33

I come here from time to time, mostly to read the month top questions and answers. Even if I haven't been very active, I've been around for quite some time (my first recorded interaction with this site dates back to 2013.

I have the feeling that the activity has been decreasing. I don't have access to mod tools, but I tortured the search tool and got these hints:

There are 202 Q or A with a score above 50.

There are 120,095 questions in total. From which:

There are 26415 questions without an answer. From which:

Which means, the percentage of questions without answer is around:

  • 30% between 2017-01-01 - 2019-12-31
  • 19% between 2014-01-01 - 2016-12-31
  • 11% between 2011-01-01 - 2013-12-31

I'm not willing to compute more stats, but these numbers are not looking really good to me. However, I see from this page, that there is about 57000 visit per day. And that the site is ranked in the top third of all sites.

So what is happening? all these visitors not engaging? Not answering? Not voting? Am I the only one to have this kind of feeling?

| |
  • 3
    Mod Analytics state there has been 2.4 Million page views in the last 28 days. (+36%) there is always a lull in December for 'holidays'. Note Stack Exchange had less sites way back (there is Earth Science, Open Data and Computer science now). – Mapperz Feb 4 at 17:04
  • @Mapperz Interedring. But still, it seems that not many users are interacting (answering, voting,...) do you have stats about that? – Legisey Feb 4 at 18:19
  • 1
    This not exclusive of GIS SE but all of Stack Exchange en.wikipedia.org/wiki/…. – Mapperz Feb 10 at 22:43
  • 7
    Have you considered that many of the existing questions/answers are applicable still? Many of the concepts have not changed significantly and a good question many years ago is still a good question worthy of an upvote, I've upvoted and/or linked to a few posts with 30+ votes because they were still relevant to the question as most platforms are extended by adding new functionality to an existing base; Emergent software (like ArcGIS Pro) is an exception... so there aren't very many new good questions. – Michael Stimson Feb 12 at 5:05
21

While I actually agree with your feeling here, I think it's important to keep in mind that the community (both this site specifically and SE in general) has changed quite a bit over the years, so comparing content across that time may not work out so easily.

Looking at the Q/A with a score above 50, the top 3 are questions that would be closed within minutes if they were posted now. They have been preserved since they are actually useful, but it seems the community now feels that moderating questions like that is not feasible given how large the community itself has become.

For unanswered questions: Improving on 19,000 unanswered questions is definitely relevant. While there is some discussion around why so many questions are unanswered, there overall push seems to be just improving the total number.

Finally, there's the rise of what I'd call meta-moderation. Moderation has slowly gone from loosely enforcing some minimum standards in an effort to improve the community to a highly structured rule based enforcement that can be downright oppressive at times. So much so, I've had to justify answering a question instead of closing it, when an answer would have been more helpful.

Couple all of the above with recent issues with SE itself, I would expect some users to start taking a step back from the community. I certainly have.

So, is GIS Stackexchange ill?

No, it sounds like the numbers show that the overall visits are up, and there were probably multiple questions posted while I wrote this answer.

But, that doesn't mean that this is the same community as it was in 2013. It has most certainly changed.

