17

Background. SE has periodically pressed forward campaigns trying to make the site friendlier to new users. Summer of Love (several editions), Be-Nice-policy,... Apparently they have largely failed. At least Sara Chipps seems to think so. The latest attempt was to increase rep rewards for questions. While a nice try, I doubt it will do much, because it does nothing to the root causes. I, for one, will continue to insist that the new user searches the site for duplicates before posting, and shows a due amount of effort.

My recovery from the events that lead to this outburst made me realize that I really don't have so much against new users. I quite understand that they have difficulties coming up to speed with what is expected. I have been teaching college math for 30 years, and I know not to stoke newbies math anxieties. Mind you, we actually get enough new users at Math.SE who actually bother to research their questions (I am willing to dismiss homework help seekers, I will educate them on my paid time).

I want to draw attention to a couple of comments:

Shog9: "A big part of solving that comes from recognizing that these sites are a shared resource, and working from that toward some sort of compromise as to how the site should be used."

Shog9: "No amount of pleading or nicely-worded signs are going to convince water to wet your parched plants when it wants to tear out a gully and carry away your precious topsoil."

So why I insist on the strict rules?

At the dawn of the SE network users started developing different ideas about how the site should be best used. Some emphasized quality, organization and generality of answers, others decided that speed and volume (of their own posts) are more important. The notorious 4 camps are just isolated points on a close to continuous spectrum. As explained there camps 1 and 3 have all the moderation power:

  • camp 1 = users emphasizing content quality and organization, aka caretakers,
  • camp 3 = users emphasizing content volume, I call them Sandmen from the sand vs. pearls dichotomy for the alternative terms are not conducive to dialogue.

The software imposed rules on voting heavily favor camp 3.

  • No matter how many duplicates such a user answers per day, there are no consequences to their ability to answer.
  • Even if camp 1 deletes 10 dupe answers to FAQs by the same user per day, that user can post many more - to reach the rep cap, or whatever other reason.
  • If I downvote all the FAQ/dupe answers of such a user, the system will revert those votes as targeted downvotes.

Let me spell it out for the benefit of those developing the site SW:

No amount of pleading or nicely-worded signs will convince me not to vote to delete unpolished questions from new users if letting those questions survive will give fodder for [term redacted].

Why dealing with this camp 1 vs. camp 3 is important in relation to newbies?

Right now, a few chatrooms in Math.SE are a war zone. I don't think that the relations between these two loosely defined camps are amicable elsewhere in the network either. The war consumes a lot of resources, and probably many would prefer a lean compromise to a continuation of this bloody war.

The two sides have dug deep into their respective trenches. They cannot talk it over in meta for the camp 3s never come there. Or, they only show up to complain about an occasional deletion, but won't change their behavior. Even if some adjust their answering policies, the war will continue as new answerers join their ranks.

In the eyes of caretakers disenfranchised newbies are simply unavoidable collateral damage when fighting the true enemy of the site. Mind you, we will post nicely-worded comments when voting, but the newbies will still get their questions closed/deleted.

What I think should be done.

[Edit: I deleted a suggestion for a newbie tag, albeit with the extra piece of answers not being rewarded generously or at all. Apparently that has been discussed to death already. Concentrating on the (over)eager answerers for now.]

  • If I were a dictator the wrongfully earned rep from answering FAQ/dupes would be gone, but... A) that's near impossible to do fairly and algorithmically, B) I said that we need a lean compromise, so the new rules should probably apply in the future only. Suggestions below.
  • If you answer a duplicate, or your answer gets deleted because the question was deleted as poor, you get a speeding ticket limiting you to 6 answers per day and 50 per month for the next month (the recent change made the rules more symmetric between the askers and the answerers, and this would be in the same spirit).
  • The rep lost on the deletion of an answer is deducted from the daily rep cap of the day the answer was posted (remove the possibility to "prepare" for deletions by answering more than is needed to reach the rep cap).

I think these would be very mild consequences. With 98% of the questions already answered, being limited to 50 answers per month should not really hamper anyone who pays attention. But these SW limitations would show that the software is not all in favor of the camp 3 lifestyle, and would be taken as a sign that the SE is not picking the side of camp 3 (more about that in another thread). Furthermore, these rules would steer the users away from trying to compensate lack of quality by quantity.

