12
$\begingroup$

I can see in this answer that questions shouldn't be asked with "thank you"s, "cheers", etc. But should we remove these during editing? I'm asking because the policy on the Mathematics site is to leave "thank you"s in place, so I don't want to make any assumptions.

Also, sometimes I see posts that begin with "This question is too simple for this site", or similar. It seems less obvious that that should be edited out, but should that text go too?

$\endgroup$
19
$\begingroup$

Like so much around here, this pivots on the slight tension between

  1. Individual posters with short-term goals of getting an answer, who almost always want to express appreciation of your attention, their need for an answer, their gratitude if you provide one, and so forth. That's all fine, and I have no inclination to dismiss or downplay politeness.

  2. The long-term goals of the forum, of building up posts with concise, clear and correct questions and answers as an archive for all interested. These goals are what drive a desire to cut out material that doesn't help technical understanding. Even cutting 5% of a post that isn't needed is 5% cut for possibly many future readers in a busy world. Questions and answers are public statements that happen to be written by individuals, rather than personal statements that happen to be visible widely.

I say: We really can assume goodwill from essentially all concerned, except in a very few cases, and those separately require different treatment. So, although naive or new posters sometimes pad their posts, the silent reply is Thanks for the thanks, but we don't need that. Sometimes I will comment on the point if it seems appropriate, but I can't claim to be consistent on that. Sometimes I am in a hurry too, and irritated by someone's sloppy and careless presentation (with latitude when it's evident that English is not the first language).

I edit quite frequently. I usually edit for concise style, improved grammar, spelling and punctuation, and improved presentation of mathematics and code; but I will also cut down emotive content, including not only extended thanks, over-lengthy personal statements, and pleas of desperation or urgency. I think it's germane that those edits are very rarely reversed, although it is clearly very possible that many people never noticed, or did notice and disapprove, but lack the time and inclination to do anything about it.

$\endgroup$
12
$\begingroup$

I don't think there's a formal policy, but I'll talk about what I've noticed myself doing.

If someone just says 'thanks', I tend to leave it, even when doing a lengthy edit, even though it strictly doesn't belong.

However, if I'm removing a subsequent signature, I may take 'thanks' as well.

If it's a long rambling thanks (say a couple of sentences) I take it out.

I object to the distraction of put-downs of the question or apologies for asking a basic question by the asker and tend to have a strong impulse remove those or encourage the OP to. (Not least because it puts down other users of the site with the same problem. They may think they're only calling themselves dense or uneducated, but since the questions are permanent, and often posted questions are questions other people have, ... they aren't)

There's potentially a similar put-down with the basic-concepts tag. I'd say easily 2/3 of the posts with the tag don't come remotely close to being basic (people tend to underestimate the non-basicness of their problems, sometimes drastically), and many of the ones that could rate such a tag ... well, I often wish they wouldn't, for the same reason I don't like "sorry this is such a dumb question".

We were all ignorant once - what's so wrong with a basic question that we must apologize for it? I don't come here only to answer questions at the outer edge of my knowledge. Indeed, sometimes formulating a good answer to what looks like a relatively simple question can be a surprisingly subtle task, and the review of the basics from a variety of viewpoints can often be helpful even when it's reasonably familiar territory.

Of the emotive statements that come up, the kind that tend to get me the most is begging. I find those particularly objectionable. (I don't so much object to a simple factual statement about time pressure, but multiple lines of "PLZ answer my question" or even worse, a title with it... that stuff pushes my buttons for several reasons at once.

$\endgroup$
  • 1
    $\begingroup$ Very close agreement. "Dumb questions": I sometimes think if it really is a dumb question, then don't ask it, or why can't you find out; but naturally beginner questions are fine, not least because lots of statistics is weird even when you are accustomed to it. Claims of urgency or desperation and other attempts at emotional blackmail are unacceptable; the principle of charity here is answering interesting questions because it benefits many, not addressing the personal situation of the questioner. Add for me "I am not a statistician" as a plea for extra help; well, I am not one either! $\endgroup$ – Nick Cox Mar 14 '14 at 9:32
  • 1
    $\begingroup$ Good point about the [basic-concepts] tag. I never really thought about it. I have added it to a couple of Qs, but never in a pejorative way. Come to think of it, do you suppose it's a meta-tag? I haven't had any big objection to it, but maybe it should be banned. I always thought of it as a way to collect several categories that might be too small otherwise; basically for the type of info that comes in the 1st 1/3 of stats 101 (eg, difference b/t mean & median, what is a histogram, etc). $\endgroup$ – gung - Reinstate Monica Mar 14 '14 at 14:15
12
$\begingroup$

There are some good answers here. My policy (and I think the default overall), is that if I edit, I remove all peripheral comments, be they 'thanks', '-Bob', 'I'm not a statistician, sorry about the dumb question...', etc.