| |
  • 23
    I largely stopped contributing to this site because of the over-moderation. I'm not the only expert to have done so. – dbaston Feb 5 at 15:57
  • 12
    Agree with much of what has been said here. I quit/come back to this site based on exactly what @dbaston says. Getting off topic from the post: but I would love to see the questions which have list answers or 'best' come back. Dont think that would change the overall usage of the site, just offer up more 'help' to the casual user. – KHibma Feb 5 at 17:30
  • 7
    Some of this answer is incorrect or misleading. The top two threads definitely would not have been closed at any time in the site's history--and the third is closed. The site's activity, on most metrics, peaked around 2015 and has been declining since then. (That's not unique to this site, btw.) I do agree about the change in moderation and that's why I have been only minimally active for the last several years. – whuber Feb 6 at 17:24
  • 3
    Additionally, the top question, currently locked, states "This question and its answers are locked because the question is off-topic but has historical significance. It is not currently accepting new answers or interactions." which would indicate to me that this question would be closed as off topic if asked today. – Evil Genius Feb 6 at 17:45
  • 1
    Also, none of that changes the fact that those questions, while useful to this community as evidenced by their score, are opinion based and/or shopping list questions, which I think have pretty clearly been defined as off-topic by the current moderation staff. – Evil Genius Feb 6 at 17:54
  • 1
    @whuber Click on the link provided in the question. Look at the first three posts listed. Not top voted, but top as displayed. Which is even more ironic, since the 'most relevant' post in that list is now off-topic. – Evil Genius Feb 6 at 18:56
  • 1
    "Top as displayed" is irrelevant in this search, because it contains no search terms to determine the sort order, which therefore is arbitrary. It is a mistake to read anything into the sequence in which the results are displayed. – whuber Feb 6 at 19:50
  • 2
    I'm not nit picking: you have made a broad claim that just doesn't stand up to examination. It is false and misleading. – whuber Feb 6 at 23:22
  • 12
    "Finally, there's the rise of what I'd call meta-moderation. Moderation has slowly gone from loosely enforcing some minimum standards in an effort to improve the community to a highly structured rule based enforcement that can be downright oppressive at times" I agree – user2856 Feb 7 at 0:39
  • 2
    @whuber I've been clear in the past who I think is at least partially responsible for that. What I'm pointing out is that a vocal part of the community has an expectation that for something to be important it must follow a rule, be statistically significant or otherwise be scientifically provable. I just clicked the link in the question, out of curiosity, and noticed the top 3 posts I clicked on were questions I'd consider important and were also questions that would be closed today. While not statistically significant, it is relevant to the discussion. – Evil Genius Feb 7 at 11:50
  • 4
    In case you are curious, or want proof of what I've said in the past: you can read some of it here. I'm not spending anymore time trying to find more. – Evil Genius Feb 7 at 12:35
  • 10
    "a highly structured rule based enforcement that can be downright oppressive at times." Yep. I got sick of asking questions when they would constantly get closed for technically being two questions. (Usually the "two questions" were just different ways of phrasing the same thing, or slightly different angles to help explain what I was having trouble understanding.) – Steve Bennett Feb 13 at 4:28
  • 4
    Yes, it's your opinion that the "one question per question" rule is sufficiently important that it's worth closing questions that violate it. It's not mine, and based on the number of upvotes to my comment, you might wish to reconsider. (The historical context shows that it was never meant to be such a rigid rule, but rather discouraging a specific behaviour about follow-up questions, not multiple closely-related questions posed at the same time). – Steve Bennett Feb 17 at 12:13
  • 4
    I didn't say "no flexibility", but...not much. You strongly believe in rules, and mechanisms such as close/reopen votes. These work for you. They don't work for everyone, nor do I see them deployed as zealously elsewhere on the network that I participate. And your overwhelming presence everywhere on this site makes that feel even worse, like having a teacher hovering behind you while you work. It's the difference between getting a parking ticket once a year, vs a parking inspector following you around and waiting by your car. Same rules, vastly different experience. – Steve Bennett Feb 17 at 22:35
  • 7
    I feel like a lot of this conversation is me attempting to share with you how other people use the site, how they react to heavy-handed moderation, and why a lighter touch would be appreciated by a lot of people. And you responding by telling me how great the moderation tools are and how we should all use them more. Has anything I have said opened your eyes at all? Been valuable in the slightest? Or do you see my views as simply incorrect and ill-informed? – Steve Bennett Feb 18 at 11:19
20

It is true that the industry has evolved and now many questions have found a new place in other communities. I would like to start a conversation about the usefulness of some of these sites (like the Open Data) from a GIS perspective, but that would be another topic.

It is also true that some of these stats are time-depend and the more time a question is active the more chances it has to get an answer (although I don’t think the relationship is linear at all).

However, I couldn’t agree more with @Evil Genius answer, I have this same feeling.

In my case, even if I regularly enter and check the site, contributing is becoming less and less appealing to me due to this regime of over-moderation.