After all, Shog9 called for a compromise. Policies where camp 3 carries on unchecked are not compromises.

I didn't make the rules of this game, but I would very much like a level playing field in relation to the camp 3s.

===

Edit: Several points have been raised (some I foresaw but left out because this post is long enough already). A main objection being that accidentally answering a duplicate should not have severe consequences. I don't think being limited to 6 answers per day + 50 per month is severe (it would simply steer the answerers towards making their allotted posts better). But how about a "three strikes" -rule with warnings after strikes 1 and 2. Similar to the messages we receive from the review audit engine!

Anything that signals to these answerers that they should do their part in organizing the site, and not simply merrily pick low hanging fruits.

Acknowledging the difficulty of coming up with a rule limiting a practice that is tolerable when accidental, but disturbing and unfair when done excessively. Suspensions are IMHO too harsh a solution. That's why I suggest limitations on the number of posts.

  • 21
    You had me at the beginning, but your proposed solution isn't something I would support. Making any sort of distinction between new and established users carries the implication that there's a difference that needs catering to. And will result in a jarring feeling when we take the training wheels off. Crashing at speed, if you will, is far worse than crashing when you first get going. – fbueckert Dec 6 '19 at 17:24
  • 7
    "Gamification?" Well, gamification at least generated those thousands, evil ole veterans (SE rep "junkies") who keep the communities running at a halfway acceptable quality level of a Q&A oriented research repository. And suddenly it goes: "Forget about what you think you know how SE works. We're going to get rid of "these ghosts we called" to make our company that big. – πάντα ῥεῖ Dec 6 '19 at 17:49
  • 10
    ... If a new user want's to play and participate the game in real (and not a simulation sandbox), they should simply inform themselves before they start. I believe there are enough hints and informations available. The problem is that people don't read or research,, but they want an answer for their poorly stated problem. – πάντα ῥεῖ Dec 6 '19 at 17:55
  • 1
    Continued "gnatting" after the Q edit: meta.stackexchange.com/questions/328609/answering-duplicates – πάντα ῥεῖ Dec 6 '19 at 19:13
  • 1
    That single user you mentioned who wrote those 2000 FAQ answers, that sounds like they would probably make up a significant fraction of such traffic on math se then? Is it really not possible to flag that person, and have the moderators sanction them? If that user continues to do that, why aren't there millions of flags on their content, and the user gets suspended?! – GhostCat salutes Monica C. Dec 7 '19 at 15:36
  • 2
    @RouninsaysJesuisMonica I see. The "feature" I most dislike is users posting minor variants of same material, day after day, sometimes many times on the same day. And its low level, too. Math equivalent of debugging Hello World and bubble sort. And these people get all the rep. Hundreds of thousands of it. Without adding anything worthwhile to the site. I will vote to close those threads at the slightest excuse to stop the repfarming. Which brings us to the real question: "Where is that lean compromise?" – Jyrki - No Shog9 - No SE Dec 7 '19 at 17:19
  • 3
    I don't disagree at all with what you're saying. Just, I feel the "minor variant" questions ought be moved from their own "dead and closed page" to an accordion section (ie. not initially visible) on the page where the original question appears. Arguably, a dead and closed page which isn't accepting answers doesn't benefit anyone. – Rounin Dec 7 '19 at 18:51
  • 3
    I see, @RouninsaysJesuisMonica. The idea of an accordion is new to me. Might work, and would definitely improve the site – Jyrki - No Shog9 - No SE Dec 7 '19 at 19:02
  • 5
    We can be strict without being unwelcome. Why are some people arguing that being welcoming means forgetting about quality control? We are not being asked to ignore the flag queue and stop flagging and voting, but to be nicer and more helpful. – Von Huffman Dec 8 '19 at 2:26
  • 2
    At math.SE, it's irksome watching an expert (e.g. someone with a PhD) endeavor to interact with a user to improve their question (typically a PSQ), and then someone would obliviously post an answer after which the OP ignores everything. (It reminds me of the Simpsons: "Oh, there's something unsatisfying about scrubbing these rocks and I think I know what it is." [a wave washes a new coat of oil on the once-clean rocks]) Once I mentioned math.SE to my colleagues (in real life), they reacted as if I were helping students cheat on assignments. I lost much interest afterwards. – Rebecca J. Stones Dec 8 '19 at 6:20
  • 2
    @I am not the way you speak, @Rounin says Je suis Monica: That is why a help desk-like place could work. There are many people on Stack Overflow whose primary motivation is helping people (who don't care about reputation points or building a knowledge repository). Stack Overflow could be split into two: One for a repository and one for a help desk. The gamification, if any, could be vastly different on the two. (By extension, perhaps also for other Stack Exchange sites.) – P.Mort. - forgot Clay Shirky_q Dec 8 '19 at 6:52
  • 1
    @PeterMortensen I would welcome/support such a division in Math.SE also. Warmly. – Jyrki - No Shog9 - No SE Dec 8 '19 at 7:00
  • 5
    I would have loved to have seen this question appear a year ago. The past few months have made it exceedingly clear that SO is catering solely to camp 3; maintaining quality is just a means to attract more traffic, not an end in itself. And maintaning quality is a lot less effective at that than instant gratification through quick and easy answers and internet points. And encouraging others to learn has no place in their model at all. – Servaes Dec 9 '19 at 10:50
29