A topic that hasn't been covered much yet is deciding when to edit, or to approve suggested edits.

  • If the only thing that needs to be done is remove 'thanks', I won't edit so as not to bump the thread. If someone else has suggested an edit that only removes 'thanks', I click reject as 'too minor'.
  • I am slightly more likely to edit (approve) to remove only '-Bob' or 'I'm not a statistician...', but the older the thread is, the more likely I am to leave it.
  • The more of such stuff there is, the more likely I am to edit.

  • If I edit for whatever reason, I will always remove this sort of thing as well. If I review a suggested edit in which these have been left in, I will click improve and remove them before approving.

$\endgroup$
  • 4
    $\begingroup$ This seems to match widespread current practice. The principle could be "If it's worth editing, it's worth editing well", but doesn't exclude correcting one non-trivial slip. $\endgroup$ – Nick Cox Mar 14 '14 at 15:38
  • 1
    $\begingroup$ Given the worry over bumping the thread, is there any stackexchange support or hope of support for a minor/major edit distinction? Wikipedia has a minor edit checkbox, but I can see making minor edit the default here, with an explicit opt-in for new material that create a bump. Dreaming, I assume... $\endgroup$ – xan Mar 14 '14 at 18:49
  • 1
    $\begingroup$ I'm afraid I'm not familiar w/ the minor / major edit distinction, @xan, but might be asking this: Feature to edit without bumping? $\endgroup$ – gung - Reinstate Monica Mar 14 '14 at 19:13
  • 1
    $\begingroup$ Yes @gung. Good to see it's already been discussed even if not with the outcome I was looking for. $\endgroup$ – xan Mar 14 '14 at 22:54
3
$\begingroup$

Personally I leave "thank you" statements or the like. It is nothing that clutters up the question and at the end of the day it's nice if people are friendly. A different thing is when someone has a lengthy introduction which is about anything else but statistics. Fortunately this rarely happens.

I've never seen anyone starting their answer with "this is too simple for this site" or likewise statements, though I'd apply the same rule of thumb as before: if it clutters up the text due to excess length then it has to go. Otherwise I would just leave it even though I think it's not very polite to begin an answer this way. What is a trivial problem for some users can be a huge challenge for others.

$\endgroup$
  • 1
    $\begingroup$ "I've never seen anyone starting their answer with "this is too simple for this site" or likewise statements" - Let me present Exhibit A: A question that I posted some time ago: stats.stackexchange.com/questions/41789/… $\endgroup$ – Speldosa Apr 9 '14 at 15:29
  • $\begingroup$ Just that I haven't seen it in almost one year of active membership doesn't mean it wouldn't happen. I'm sorry for your experience but in general this isn't how the community treats new users $\endgroup$ – Andy Apr 9 '14 at 19:10
  • $\begingroup$ Sorry, I misread your statement. It was I who started my question with a statement saying that it probably was a "n00b question" :) $\endgroup$ – Speldosa Apr 10 '14 at 11:30
2
$\begingroup$

From the perspective of a beginner trying to gain a foothold in machine learning through CrossValidated, a reasonably lenient policy on things like "thank you"s seems to be helpful, for several reasons that I haven't yet seen brought up in this thread:

  1. I haven't seen any of this yet at CrossValidated, whose members have impressed me with their civility and generous help so far, but that isn't the case at a lot of other forums for related topics at StackExchange and other sites, particularly programming ones. There are two principal temptations which can ruin sites like this (both of which come from hubris): the experts who think they can't possibly be wrong and let everyone else know it, and their nemeses, amateurs who pretend to know a lot more than they do. If either one gets out of control, people become reluctant to post because they have to brace themselves for what sometimes amounts to outright abuse (I can name specific forums where semi-famous experts in various fields make a sport of ridiculing newbies and engage in shocking public meltdowns, but I won't). The damage from that sort of atmosphere is incalculable because we'll simply never know how many lurkers never signed up for the site or posted (as I have for many years in the forums I'm referring to). It can occur silently, with no outward signs like falling membership numbers, but with devastating effect.