According to Jeff Atwood, the Art of Moderation lies in doing as little as possible:

The ideal moderator does as little as possible. But those little actions may be powerful and highly concentrated. Judiciously limiting your use of moderator powers to selectively prune and guide the community — now that’s the true art of moderation.

There have been multiple complains about improperly closed questions or excessive edits, you can simply check the last questions of Meta.

As for the closed questions I would like you to check these two queries:

https://data.stackexchange.com/gis/query/1193105/number-of-created-and-closed-posts-per-year

https://data.stackexchange.com/gis/query/1174445/unilaterally-closed-posts

I encourage you to export these tables and do some joining and aggregating. The proportion of unilaterally closed posts by some users is humongous.

As for the edits, there is an easier way to check of the amount of edits per user: https://gis.stackexchange.com/users?tab=Editors&filter=all

I don’t like to point fingers but if there is a problem of over-moderation, the data seems to be pointing in one direction.

My impression is that instead of doing as little as possible some users are doing as much as possible to find whatever reason to modify anything.

In my opinion this is discouraging for new users and tiresome for old ones. Personally I would like to see just the minimum edits to maintain comprehension and site functionality and less censorship, simply let the voting system do its job.

| |
  • 7
    That edits by user page. Wow. I knew that, but I didn't realize it was by an order of magnitude. – Evil Genius Feb 6 at 14:43
  • 6
    Hahah yes, try plotting the data! Another fun fact: if you sum all the edits from the top 2 to 62 users you won't get to the top 1! – Albert Feb 6 at 14:59
  • 6
    Hopefully they are using more than 16 bits to store edits per user. We'll find out soon! – dbaston Feb 6 at 15:56
  • 8
    I have not seen an edit by another user than the author so far which would not improve the quality of the post. I still think fixing grammar, typos and formatting should be encouraged. – MrXsquared Feb 6 at 20:49
  • 4
    I also vote a lot (43,872 times to date) rewarding the posters of content that I think looks useful with 235,510 reputation points. I do my best to try and "let the voting system do its job" but I cannot recall ever seeing an upvote that I've cast directly lead to further improvement of a post. I have certainly seen a very large number where close votes and downvotes have led to that. – PolyGeo Feb 9 at 23:33
  • 14
    Look, I don’t doubt you do your mod tasks thinking you are bringing quality to the site. The feeling is there are many users, new and historically important, who think and even have said this is too much. Maybe a blind or obsessive person wouldn’t be able to read between their polite lines but I’m sure this is not your case. Some users have said they are stepping back due to their feeling of the site being over-moderated. This is SERIOUS and eventually something will have to be done. – Albert Feb 10 at 8:59
  • 9
    Imagine this is a big house where all the users of GIS SE live all together in community. Imagine there is a person in this house that is moving all the books, pans, pots, chairs, plants, clothes. This person even enters in your room and after making your bed in his fashion removes all your junk from your first drawer and puts all your underwear 0.7 centimeters apart from each other because this of course, is the best place for them to be. – Albert Feb 10 at 9:00
  • 10
    This is how some of us are feeling. We feel invaded. You have more edits than the top 60 editors of the site COMBINED. You say you do this to bring quality to the site but here is the thing: GIS Stackexchange is not your backyard. – Albert Feb 10 at 9:01
  • 5
    The funny thing is, to me anyway, is I feel like this site is already in a feedback loop. @PolyGeo seems quite content constantly tinkering away at, well, everything. That has caused some users that would normally engage with the site for more than just answering the question they are interested in to back away. While GIS does have quite a lot of views and questions per day, I feel that the average user is not interested in the community. This in turn gives PolyGeo more to do (edits, closing, etc.) which then drives more of us away... – Evil Genius Feb 12 at 13:01
  • 7
    Just my little datapoint here. The heavy-handed moderation is literally the reason I don't contribute here. Each time I think about maybe posting a question, I think about either a) Having yet another confrontation with PolyGeo, or b) Modifying my contributing behaviour (which seems to be just fine on every other SE site) to suit PolyGeo's whims. Both of these are really unpalatable to me, so I just don't. – Steve Bennett Feb 13 at 4:35
  • 3
    So...just some feedback here. My comment was that participating here is unpleasant because of your heavyhanded enforcement of arbitrary rules. Your response to that is...to nitpick the word that I used and to request that I use a different one instead. Does this give you any insight into my experience at all? – Steve Bennett Feb 16 at 11:49
  • 4
    I'm not even sure you can tell someone that their interaction with someone is not a "confrontation", if they're telling you it feels like one. Maybe it isn't to you. Many of these interactions feel confrontational and stressful to me. I'm guessing that many others feel the same, or more strongly - I'm not particularly sensitive to such things. – Steve Bennett Feb 16 at 11:56
  • 4
    I think what you are saying is "I will curate the way I choose to, thank you very much". Which is consistent with my view. It turns out that the way you curate is a serious discouragement to my willingness to participate here (and apparently others). I don't expect you will change. But the question was asked, why has activity decreased, and this is my part of the answer to that question: I don't participate much, because I don't enjoy experiencing the heavy-handed actions of this particular moderator, and I can't avoid them. – Steve Bennett Feb 16 at 23:03
  • 5
    Sure, happy to accept that correction. I don't agree with it - I think having answers is much important than how those answers are presented, etc. And, personally I think you continue to underestimate the negative impact that "content curation" has on creators and their willingness to continue creating. – Steve Bennett Feb 17 at 2:03
  • 5
    Right, the inevitable reference to the depths of the Meta canon. I don't think anyone's disputing that it's possible for editing to be helpful. That doesn't mean it always is. – dbaston Feb 18 at 4:35
16

