Report: Wikipedia Losing Volunteers. Is Open Street Map Next?
If you build a system based on volunteerism, what will you do when the volunteers don't want to volunteer any more? Proponents of various community mapping efforts seem certain of the existence of an unlimited supply of volunteer (read 'free') labor. But they should take note of this cnet news article, which shows that a volunteer's passion can be a fleeting thing.
Collaborative mapping and its sibling, Public Participation GIS, are all the rage these days. They appear to be attractive alternatives to commercial or governmental mapping systems. Often built with open source technology using VGI (volunteered geographic information), these systems cost less to build and maintain than their commercial counterparts. They have other advantages as well, such as faster turnaround for data updates.
But the lack of due diligence and quality control in the data collection process casts serious doubts on these systems' usability for anything but the most casual recreational purposes. Factor in the fleeting and unpredictable nature of the volunteer effort needed to keep these systems running, and you have an exotic flower that inspires, but may not be there next time you look for it.
Collaborative mapping certainly has its place in the GIS ecosystem, but it is often misunderstood, and on occasion misrepresented. You get what you pay for. There is no free lunch.
Posted by Atanas Entchev at 11/23/2009 8:57 PM 
Categories: Data,FOSS,Web 2.0,Opinion,Business,Maps,Wikipedia,Online Mapping,GIS,Open Source,OS
Tags: FOSS Open Source Web 2.0 Opinion Business Maps Wikipedia GIS Online Mapping OS Data
Categories: Data,FOSS,Web 2.0,Opinion,Business,Maps,Wikipedia,Online Mapping,GIS,Open Source,OS
Tags: FOSS Open Source Web 2.0 Opinion Business Maps Wikipedia GIS Online Mapping OS Data
Interesting thesis, but not really borne out by any analysis.
The two reasons cited in the Wikipedia piece are "many subjects have already been written about" and "the site has also enacted an array of rules... especially on controversial subjects".
To address the first; unlike Wikipedia, OSM doesn't have a notability criterion. It doesn't need one. The only criterion is "is it there?". So with a near-infinite depth of mapping in any given area, then, the problem of "it's already mapped" - though significant - is less.
Second, rules. OSM famously has very few. Map what's on the ground; don't copy from copyrighted sources. Er, that's it. I write this as someone who enjoys OSM greatly, but has long since given up contributing to Wikipedia because of those rules.
Although OSM's tools could greatly do with improvement, they're streets ahead of Wikipedia's arcane markup language which gets more complicated by the week. The WSJ article says of Wikipedia in 2004: "Everything was a little less complicated." Of OSM the reverse is true. OSM is easier than it was in 2008, and that was easier than it was in 2007.
But more telling, I think, is that you're trying to fit this into a GIS viewpoint. Collaborative mapping never really sought "a place in the GIS ecosystem". If you describe your needs using the phrase "GIS", OpenStreetMap is never going to be for you.
Reply to this
Richard:
Thank you for your comments, all of them spot on.
My thesis was, indeed, the product of more gut feeling than analysis. My main point, which I clearly failed to articulate (Twitter feedback to the same effect) was that large reliable systems are unsustainable through volunteer effort; volunteer motivation is fleeting by its nature.
There are so many municipal governments (my primary market) in my area who do nothing about GIS, expecting the various community mapping projects to do it for them. This little piece, triggered by the cnet article on Wikipedia, is my frustrated response to them.
Reply to this
I also think that the OSM effort is incredibly influential beyond the simple idea of providing free data, and discussing it just on those grounds misses the point a bit. Firstly I don't believe that the UK government would have moved towards freeing Ordnance Survey data without the work of OSM. Secondly I think OSM has contributed enormously to the development of the "neo-geography" ecosystem- there are numerous side projects that only exist because of it.
Reply to this
I would agree with what Richard has said above, with one caveat. As a person whose needs can be almost fully described by the phrase GIS, I find OSM to be a very useful addition to my magic box of data. There are situations where OSM data is sufficient, and others where it is not.
'Quality' issues are being examined elsewhere (Muki Haklay's blog has details here: http://povesham.wordpress.com/2009/11/14/openstreetmap-and-ordnance-survey-meridian-2-progress-maps/ )
In my mind the data quality arguement for VGI centres around fitness of use rather than trust. Of course OSM data has limitations (it has significant advantages too, not just cost) Once these are clearly understood by clients / data users then informed decisions can be made.