A system like this presumes that duplicate closure is flawless.

And my God, it is far from that.

You're basically punishing people for the fact that the biggest broken thing in all of this is that no one could reliably search for a duplicate question.

Can we just fix broken search instead of shooting someone who may be acting in good faith?

|improve this answer|||||
  • 18
    One dupe answer is an accident. Two dupe answers is an accident. After we have exposed 2000 answers by the same user to be to FAQs you still assume good faith? I realize that good faith (or lack thereof) is difficult to detect algorithmically. But, the proposal is not an on/off thing. The suggested consequences are VERY MILD. – Jyrki - No Shog9 - No SE Dec 7 '19 at 4:15
  • 5
    And users posting an answer in under 5 minutes certainly did not even attempt to search. Anyway, thanks for bringing up good faith here. It is one of the key factors. – Jyrki - No Shog9 - No SE Dec 7 '19 at 4:27
  • 3
    +1 –– If duplicate questions were as easy to find as some seem to think, the system could automatically point out to an asker that their question was a duplicate and automatically disallow that they post it. In my experience, those who mark a question as duplicate happen to know of the previous similar question and search for it based on what they remember of it. – I am not the way you speak Dec 7 '19 at 8:03
  • 4
    @Iamnotthewayyouspeak Finding duplicates is not easy, but that's not the point. I don't know about StackOverflow but I know about Math.SE and given that SO is older, I suspect it is very much the same: we are talking about questions that anyone who has spent a year or more on the site can tell at a glance that There is snowball's chance in hell that this is NOT a duplicate. – Jyrki - No Shog9 - No SE Dec 7 '19 at 8:25
  • 1
    @JyrkiLahtonen Newbies have not spent a year or more on the site. Why would you want to punish them for failing at the impossible? –– Oh, and by the way, this here question of yours is a duplicate ;-) – I am not the way you speak Dec 7 '19 at 8:36
  • 7
    @Iamnotthewayyouspeak I am not out to punish the newbies. It's the repmongers I'm after. They are the ones I want to kick. – Jyrki - No Shog9 - No SE Dec 7 '19 at 8:42
  • 2
    And, the irony of this being a duplicate did not escape me. The question started out as a call for discussion on the theme that the two major problems of A) how to be welcoming to new users, B) how to bring about peace between camps 1 and 3, are connected. And a proposed solution to problem A that amounts to camp 1 unconditionally surrendering to camp 3 is not acceptable. – Jyrki - No Shog9 - No SE Dec 7 '19 at 8:46
  • The need to have a specific proposal then forced me to edit out the constructive parts of dealing with problem A, and concentrate on B. That is a shortcoming of the meta format - it is difficult to discuss two intertwined problems in one go. – Jyrki - No Shog9 - No SE Dec 7 '19 at 8:48
  • 4
    I remember a user not too long ago asking a question that was almost an uncanny duplicate of a question I had read weeks prior. I tried for half an hour to find that other question, as it contained the answer too. I couldn't find it. – MechMK1 Dec 7 '19 at 13:45
  • 1
    I don't think improved search is sufficient (though certainly welcome). It can be surprisingly hard to find a duplicate, even if you already know it exists. – P.Mort. - forgot Clay Shirky_q Dec 8 '19 at 7:05
  • @JyrkiLahtonen: I can appreciate your zeal for wanting to rein in those who would blithely answer duplicates over and over again, but I maintain that there's still a problem at the application level which makes this more convenient. We should be prepared to assume good faith on behalf of the answerers and sew up any loose ends with duplicate search so that this problem manifests less. You're always going to have people who want to answer the low-hanging fruit, which is a symptom of the larger problem of not exposing the existing answers in a more accessible fashion. – Makoto Dec 12 '19 at 17:06
  • @Makoto: Yes, there will always be such people. I just want to curb it not being the same selected few people dozens of times every day. That is why I suggested very mild restrictions. Such that errors can be tolerated. – Jyrki - No Shog9 - No SE Dec 13 '19 at 8:20
17
+50