In addition to several good reasons already posted in answers here, there is another factor that I don't think I've seen addressed yet...

The longer the site has been around, the more likely it is that any given question has already been asked. Or to put it another way, the more likely it is that the question a user wants an answer to has already been asked and answered.

The more time passes, the more likely it is that a user searching for answers will find the answer here, and not have to ask the question at all (and if they did ask the question, it would be flagged and closed as a duplicate, and therefore not be answered).

| |
  • 1
    I do agree with your view. It would be helpful to compare with other SE sites to see where SE GIS settles in the trend analysis. Perhaps someone can run a Query - data.stackexchange.com – whyzar Feb 10 at 15:50
  • 2
    Yes, fully agree with that. For me it is a good sign that relatively more complex question are emerging. Maybe the stats should compare tags of "new" softwares (e.g. ArcGIS Pro, GEE ) to see if the rate is different in there. – radouxju Feb 11 at 7:52
  • Yet, I am in constant bewilderment about how, despite our best efforts, there seem to be absolutely no trace of information at all archived on this board about the mysterious topic of 'reprojecting vs. setting a CRS'... – geozelot Mar 10 at 22:43
  • @geozelot This is a focused Q&A site rather than a bulletin board but what you mention is our most frequently asked question and answered at gis.stackexchange.com/a/27059/115 for ArcGIS Desktop. If you're using something else I think we'll already have it too. – PolyGeo Apr 2 at 9:35
10

As a relative newcomer to both the GIS stackexchange as well as GIS itself, I can't comment on historical trends. But I would point out quantity does not equal quality.

My experience with this community is better than with some other online communities (SE and elsewhere) whose stats would doubtless be higher. Questions span the range from beginner to expert, answers are generally constructive and comments free from useless sniping, and an amazing number of challenges I have had as I have gained experience in GIS have had good questions and answers on the GIS SE already. Ergo, I am not sure my benefit from the site would necessarily have been better if some of the stats had been higher. (To a certain extent yes, of course, more activity = more eyeballs and so more specs of gold somewhere to be found, and simple answers likely to be given a bit sooner, but beyond that, I'm not convinced.)

Does moderation feel heavy handed to me? Sometimes. And (less often) occasionally too light as well. But the impact of that on my interest is far less than the availability of quality information.

| |
  • 3
    I have a feeling that for couple of years now when the first GIS lessons start somewhere in the World the students are sending their homework to gis.stackexchange. Probably it is not easy to analyze if my feeling holds water and if such questions show in statistics of closed and unanswered questions. – user30184 Feb 17 at 15:51
10

I've been here since the beginning, and served as pro-tem moderator for a time when the site was finding it's legs. My contributions have been downwardly trending for several years. It would be wrong to say the distance is because of over-moderation. Home and work are claiming more of my available attention, and my public internet presence has quietened down on all fronts. However the tone of the present day GIS-se curation does give me pause from time to time when I do feel the impulse and have some time to add something new. It's not a stopping force for me, I'm heavily invested and still wear what's left of my GIS moderator t-shirt gift in pride. Not everyone has that investment to carry them through. It's clear from my reading here that curation is a block for people, it just IS.

