I know this question has been recently brought up several times, here is the most recent.

On the Area 51 site they list the progress of all the stackexchange sites, and here is the page for GIS. For comparison here is the page for the Stats forum (I did not know of this until someone recently posted it on the Stats meta forum).

I figured some of these metrics weren't brought up in the former questions regarding GIS getting the axe or growing the community, so this question is worth revisiting in a new (and hopefully well viewed) thread.

So we are doing well or excellent in percentage answers accepted, number of answers per question, and okay in the number of views per day. Where the metrics say we are lacking is the number of questions and the number of high rep users (high rep defined as simply 200 reputation, a seemingly quite low standard).

As a researcher I feel exactly the same as Matt Parker listed in this answer, and we need to ask more questions related to such topics (myself included) to both help growth and develop a core body of experts. I think one thing that we are lacking, both in analysis related and programming related questions, are simple questions that are oft encountered. Even if the user knows the answer such questions will be good to drive up traffic and allow other users to gain more reputation. They will also create templates of simple "FAQs" to point users to.

I know the dichotomous reading of the metrics are silly, and these statistics are as comforting as they are disconcerting, but if the Stackexchange sites care about them I feel we should care about them as well. It seems an easy solution to increase the number of questions asked, even if the expert knowledge to answer those questions is quite low to simply help the site grow at this point.

share
    
for what it's worth, the Ask Ubuntu site had stats pretty similar to ours when it was launched :) –  matt wilkie Nov 25 '10 at 21:08

8 Answers 8

up vote 11 down vote accepted

Please make note of those users who are under 200 rep, spend time to scan their questions & answers, and upvote those which are worthy. This is not a request to make pity votes or start passing votes out without thinking. Just make a little extra effort to find those which have been previously overlooked and are perhaps outside of your normal browsing path.

share

I'll confess that I find most of the questions on this site uninteresting, so I'm not surprised at the "worrying" numbers of participants. It's probably just my idiosyncratic impression (so please don't take this as criticism), but it seems this site is heavily Balkanized into extremely limited and technical domains: the questions are often software specific and only a minority of people will be interested in any particular one. SO, due to its size, can get away with that, but this one can't. Where are the questions about principles of GIS? About GIS science? About programming principles, GIS database techniques and strategies, cartographic techniques, appropriate use of projections and coordinate systems, and on and on? These are the kinds of things that could attract a wider audience, go to the core of our shared interest, and distinguish this GIS community from others. But these are exactly the questions that one hesitates to ask after seeing the stuff that actually gets posted here.

Anyway, I suspect I'm in a small minority, but thought it might nevertheless be constructive to give voice to this opinion in case anyone else has similar thoughts or helpful reactions. And if you all vote me down that will be a clear enough sign too and I'll be happy to retreat to alternative venues. Here's wishing you the best.

share
2  
Part of the problem (I think) is that most of the people who came here are from SO, which gives the site a pronounced slant to the programming side of things. We need to reach out to more non-programming GIS users, open source or not, to build up the community appropriately. So, how about you post some great GIS questions that aren't limited and technical? Even if you know the answer, we could at least use those questions as leverage outside of this domain to attract those we feel could benefit from this site and provide a benefit to it. –  Michael Todd Oct 20 '10 at 16:07
    
(And I would do it myself, too, but I'm one of those "kinda-sorta new to GIS and don't really involve myself with the day-to-day issues that arise in GIS" programmers who came over from SO.) –  Michael Todd Oct 20 '10 at 16:08
    
@Michael I was worried someone would challenge me like that ;-). I didn't want to proceed unilaterally in such a direction: this has to be a community effort, so let's see where the community would like to go first. But I'm willing to help: I have a huge collection of interesting problems posted over the last 15 years and ought to be able to mine it for one or two that are still current ;-). –  whuber Oct 20 '10 at 16:13
    
Thanks Bill, I think you've neatly described our current state. Part of the force directing questions to narrow techincal domains is the strong de-emphasis on subjective or multiple right answer questions. This is something I personally think we can ease up on. I welcome cross domain and cross application questions. –  matt wilkie Oct 20 '10 at 20:45
2  
I have to agree with the original comment. I do find the site has very rigid questions referring to a specific technology. What I think would help is if we can answer such questions with general concepts and alternate technologies w/o being voted down as irrelevant. For example, if a question is "how do I find a point in a polygon using ArcGIS?" perhaps the "using ArcGIS" part should be edited out, and the question opened up to both point-in-polygon theory, and how different packages implement those techniques. Then the question is answered for many folk, regardless of what system they use. –  Mark Ireland Oct 26 '10 at 18:05
    
I would warmly welcome an approach like the one Mark Ireland is suggesting. I think it can lead to a more permanently valuable record and be more generally applicable to a wider variety of people. –  whuber Oct 26 '10 at 19:25
    
Not sure what size minority you might be in Whubber but I totally agree with your answer. As someone new to GIS, but not maps, I wish I felt I could ask the sort of questions you describe. What puts me off? The thought I am going to get blasted because my question is "off topic" or that I will appear to be lazy for not having done loads of research and answered my own question. I might be worrying about nothing...... –  Nigel Law Apr 5 '13 at 9:10
    
@Nigel I am sorry you have developed these concerns, but I can guess why. However, one advantage of this "Balkanization" I described here on GIS is that a tremendous variety of questions is tolerated compared to other SE sites: just about anything goes, provided it's not overtly commercial. I don't see people blasted for posting off-topic or even lazy questions (and IMHO there are an awful lot of the latter: e.g., there are infinitely ways to make a syntax mistake in a field calculation, it seems, and an endless supply of people who have no clue how to diagnose or fix it). So ask away... –  whuber Apr 5 '13 at 12:27
    