I believe that part of the problem that SE has is a contradiction between two goals. On the one hand, SE intend their network of sites to become a repository of canonical questions and answers. That is, they want the sites to become a reference work or encyclopedia. On the other hand, they want the sites to be a place where anyone can ask a question and get help.

The first goal (the "encyclopedic goal") requires that answerers are experts and that duplicate or otherwise "low quality" questions are prevented or pruned. The second goal (the "providing help goal") requires that the sites are open for anyone and that anyone may ask anything (as long as it is within the scope of the specific site).

The first goal requires that the community has a high participation threshold; the second goal requires that the community is open and welcoming. Both goals are not easily possible at the same time. In other words, the basic problem is conflict by design.

|improve this answer|||||
  • 2
    A possibility that appeals to me would be to move "tutoring" to chatrooms, and keep the Q&A are well organized. As another veteran user put it "Confused newbies are confused in their own unique ways". There is nothing of permanent value in storing the cure to each and every individual confusion. – Jyrki - No Shog9 - No SE Dec 7 '19 at 8:38
  • 1
    When I don't find a precise enough duplicate to a minor variant of a FAQ from a new user, I try and engage them in a socratic dialogue of hints, pointers, feedback etc. With a goal of making them answer their question themself. After the newbie has posted a question, the comments are the only place for such a dialogue. But a chatroom would be better. – Jyrki - No Shog9 - No SE Dec 7 '19 at 8:41
  • ^(two comments up) "... keep the Q&A area well organized". Apologies for lack of proofreading. – Jyrki - No Shog9 - No SE Dec 7 '19 at 8:49
  • Then you need 2 areas, one that has any and all questions no matter how duplicatory or poor, and if they pass some sort of validity test, they get moved to the site proper as Q&A references. Users never see their Qs deleted, they may get answers saying "look here, does this solve your problem" as an answer (which is valid TBH, links to other sites are always valid), so they will never appear to be pushed aside. – gbjbaanb Dec 7 '19 at 23:37
  • 3
    Do you honestly think we can't do better in the being nice department, without affecting quality? Is it really impossible to improve the current new user experience without turning this into google answers? I feel like a lot of us are approaching this through a pretty severe false dichotomy. I don't see the dilema. Some people are full-on disrespectful, patronizing, rude, non-constructive (etc) while dealing with new users. We can't do better without turning into quora? That doesn't make sense. – Von Huffman Dec 8 '19 at 2:33
  • 3
    @Lyd Sure we can. And I strive to do just that. But the need for speed in putting questions on hold is overwhelming. And also THE BURDEN IS THEN MISPLACED. The people wanting to answer the umpteenth duplicate should be the ones who guide the asker. But, no, they want to show that "hey, look, I can do this piece of freshman homework, can I have upvotes?" – Jyrki - No Shog9 - No SE Dec 9 '19 at 5:41
11

As the "sandmen" are going to get rep, what about "Give an incentive for finding duplicate questions" along with a way they can explain to the questioner why it is a duplicate.