Curation is necessary, it's vital, a core principle that makes this community valuably different from many others. It is not going to be set aside.

However we should recognize that manner of delivery has problems. Let's acknowledge that at least, even if there's no solution in the same breath, or those soon after.

Problems grow when there's a only a few voices that express a significant and vocal opinion on every item, all the time. It's not that they don't have valid points, that the contribution isn't valuable, isn't true. It's that other opinions and voices get crowded out, the soup's flavour goes flat, and over time the community is undernourished.

Okay, so what's a remedy then?

I don't know, I'm not that insightful or wise. I do believe this very conversation is a step in the right direction though.

I encourage people to continue voice their opinions, with reflection and thought, with enough personal flavour that it's real, and enough consideration-of-other that it transcends flame-war rhetoric.

Equally important is for readers to take the time to absorb and reflect on what is said, especially those things that run counter to our own certainties, to allow the truths conveyed to seed and grow, to be open to change, to the possibility that the view one starts with can be altered without sacrificing core principle.

| |
  • 4
    I encourage people to continue voice their opinions, with reflection and thought, with enough personal flavour that it's real, and enough consideration-of-other that it transcends flame-war rhetoric. it seems like there is a number of people who have been voicing these concerns over the years and nothing has changed in terms of "curation" or reflection among some of these readers. At what point do we reach a critical threshold? This site is already losing skilled users at the whim of one's curation standard. – GISKid Feb 18 at 12:58
7

I'm going to suggest that, yes, SE GIS is ill.

Answers so far have noted that: There is usually a lul in activity over the new year and holiday periods. The userbase has been steadily increasing, as has activity.

I'd suggest that at least in terms of the second, this is because there is little other place for GIS users who have questions to go for help. Stack Exchange and Stack Overflow are well embedded into technical disciplines and search algorithms that they continue to beat out the competition, and even vendors own preferred support mechanisms, almost as if by default.

As someone who has had to learn a wide range of programming and extended GIS skills over the past 10 years, and who has moved from viewer to interactor in that time, I'd say the number one reason I see for the percieved dropping off of quality interactions in the community is over-moderation. There is more than one moderator on GIS-SE who over-zealously penalises people who are new to GIS for asking questions that are obvious to most readers, but do not necessarily fit the format they wish the questions to be asked in. At least one of those same moderators expends a great amount of time investing in minor edits on every other post. When they do reply, the replies are often condescending, overly negative or disparaging, or do not make clear what they would like to be fixed about the post / how to do it.

The moderators primary job is to keep the community in check, to boot out idiots, and keep things civil and on track. The moderators job should not be to correct users spelling and grammar mistakes, shut down legitimate lines of questioning, or spend their time informing clearly not native english speakers that they are not putting in enough effort to clearly articulate their question to the unreasonably perfect standards that the mod team (or is it just one mod?) expects.

IS GIS-SE ill? Yes. And the reason begins with the leadership.

| |
  • 6
    but some many of our users can't be bothered to take the tour to find out what our standards are. How do you propose we make the site useful to everyone, with out 1000 but my shapefile doesn't line up questions none of which have a correct answer? – Ian Turton Feb 10 at 17:28
  • 3
    One of the problems is that the tour doesn't actually describe the standards of the most active moderator. It doesn't tell you that you must restrict the solution to a single piece of software, or that you must not ask questions about PostGIS without providing a block of SQL, whether or not it is relevant, or that you must include only one question mark symbol in your question; you may not phrase the same question multiple ways, etc, etc. – dbaston Feb 18 at 16:32
  • @dbaston The Tour is necessarily very concise but if you follow this link sequence from within it Visit the Help Center - What topics can I ask about here? - What makes a good question? (or go via any other path to that Meta Q&A FAQ) you'll see the suggestions I offer for what I think makes clear and focused questions. I just updated my answer there to correct two things I would have said once but have changed my view on with help from various Q&As here. – PolyGeo Feb 18 at 23:36
5