Reply to this
I'm not sure I agree with the argument that large reliable systems are unsustainable through volunteer effort. If you take the wider issue of open source software, rather than open data, then the volunteer component is massive, and essential. OK so the larger initiatives have paid-for developers, but even there it is the volunteers that provide the momentum, publicity, bug-fixes and, dare I say, the reason for keeping the project going.
I also think that the OSM effort is incredibly influential beyond the simple idea of providing free data, and discussing it just on those grounds misses the point a bit. Firstly I don't believe that the UK government would have moved towards freeing Ordnance Survey data without the work of OSM. Secondly I think OSM has contributed enormously to the development of the "neo-geography" ecosystem- there are numerous side projects that only exist because of it.
Yes, there are limitations to it's use- and Muki's work on completeness is going a long way to making it a more reliable product, but I'm convinced that we get considerably more than what we pay for with both OSM and open source software when we look at the wider issue, and will continue to do so.
Reply to this
Quality control seems like low-hanging fruit to anyone trying to find fault with a crowd-sourced dataset, but that fruit is sour. When something unusual happens OSM can respond in a way that no commercial map can. When a disaster strikes in the Philippines, OSM volunteers produce a map of what the area *really* looks like to help with the aftermath. When bridges are washed out in extraordinary rainfall in NW England, OSM shows gaps in the highway - other maps still show a passable road. That's contemporary quality.
There are still areas that need a basic level of coverage, but that is happening; in the areas that have this coverage, accuracy and level of detail are very good.
Reply to this
It is clear from the comments (thank you all very much for them, by the way), that my message did not come across as I intended. I will clarify this in a subsequent post.
For now let me state for the record that I am not knocking OSM. At all.
Reply to this
I can understand the thinking behind this blog post. Wikipedia is actually a great example of how this seemly sound logic is faulty. Microsoft's Encarta was around and doing well in 2001, funded by a huge company with proprietary content gathered from "expert" sources. That year Wikipedia launched, with open content provided by volunteers. If you had told me in 2001 that Encarta would now be dead and Wikipedia would be dominant, I would have thought you were nuts. I would have been wrong.
Aside from the value argument (with open data and open source software you don't get what you pay for, you get what everyone has paid for), I think it's a fundamental misunderstanding of what motivates and improves performance in people. I would be a hypocrite for giving anybody a hard time over that meme however - I was convinced of its accuracy until the first time I installed the precursors of Firefox.
Reply to this
I am trying to clarify my position in a new post. Summary: I am not knocking OSM.
http://blog.entchev.com/2009/11/24/relying-on-vgi-for-your-mapping-needs-is-like-relying-on-twitter-for-your-news.aspx
Reply to this
I agree with all mentioned above, but I miss one crucial thing. The volunteer.
Most active volunteers on OpenStreetMap have a more or less tech background. They are, so to say: "The head of the herd". OpenStreetMap its most important strategic target right now is to "bridge the chasm" between the head of the herd and the masses.
This requires:
- Effort to make it easier for non-techies to get on board
- Effort to respond in a more constructive way on mailing-list posts then "read the manual, or DIY"
The strength of OSM is in crowd-mapping. Lets work on making a real crowd.
Reply to this
Miblon is right. The biggest single problem with OSM right now is the utterly terrible documentation.
Reply to this
There are always plenty of issues with large, complex and open systems such as the OSM. I set out to highlight some (fleeting nature of volunteer enthusiasm), you point out others. This is a very healthy discussion.
Reply to this
OpenStreetMap is a large complex open system with (compared to the scale and global coverage) a relatively "small" subgroup of contributors. This should be taken into account because it greatly determines the effect on the community as a whole when people leave.
Reply to this
I think new incentives need to be found, perhaps coupons or some sort of official documentation as proof of volunteer work that could be added to a CV. Otherwise they will simply have to start doing it themselves, as for Wikipedia though, maybe people are actually starting to run out of things to write?
Reply to this
This comment actually makes a lot of sense, even though it is clearly designed to generate traffic to the commenter's website. So it stays.
Reply to this
I'm doing my research papers with the help of reading from different wikis like wikipedia. Contents from Wikipedia is a great help to anyone searching for particular topics or subjects.
Reply to this
All of the volunteers I've met had a secret agenda. Although they did all that work for free they wanted it to appear in their resumes or wanted a letter of recommendation. All and all, no volunteer is doing anything out of pure passion.
Reply to this