|improve this answer|||||
  • 6
    +1. Very sensible. Make finding duplicates of questions more rewarding than answering dupes. Prevent users from flagging everything as a dupe (no matter how tenuous) by requiring the flag be verified by a high-rep user. When it is, both the flagger and the verifier receive rep. – Rounin Dec 7 '19 at 15:01
  • 2
    @RouninsaysJesuisMonica Maybe just awarding 15 rep if the duplicate is comfirm as solving the problem by the person who asked the question will be enough. – Ian Ringrose Dec 7 '19 at 18:01
  • 1
    @RouninsaysJesuisMonica I don't think the verifier should get a reward. As it gives an insentives without any quality control on what is being done. – Ian Ringrose Dec 7 '19 at 18:03
  • Yep. I agree with both your points. – Rounin Dec 7 '19 at 19:08
  • Hmm. May be the reward should depend on how good the question is? May be the rep reward should come from the bag of rep earned by the answers to the dupe, to be sure that answering a dupe is less rewarding. – Jyrki - No Shog9 - No SE Dec 9 '19 at 5:45
  • 1
    This might have the effect of turning the site activity into one of dupe finding frenzy. While a nice idea (I upvoted), I'm not sure I cherish the prospect of "the race to answer" getting replaced with "the race fo find a dupe". Neither really measures the skill and expertise of the participant. – Jyrki - No Shog9 - No SE Dec 9 '19 at 5:54
10

The main problem for Stack Exchange is that their platforms are unwelcoming by design. When new joiners ask a question, they often get dupe hammered, heavily downvoted and/or their question gets closed because of low quality. After asking their first question, their first experience is generally very negative. This has been discussed in many blogs and Reddit threads that I shall not reference here.

The real problem is that this is by design to ensure Stack Exchange sites remain a canonical question and answer site, rather turning into Quora where anything goes. Stack Exchange management basically need to make a decision what direction they want to go in. Being more welcoming will likely compromise the overall quality of questions and answers, however, if they don't become more welcoming, they may not get enough advertisement revenue to survive.

|improve this answer|||||
  • The earliest versions of my question were trying to discuss "under what conditions could I do my part of the compromise and not vote to close poorly researched questions". The spirit was: "I see the need for a compromise here, and could get behind one. But only if the grand compromise also affects camp3". I edited those out, because the question was criticized for being unclear, and my suggested way of helping new users was old hat. – Jyrki - No Shog9 - No SE Dec 7 '19 at 4:22
  • 2
    Having seen a lot of different ways and approaches that the SE community chooses to communicate rules and guidelines to new users, I think we can do a lot better. As I see it, it's less about completely ignoring quality controls, and more about implementing them in a more friendly way. I don't think new users are surprised by the house rules, or quality standards, but by how they are treated during this "discovery" phase. – Von Huffman Dec 8 '19 at 2:21
  • How do you know how mych revenue SE gets and how much it needs? – Renan Dec 8 '19 at 7:41
  • @Renan I don't know how much revenue SE receives, however, as a commercial company, they need to make sufficient revenue to please their shareholders. It could well be that the advertisement revenue is sufficient, however, if not, they maybe looking for alternative revenue sources. – Alex Dec 11 '19 at 19:40
4

Update: I agree, situations where you have blatant "rep only" hunters that write 20+ low quality answers on low quality questions, that is a nasty issue, and the rules outlined above might help for that case.

But: not every situation is that clear. Grey areas exist, and I see plenty of potential that the proposed rules also hurt other people:

If you answer a duplicate, or your answer gets deleted because the question was deleted as poor, you get a speeding ticket limiting you to 6 answers per day and 50 per month for the next month (the recent change made the rules more symmetric between the askers and the answerers, and this would be in the same spirit).

Problems with that:

  • Remember that the relatively new Lifejacket and Lifeboat badges suggest the exact opposite. The badges encourage users to spend time to write up high quality answers, even on questions that aren't too great. And also to invest time to help improving the question quality. The above proposal basically puts a distinct "risk" on people writing answers. An answerer might start with "others just don't see that this is a reasonable question, and I will prove them wrong by writing a great answer". Maybe the answer is really great, but the question stays low, and eventually gets closed and deleted. And now the person who actually did something useful gets kicked for doing so?!
  • Then: just this week, the vote count for closure has been reduced from 5 to 3. I personally like that, because quickly putting questions on hold gives feedback and suggestions to the OP asking. That is much better than having them look at -2, -3 votes, for hours, and "nothing else" happening. But just 3 votes also means: it is easier to get the vote wrong. There are not only people hunting reputation by answering all kinds of nonsense, there are also robot reviewers coming to wrong conclusions too often. Now: the sanctions you propose when answering "in the wrong place" will lead to pushback. When 3 votes are good to close, then it only takes 3 well connected camp3-members working together to undo that close, to prevent deletion.
  • Leading to: sanctions are nice, but what if a question gets in fact undeleted, re-opened. Maybe the same day, or worse, a few days later. So not only do you increase the potential for people fighting on close/re-open/close/... harder, you also put a sanction in place that can't be "undone" easily (from the perspective of folks who try balance camp1 with camp3 desires).