I am not a statistician, but I think something that needs to be taken into account when considering the numbers that you have presented, is that the more time that passes, between when a post is made and the present, the more likely it is that a:

  • post will be voted on; and
  • question will be answered

This is because the more times a post is viewed, the more likely it is that 1 to many people will take an action on it.

Of the four sets of statistics that you have presented, I think only the second one will not be heavily correlated with time since posting.


Another observation that may or may not add value to this discussion is some manual copying of data that I did from https://stackexchange.com/sites#oldest. It is a table with data from the 11 sites that launched 9 years 7 months ago (of which we are one). On most measures we seem to be in about the middle of an illustrious field.

enter image description here

| |
  • 2
    I'm not really sure about that. Can you see how much activity there is in average on years old questions with 0 answer and 0 upvote? I think that if question is neither upvoted, neither answered within a relatively short time of its creation, it will never be. – Legisey Feb 6 at 12:25
  • That may be in SEDE but I don’t write other than rudimentary SQL. I often vote on, edit and answer questions that are many years old. I’m pretty sure any closed question more than 9 days old with 0 upvotes and 0 answers gets Roomba-ed. – PolyGeo Feb 6 at 12:26
  • 2
    @Legisey You're correct, but it's been that way for many years. At least 75% of questions that are answered have the first answer within 24 hours. 99% of questions that have answers have at least one answer within 4 days. – Evil Genius Feb 6 at 14:22
  • 3
    Now, what's interesting is the fact that the time a question remains open before being closed has decreased dramatically. In 2011 around 28% of questions that ended up being closed were closed within a day. In 2019 that's 78%. 2012 to 2013 was quite the jump, 33% to 48%... – Evil Genius Feb 6 at 14:35
  • 2
    @EvilGenius It's just a thought, and I don't know the precise numbers, but in 2011 the only people who had the cast close and reopen votes privilege, which needs a rep of 3,000 (and I assume also did back then), may have been the three site moderators, whereas nowadays there are around 300 users eligible to cast them. – PolyGeo Feb 7 at 3:40
  • 1
    It very well could have been only 3 moderators in 2011, but that doesn't change the fact that the majority of closed questions were unanswered. The simple response to that is: they weren't good questions and they would never get answers. But, what about the questions that end up being reopened? Since we've established that over 80% of questions get answered within 24 hours of being posted, do you think that a question not accepting answers in that time frame has an effect? I think it does. – Evil Genius Feb 7 at 12:08
  • 1
    @EvilGenius I don’t have numbers but I suspect a majority of questions that are re-opened get to that state after being edited to address the concern that led to their closure rather than being left unaltered. If you think no question should be closed for 24 hours to see what types of answers arrive on them then I think you should propose a feature request for that at Meta Stack Exchange. I think if a question should be closed eventually, then it is best that it is done immediately while its asker is still around to address what would become the eventual close reason. – PolyGeo Feb 7 at 12:32
  • 4
    The facts support PolyGeo's position in these comments. It is correct that during the first few years there were relatively few people who could vote to close or reopen questions, so that most of this work was performed by "diamond moderators," whose votes take immediate effect. (It takes five non-mods otherwise.) On any site it's important to close ambiguous or ill-formed questions quickly to head off the confusion that results when well-meaning users rush to answer their interpretation of a question. Questions are almost always reopened after they have been clarified or improved. – whuber Feb 7 at 14:39
  • 3
    I think you can look at the pattern "question was asked, then closed, then edited, and reopened" as support for many things. In my case, step 3 of that pattern has often been a begrudging and irritated edit to conform to what I see as an arbitrary rule, adding little value in the process. But for a moderator it could look like "see! curation works! by closing the question, it got improved!" – Steve Bennett Feb 19 at 0:34

You must log in to answer this question.

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