    Newcomers like me can't do much about the problem of supercilious experts (which I've seen no signs of so far at CrossValidated), but we can put a drop in the bucket towards solving its polar opposite by short expressions of gratitude and the occasional short disclaimer of inexperience. I would be less comfortable about posting if such niceties were regulated too aggressively, because it allows me to consciously nip a potential source of tension in the bud. Other posters, particularly ones who've spent years earning the right to speak on these issues, are less likely to misconstrue posts and take offense if gratitude and humility are expressible in reasonable proportions.

  2. The subject matter in forums like these can be highly technical and taxing to think about, plus working with data mining software can be frustrating at times. This can raise the tension level further, whereas the occasional expression of basic courtesy can reduce it.

  3. Tone of voice is difficult to get across in Internet discussions, where people tend to freely mix on-point information with casual conversation; that's why in chat I consciously use emoticons a lot to avoid misunderstandings. "Thanks" etc. sometimes serve the same purpose.

All three points suggest that an overzealous approach to editing out "thanks" and disclaimers etc. can actually reduce the quality of the answers in the long run. Why? If people don't post questions, there are no answers, regardless of quality or length. Fanatical text slashing can be as self-defeating as fanatical accounting, where 0 expenses and 0 income also equals a balanced budget. I don't see that happening here: in fact, the moderators have been generous in letting me ask questions verbosely, which can also end up reducing the amount of text in the long run by avoiding unnecessary replies based on misreading of questions. "Thanks" are in order for maintaining that balanced policy so far...when in doubt, I'd always lean towards a more lenient response.

