Culture: Website: Map
wikimap   (+3, -2)  [vote for, against]
collective GPS data

This is a website where people can upload their GPS data to create collective maps. It relies on the open nature of wiki projects to keep the maps current and updated. You can even upload data anonymously.

You can identify features on map data that you've uploaded. This creates a system of certainty rating. For instance, if 100 people have uploaded a highway they call Route 66, you can be fairly certain that the GIS data you're looking at is route 66. However, if some 'road' you are looking at was uploaded by an anonymous user, winds pecularly, and stops in a lake, and is called 'Main street', you might be inclined to question the validity of the label.
-- lawpoop, Jul 07 2004

Wikipedia: Blocked IPs and usernames http://en.wikipedia...ecial%3AIpblocklist
".. open ... No one is disbarred ... "
If you think you can run a genuinely useful service without banning idiots from your site... well, then running such a service will turn out to be a valuable learning experience. [jutta, Oct 04 2004]

what are the advantages of this over existing maps? whatever they are would be undermined by the questionable validity.

wiki reference sites in general share this disadvantage.

i think it would be more viable to use your system to update an existing map with points of interest tied to comentary from the uploader, something non "mission-critical."
-- xclamp, Jul 07 2004


The advantage over existing maps is that changes (e.g. from road construction) would happen faster than with traditional sources, e.g. yahoo maps. Businesses have a disincentive to keep map data updated, as it costs them money (while they do have an incentive to keep them reasonably updated, or else they might lose users). However with wikimaps, 5 people might upload road changes that they just drove through 40 minutes ago on their way home from work. I don't think any online mapping business could make changes that fast.

Also, I think it would be *harder* to pass off bogus data. Think of it thisway: If you want to get some river mislabeled as a road, you have to get x number of others to go out, create, and then upload and mislabel the same bogus route. On the website, can have filters that would remove identical uploaded data sets. You can also have filters that remove features that have less than, say, 10 people vouching for this. In short, fooling the system would be a *lot* or work, compared to posting false info on wikipedia.
-- lawpoop, Jul 07 2004


It's easy to do things multiple times in cyberspace. It's easy to prove that you are the same person, but it's very hard to prove that you're *not* the same person. This is an interesting issue that a lot of reputation-based systems need to address.

In practice, you can have a human look at updates, and patterns will be apparent - but that's a far cry from automatic. I disagree that a computer can "easily identify random GIS data".

Motivations: Social network sites have a bit of a problem with fictitious characters, which are beloved by many of their real users as a form of creative expression. The IMDB has a problem with users entering their friends as having had roles in movies. Similar motives could apply to geographical features; wouldn't you like to have a street or a mountain named after you?

Maybe it would be fun to go to the other side of this and just allow people to make over the world.

In any case, I'd like a user-updated map, even if it requires moderation.
-- jutta, Jul 07 2004


[jutta] - There is a difference between wikipedia data and GIS data.

The servers on wikipedia can't filter out trollish or false text. It's very easy for idiots to create false entries, or deface a site.

However, an computer can easily identify random GIS data. So a troll actually has to go through the trouble of creating GIS data that seems accurate. Then, with the voucher system, they have to get others to upload similar data (but data that is not identical). So the trolls have more work cut out for them to fool a GIS wiki. Maybe the extra work will make it not worth the effort.
-- lawpoop, Jul 07 2004


This is a really good idea and one that I've been thinking about recently. However, these annotations are useless - could we try not to re-cycle the whole "wikis don't work" argument, peeps? We've heard it all before and hence it's rather boring... thanks.

Creating and maintaining detailed world maps is a massive undertaking, and no matter how hard anybody tries, there will always be limits to the 'freshness' or level of detail. A user-editable map could potentially help overcome both of these issues.

The only potential difficulty might be the user interface; text documents can be easily edited by anyone using a very simple UI, but GIS applications can get quite complex. However, it would still be very useful to have users highlight where the base map does not accurately represent reality, such that the map maintainer can direct his attention toward these areas.

On top of this, a wikimap would have great potential for adding other "layers" of geographically-situated information. More obvious examples include local eateries (perhaps with reviews/ratings), petrol stations (perhaps with prices?), places of interest, etc.

Although rather more complex and almost a seperate idea in itself would be to link this map into public transport information, such that it give directions to the appropriate stop or station as well as planning the journey itself.

It might be fun to also incorporate a system of ad-hoc lift-sharing, whereby hitchers (perhaps with eBay-style ratings) could find car drivers who are making a journey that passes by their location and is heading in the right direction.

Such user-contributed real-time information makes the "map" a lot more dynamic than any single organisation could ever hope to acheive alone. Those contributors who doubt that the wiki concept has anything to offer to mapping are really not using their imagination!

If anyone is working on any similar idea, or wants to start something, or knows of any similar projects, please get in touch.
-- orangejon, Dec 31 2004


Just an update: it looks like Google Maps provides a superb basis for building a service like this, by providing high quality maps and satellite images (US and UK only at the moment, but I'm sure that'll change) and a well-documented API. All that is needed is the database to go on top of it to describe the additional features, or mark inaccuracies.

I'd really like to start this as an open-source project, but I'd need a fair amount of help! Is anyone with me?

Contact me via www.orangejon.com if you're interested.
-- orangejon, Jul 27 2005


check out google maps. they're doing this already i think.
-- jonplackett, Jul 27 2005



random, halfbakery