The rep lost on the deletion of an answer is deducted from the daily rep cap of the day the answer was posted (remove the possibility to "prepare" for deletions by answering more than is needed to reach the rep cap).

Nobody can know upfront "what will be needed" to reach the daily cap limit. Beyond that, that cap limit is in itself loaded with emotions, and adding "more rules" and sanctions has a high risk of conflict, for relatively few gain.

And remember: in the end, you need at least 20 individual upvotes per day to get to the cap. That means that often 20 different individuals found your content worth upvoting. Now, if one happens to add an answer in the wrong place, is it really helpful to turn the reputation gained there into a sanction?! Remember: you talk about answers that probably saw multiple votes, so maybe: they weren't that bad in the first place.

Finally, just for the record: I occasionally comment on "hmm, well okay" answers on bad questions, and explain to the answerer that he might A) send the wrong signal to the OP asking the bad question, but that B) most likely, the answer will be deleted together with the question anyhow (so: waste of time). You now what comes back pretty often: "I don't care about the reputation. I want to help people, even when they struggle to put up a great question". Okay, then their answer gets deleted ... but would you really want to extra kick the people trying to help? Do you really think this will overall improve things?

|improve this answer|||||
  • 3
    Answering the last sentence. Yes, it would improve things. You apparently don't appreciate how wide the chasm between the opposing camps has grown. The rest of your answer has the air of tinted glasses. There are no nice answers to the 20th duplicate. And that nice answer, should there be something new to add, could be posted to the older version equally well. – Jyrki - No Shog9 - No SE Dec 6 '19 at 20:14
  • 1
    We are talking about the cases where the answerer just has an overwhelming desire to post... anything. The time stamp of their answer tells that they did no searching, the first answer is just a sketch. This is the reality caretakers try and cope with. Nothing of the lifejacket sort. – Jyrki - No Shog9 - No SE Dec 6 '19 at 20:17
  • 1
    @JyrkiLahtonen Probably there are differences ... in different communities. Talking about programming questions on SO, there are plenty of situations, where yes, some DUP exists, but where it is also possible to address specific "special" aspects in the question asked. Meaning: there can be a lot of areas where subtle differences between all those 20 DUPs exist. – GhostCat salutes Monica C. Dec 6 '19 at 20:19
  • More tinted glass. If I try and convert a math situation to programming, we have users who would insist that a question about bubble-sorting an arrray of rectangles according to increasing width would NOT be a duplicate of another where they are sorted according to decreasing width, or increasing height. – Jyrki - No Shog9 - No SE Dec 6 '19 at 20:22
  • Anyway, if caretakers cannot get any help from software, I will need to redouble my efforts in getting unworthy content deleted. – Jyrki - No Shog9 - No SE Dec 6 '19 at 20:23
  • 2
    @JyrkiLahtonen the community getting help on such requests from SE Inc. got harder and harder over the last years anyway, and most likely: even harder by now (but that is a different story which I didnt mention in the answer on purpose). – GhostCat salutes Monica C. Dec 6 '19 at 20:25
  • 2
    IMO there should be no reward to FAQ answers for they don't add any knowledge to the pool we are to build. – Jyrki - No Shog9 - No SE Dec 6 '19 at 20:31
  • 3
    What are the 3 of us to do to a user who willfully answers 20 or more duplicates per day. If we downvote all those, the system will revert our downvotes as being targeted. – Jyrki - No Shog9 - No SE Dec 6 '19 at 20:33
  • 2
    Anyway, there are consequences to asking bad questions, why are you so adamant about there not being consequences to answering bad questions. – Jyrki - No Shog9 - No SE Dec 6 '19 at 20:35
  • 5
    Anyway, thanks for your time. You see, I'm dangerously close to the point were I begin to prefer no StackExchange at all to a StackExchange with repmongering running rampant. – Jyrki - No Shog9 - No SE Dec 6 '19 at 20:37
  • 1
    Like I said in the deleted part of the question, I would happily allow and answer questions from noobs, if and only if I know that no rep is on the line. – Jyrki - No Shog9 - No SE Dec 6 '19 at 20:39
  • 1
    @JyrkiLahtonen Really last comment for today: maybe the real point is: it really doesn't make sense about somehow highly technical rules ... when the underlying social problem/conflict isn't addressed. In the sense of: yes, "camp3" isn't declared as totally unwanted, behavior. A lot of our documentation emphasizes "trying to answer" is a valuable thing. You would really first have to convince the majority (and SE Inc.) that "no, we ONLY want good, almost great stuff, and everything else shall be kicked out". When you have that consent, then you can talk about rules and stuff. imho. – GhostCat salutes Monica C. Dec 6 '19 at 20:41
  • 2
    I'm not asking the camp 3 to be booted out. I am asking for some kind of a balanced treatment as opposed to members of SE staff sending "congratulatory" e-mails to our worst [redacted] for reaching 100k rep with 7000 answers in a span of over a year - a whopping 1.4 upvotes per answer. And you call that many nice things! – Jyrki - No Shog9 - No SE Dec 6 '19 at 20:50
  • 2
    Sorry about having to downvote this. I understand that you believe in this, but I don't see anything that would be called a compromise here. A compromise needs software assistance when one faction never shows up for peace negotiations. – Jyrki - No Shog9 - No SE Dec 6 '19 at 20:59
  • 2
    Re posters of "answers on bad questions": They shouldn't be helping people in inappropriate ways. Kick them. They asked to be kicked. "Overwhelming desire" justifies nothing. That's how abusers try to justify their actions. – philipxy Dec 7 '19 at 7:41
3