Thanks Whuber :) –  Nigel Law Apr 5 '13 at 12:58

As someone that teaches, I know that immediacy of response is paramount and that most questions will generally be software specific. This will never change. What might change is in how responses are addressed. I recently responded to a question dealing on polygon width estimation, and I fell into the trap of which I speak. I provided a software specific solution but failed to provide further information or links to literature within computational geometry and GIS that may have broadened the general discussion or provided food-for-thought to others that may have come upon this thread at a later date. We as individuals, need to contribute to the "broader issues" by providing this information.

share
2  
Most answers may have software specific solutions, but I certainly don't think most questions should be orientated to how to do action X in software Y. –  Andy W Oct 28 '10 at 20:42
    
I agree, but my experience with undergraduates is that the big picture is of less importance which is unfortunate since the generic information may allow them to approach similar questions on another platform. –  Dan Patterson Oct 28 '10 at 22:08
    
I don't disagree with that comment either, but I'm not sure if you can make an answer insightful to questions of such a limited scope. I think we have people participating on this forum who can give deep answers, we need more deep questions. As is the forum is close to tech support for ESRI products. I'm pretty sure over half of the questions have some sort of tag relating to ArcGis, gis.stackexchange.com/tags –  Andy W Oct 29 '10 at 2:12
3  
re: the high prevalence of ArcGIS questions: I believe that's a reflection of two things: 1) ESRI is very nearly a monopoly and there are far more arcgis users out there than everything else combined, 2) the non-esri user communities are healthier, on their mailing lists the people who write the software participate directly with the users, so there's less need of a site like ours. –  matt wilkie Oct 30 '10 at 8:10
    
Dan I agree with the thrust of your suggestion. We should make effort to note and reward those who do spend the extra time to point out how to direct the principles beyond a particular application. –  matt wilkie Oct 30 '10 at 8:14

It seems like we could generate quality content more quickly if we ask really hard questions and then add a bounty.

share
    
I agree that this is a good way to distinguish between general knowledge and expert knowledge. Good questions whether they are generic, specific, hard, or easy should stimulate growth in the community (both by making already participates more active and hopefully bringing in new people to participate). Only a few people in the community as of now have enough reputation to offer bounties though. –  Andy W Oct 15 '10 at 14:55
    
Yeah I suspect the distribution of reputation in gis.se resembles that of wealth in a third world country. en.wikipedia.org/wiki/List_of_countries_by_income_equality What is our gini coefficient? Bounties redistribute reputation. If you know someone who isn't currently a user - but who might be able to anwser the question, sending them a link to a bountied question might attact them. –  Kirk Kuykendall Oct 15 '10 at 16:50
    
Consider it a form of microlending. –  Kirk Kuykendall Oct 15 '10 at 16:56
1  
Nice idea. I would like to suggest, though, that instead of "really hard" we try for "interesting," "fundamental," "broadly useful," and even "unusual or creative." –  whuber Oct 20 '10 at 16:03

Another stategy in the "rising tide lifts all boats" theme: those of us with high reputations could refrain from answering questions quickly, thereby allowing others a chance to get above the 200 mark which has been chosen as significant benchmark by Stack Exchange.

People who do this should make comments in the interim though. This provides feedback to the askers and reassurance the questions are being read. If after a suitable time no real or relevant answers are posted we should still do our best to answer.

share

I agree. Sorry not much else I can say. I'll just have to find some questions to ask now.

share

I'm new to the site, but I have to say I disagree with the notion that anything needs to be done differently here. Perusing the content, I find the range of questions being asked and technologies being used very refreshing. As a Q & A site, I think the bulk of questions are going to be people with very specific problems looking for very specific answers, with the occasional fundamental or difficult conceptual question. But I think that's perfectly fine, as it's reflective of what folks need to know.

I think being anything but warmly inviting of all questions GIS would hurt more than help the growth of the community. The best approach, in my opinion, is to try to embrace all things GIS, so that anyone with a GIS-related question - large or small, specific or general - feels completely comfortable asking it. And I think the site is well on its way to that. Basically, this is a long way of saying that, in my opinion, there's nothing to worry over. What is here is already unique on the web, and the content will continue to grow as GIS professionals and enthusiasts discover the site. You all are building it; people will come.

share
1  
I agree we should not discourage any questions related to GIS. That being said, I disagree that the content of the site is varied. I appreciate your response though, and hope you continue to participate with the site with both good answers like this and good questions. –  Andy W Oct 30 '10 at 22:41

At the same time, I thought it was frowned upon to ask questions you already know the answer to, and then answer them, and mark them as resolved?

Perhaps these type of Qs need to be community wiki? Or are you encouraging ppl to ask questions that they are experts in answering, just to build up peoples rep and the sites knowledgebase?

share
    
I was thinking more along the lines of broader general questions that can be applicable to a wider array of audiences. In that particular case the user was asking very narrow in scope questions that only took some simple troubleshooting on their end to resolve. I'm not suggesting we game the system either, but I think it is perfectly reasonable to ask a question you know the answer to stir discussion. Especially if the question generates knowledge useful to the community. So yes to knowledge base not so much solely for rep. –  Andy W Oct 13 '10 at 12:40
2  
It is not frowned upon to answer your own questions. It is seen as uncouth do so in a rapid fire and with short-just-long-enough-to-say-I-did answers. Generally speaking the community recognizes and rewards the effort put into questions regardless if it's their own or others. Here is a fine example of self answered question: Make the nodata area of a resampled orthophoto overview white? –  matt wilkie Oct 13 '10 at 20:55
    
gotcha. Ill contribute something useful at the weekend. –  Simon Oct 14 '10 at 1:26

You must log in to answer this question.

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