$\endgroup$
  • 6
    $\begingroup$ There's much truth in all that - for discussions on internet forums, the equivalents of which here are chat & comments. But when you're posting a question on a Q&A site the polite thing to do is to make it as concise as possible, shorn of unnecessary thanks, apologies, & personal details; out of consideration for the people who are going to read it now, to answer it, & in the future, to find an answer. :P $\endgroup$ – Scortchi - Reinstate Monica Feb 20 '16 at 14:18
  • 6
    $\begingroup$ Thanks for adding a different perspective. It's interesting to have the issue placed in the context of other SE sites (which I have little experience with). I think there are a couple of issues that might be specific to this site. The overwhelming majority of new posters here say something like 'sorry this is a dumb question, I don't know much statistics' etc. These statements are painful for me (I don't know much statistics either). They might be more informative if some posters knew a lot & some knew a little, but nearly all new questions are basic. $\endgroup$ – gung - Reinstate Monica Feb 20 '16 at 14:35
  • 6
    $\begingroup$ We can tell the OP's sophistication from the question, so it's redundant. Moreover, to answer a statistical question, we need certain information, possibly a good deal of it, about your situation, but OTOH the longer a Q is, the less likely people will fight through it; so there is a tension there. We want to get the right information, all of it, but only the needed information, as quickly & easily as possible. What is best for us to answer the Q, & thus best for the OP (& best for future readers who come across the thread), is to minimize the number of extra comments. $\endgroup$ – gung - Reinstate Monica Feb 20 '16 at 14:44
  • 4
    $\begingroup$ @gung: Sometimes it can be useful for the OP to say they're familiar with such & such a method but don't know how to extend it to this situation; or to ask for an answer that illustrates something graphically, & not just with equations, or without matrix algebra - it's "I don't know much statistics" & the like that can be rather off-putting because they seem to want to impose some sort of constraint on the answer while being vague as to what it consists of. $\endgroup$ – Scortchi - Reinstate Monica Feb 20 '16 at 14:52
  • 1
    $\begingroup$ @Scortchi Would specificity help? If instead of saying "I don't know much statistics," the writer says exactly which types of stats they're unfamiliar with, so that the responders can try alternate means of explanation? Something like, "I grasp matrix explanations well but don't understand tensors" or "I'm good with geometry but have a tough time with calculus." etc. $\endgroup$ – SQLServerSteve Feb 20 '16 at 20:19
  • 3
    $\begingroup$ @gung I have quite an axe-wielding approach to what I regard as extraneous comments. Having said that, I quite often leave "Thanks for giving a full answer" or "Thanks for pointing me in the right direction" (where the thanks are meant in anticipation) because they give a sense of the type of answer that the OP expects (rough intuition vs step-by-step explanation). Other than something that gives a signpost as to what kind of answers are expected, I tend to chop it out. $\endgroup$ – Silverfish Feb 20 '16 at 22:32
  • 2
    $\begingroup$ I enjoy participating on this site & maths. Compared to other stack exhanges sites this one isn't open so much to short quick-fire answers, which perhaps goes a little way to show why it has a different feel; although most of the credit has to go to the users who most frequently contribute to and help moderate the site. I don't agree with you about the benefits of retaining thankyous. (In case it happens, note that a question can be downvoted on meta because people don't agree with your suggestion, rather than it being a "bad question" as on the main site.) $\endgroup$ – TooTone Feb 21 '16 at 15:03
  • 1
    $\begingroup$ @SQLServerSteve: Exactly! When relevant - as gung says, it's often not necessary at all. $\endgroup$ – Scortchi - Reinstate Monica Feb 21 '16 at 16:38
  • 3
    $\begingroup$ Part of this interesting and thoughtful answer seems to boil down to: There isn't really a problem with civility here on CV, but there would be if we had the sort of aggressive behaviour that you have seen on some other sites; so we should not do that. That's welcome good news, but it is perhaps self-cancelling. Pedantic note: nemesis doesn't mean antagonist. $\endgroup$ – Nick Cox Feb 21 '16 at 16:52
  • 4
    $\begingroup$ I'm an aggressive editor in respect of personal or dispensable fluff, as I see the goal here as building up a repository of concise, precise, technical questions well answered. We don't need "thank you" or similar padding any more than encyclopedia articles begin with "Hi" and end with "We hope you find this useful". $\endgroup$ – Nick Cox Feb 21 '16 at 16:53
  • 1
    $\begingroup$ @Nick I'd like to see that too. My main contention is that there's a tricky point where zero tolerance can paradoxically increase the text and/or decrease the # of quality answers. Judicious thanks/specific disclaimers of technical level can prevent mixups/fruitless discussion paths that are not just testy but lengthy. In such cases, they reduce the overall text. There's 1 key difference from an encyclopedia: the public entries come from 1,000s of strangers with varied personalities & expertise. A slight allowance can be made for this slight difference & can aid in brevity in the long run. $\endgroup$ – SQLServerSteve Feb 21 '16 at 22:17
  • 1
    $\begingroup$ I agree with the spirit of the majority that such comments are typically extraneous ;I just wanted to illustrate that there's a boundary on the other side, where zero-tolerance editing sometimes leads to more text in the long run. It's like a greedy algorithm getting stuck in a local minima. I saw this happen often in journalism when editors chopped out detail that ended up requiring follow-up articles to set straight. A similar dynamic applies in forums like these too: cut in the wrong place & conciseness declines. 9 times out of 10 that won't apply, but there's always that 10th time. $\endgroup$ – SQLServerSteve Feb 21 '16 at 22:21
  • 3
    $\begingroup$ In one sense I am much more optimistic than you. Elementary or confused questions usually are self-evident. You don't need a poster to say "I am no good at statistics" if they are asking something very basic. Occasionally people get deeper answers than they want or need, but that's where others step in, and the deeper answers can help others. So, I don't think I've met a case where editing out fluff really obscured the technical question. Meeting the immediate and idiosyncratic needs of each OP is not the long-term goal of the forum, in any case. $\endgroup$ – Nick Cox Feb 22 '16 at 9:39
  • 2
    $\begingroup$ It is worth looking from the perspective of the OP. They are used to being flamed on internet forums and so they try to deflect that by stressing (a) that they know they are inept (b) they are grateful. In a perfect world that might not be necessary but we do not live in a perfect world. What they think when they find their post has been edited I do not know. $\endgroup$ – mdewey Feb 22 '16 at 16:04

You must log in to answer this question.

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