A lot of criticism of SE is that you need rep in order to do basic things (like comment) and I think that is one of the primary reasons for repwhoring. Get enough points to be able to use the site as many of us take for granted. The site becomes "normal" at about 1 to 2 thousand rep.

Now once upon a time the rep system was useful to generate the site, encourage users to answer and vote and all the rest, but I think those days are done. Now we're trying to attract new users, and the first thing they see when they come here is effectively "you can't do that, you n00b".

So I would propose making the privileges accorded to rep fall into 3: the first tier is to ensure that scammers and spammers cannot run rampant by creating new accounts and commenting or answering with their spam. Then the vast majority of the privileges becomes available to anyone who has 100 rep. Then a few of the best privileges are earned when you get to some heady height of 10,000 rep. (these numbers were taken off the top of my head, and will require more thought, but describe the principle)

Then, the rep becomes useful for the old greybeards who have participated in the site significantly, but the average user does not feel penalised for not getting so involved that they feel they have to answer with poor responses.

It also removes one impediment to retain new users, as they will not feel quite so unwelcome by the site's implicit mistrust of them.

|improve this answer|||||
  • You may be onto something. Certainly we need something like rep to handle privileges and bounties. I think that simply discontinuing weekly rep leagues would be a service. – Jyrki - No Shog9 - No SE Dec 7 '19 at 5:32
  • 1
    It's difficult to get statistics which might test your suggestion in the first paragraph, but on some stacks it certainly seems that some users are still rep-farming with over 10k rep. (Also, with respect to the cutoff for a "normal" experience, from my personal perspective the two missing privileges which frustrate me on some stacks where I haven't much rep are seeing vote breakdowns at 750 rep and editing without being placed in the queue at 1000 rep). – Peter Taylor Dec 7 '19 at 10:58
  • 2
    The r-word is banned (for instance, its entry on The Many Memes of Meta has been deleted). – P.Mort. - forgot Clay Shirky_q Dec 8 '19 at 6:56
  • @PeterTaylor We have many users over 100k still repfarming. If this loose rep from newbies questions would be spread out somewhat evenly to the brightest newbies, it would be, if not a non-problem, at least tolerable. I could also support a universal rep cap of highest privilege + 5k for bounties :-) – Jyrki - No Shog9 - No SE Dec 9 '19 at 12:20

You must log in to answer this question.

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