particular criteria is.
@Chris Be fair, though. It's not "the community", it's "the portion of the community that reads meta", which is somewhat biased on all topics (ones I agree with or not).
@PeterTaylor That may be the case; but the current wording makes it very easy to make that mistake. Perhaps either the wiki or the help center should be clarified?
@Chris Yes I am aware of what voting does in the unrelated political constructs you are referring to.
@Chris Based on your ideas about the significance and use of meta, I am not sure that I have anything more to add here. Anyways thanks for the conversation, you raise good points, I hope you see my points too and find the time to reflect on this a little. I have to get back to work.
I keep meaning to post a meta-question on the formal status of [popularity-contest] and [code-trolling], but I am very busy in "real" life right now (and for the foreseeable future). I suspect that you are right about not being able to turn back the clock, and the extra traffic may be worth it but we desperately need some kind of quality control in those tags. Actually, I think that [code-trolling] question are bad almost by definition---the whole point is that they are lazy question. — dmckee ♦4 hours ago
Ah, well in that case I don't think I've been around the site for long enough (I'm talking about 2/3 years ago) to be able to judge the "formal status" of that.
I am eating a sandwich and typing at the same time
@Chris Quick point; I was thinking on the way to work just now: The political analogies for moderators aren't the best here. A better analogy is a spotter at a gym. Minimal involvement but still enough to keep the weight lifter safe. Take a moment to visualize a good spotter. Now as for which of users / PCG / philosophies map to weight lifter / weights, that's an exercise for the reader.
@ProgrammerDan In any case, my answer will come pretty far after you already have the bounty. When I'm done I'll port it to Python and beat you. Almost all of my answers start in Java and end up in Python lol
@Rusher Haha, yes, I believe it. I keep meaning to use this forum as an excuse to improve my python, but there's always a "speed" component to writing answers, and at least for now I'm faster in Java
There are a lot of ways python would make shorter code for my solution, however. Lots of array accesses that could make heavy use of [:] constructs :)
PPCG in three acts
Act 1: Happy Times!
2011 - Nov. 2013
Not much traffic on the site, but our small community is happy and thriving, and when a new question is posted it's almost always high-quality and enjoyable.
Act 2: The Dark Era
Dec. 2013 - Jan. 2014
Systematic invasion and sneak a...
I need to convert shortened versions of month names to its longer representation (e.g "Dec" -> "December"). The months name is a String and I would rather not convert it to a date object first.
Is there an easy way to do this?
Edit:
My problem was associated with inserting "Dec" as a table name...
