On Sat, Aug 21, 2010 at 3:19 PM, Nick Black <nickbla...@gmail.com> wrote: > I would love to have an iPad app to map with. I've actually just come > back from a Mapzen POI Collector mapping trip around a town that could > have really used an iPad to add some roads and other features. But > from CloudMade (or Mapquest's) perspective its tough to justify the > cost of developing an iPad app. Mapzen POI Collector only has 7k > downloads and a few hundred users each month - so the market size for > iPhone apps like this is limited. Sure, make the UX 10x better and > there could be thousands of users a month, but its still a limited > number of contributos for the effort put in. Compare that to the 2.5M > Foursquare users or the 60M Farmville (10% of the total number of > Facebook users) users and the number is tiny. Then consider that > there are something like 11M iPhones and only 3M iPads in the US with > even lower iPad penetration in other places, and the market is even > more limited. > > A possible answer is HTML5 apps - that's what we're looking into at > CloudMade at the moment. An HTML5 POI collector, for example would > let users on iPhone, Android and iPad and other tablets join in the > party. (There are now 8M android phones in the US) It could even be > packaged into an app on app stores to make it discoverable. The apps > should be focused on doing one or two things well. Does the world > need another fully featured editor? IMO, no. It needs a suite of > tools that each make it deadly simple to do a couple of things. So > the feature set is going to get even more limited ;-) > > The next problem though is the terrible conversion rates that we see > from download to active mapper (see > http://lists.openstreetmap.org/pipermail/talk/2010-August/052476.html). > This is caused partially by some poor design decision on our part at > CloudMade, but primarily by the constraints of the current OAuth > system as discussed in the thread I referenced. Unless we fix the > problems with the OAuth sign-in / sign-up process the number of new > mappers and tools like this could attract will be severely limited. > Like I said last time - I'm keen to find a solution to this problem > :-) > >
My suggestion to this problem is fairly simple: stop trying to make map editors :-) I've tried mapping on my Android phone, and it sucks. This isn't a fault of the editor as such, the screen is small, I can't easily get stuff precise, it's far too easy to break existing stuff, and it's really slow compared to using a mouse and a decent sized screen. On the other hand my phone has a GPS, a camera, an audio recorder, a video recorder, a way to sketch diagrams, and a way to enter text.... it's all my mapping tools in one handy gadget. Maybe what the millions of people contributing to foursquare are after is a way to just take a few photos, stick them roughly on a map, and stick on a note that the road is oneway in the other direction to what we have it. ie: they probably don't want to learn how to map, or figure out why the street happens to be in 3 pieces and that reversing the oneway means altering the cycle route in the opposite direction -- all of which an editor could make easier, but all of which would make me give up if I was just having a play. Such an app wouldn't need a user account, OAuth, or any other barriers. A nice website to allow downloading the contribution for adding to OSM and integration with an editor or two might encourage some of those people to join us properly. I think we'd be much better off with a bunch of tools designed purely to collect data. Or at least, I'd like this for my phone if anyone gets round to writing it please! None of which is talking about the iPad (sorry Steve), which as it has a bigger screen might make for a fairly nice clipboard replacement. I doubt I'd go mapping with one though because it's still quite heavy. Dave _______________________________________________ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
