[OSM-talk] Big sponsors (was: WolframAlpha uses OpenStreetMap data)
SteveC
steve at asklater.com
Fri Jun 18 05:52:41 BST 2010
On Jun 17, 2010, at 10:46 PM, Ben Last wrote:
> On 18 June 2010 12:32, SteveC <steve at asklater.com> wrote:
> On Jun 17, 2010, at 10:28 PM, Ben Last wrote:
> > Well, fixing it isn't trivial, since there may be a bunch of reasons why it's wrong (or it may not actually be wrong!). We could store these and forward to OSM, but as has been mentioned, we don't want to bounce our users over to the OSM site (because we don't want them to have to be registered with OSM) and sending such comments programatically is not simple. But I do agree that it's valuable feedback if there is the capacity to respond to it.
> absolutely agreed - but the first step is to collect it, then we will respond to it (I volunteer!)
> We can't start arse backward and have a response mechanism and then start collecting it, that makes no sense.
>
> Fair enough :) However, whilst volunteering is impressive, we have to consider issues like this: our site is seen by users as "all NearMap". Yes, we acknowledge OSM in the terms and conditions and the like* but in general the average user doesn't care - it's all NearMap to them. So when they give us feedback, they expect us to respond. If the feedback's being handled by a small volunteer team who lose interest/fall under the proverbial bus/go off on a round-the-world trip, then our users see no response.
Oh totally agreed - my suggestion wasn't 'bounce all feedback to OSM' but either 'the map isn't ours we will bounce it to OSM and try to help' (maybe you can be copied on the bug, or something) or I'm saying that this is totally separate - forget about your customers/users for a second and think what would be good for OSM, it would be good for us to have such a system.
> This isn't in any way to say that volunteer effort isn't the right way to handle this; all I'm saying is that there are other concerns that a commercial enterprise must consider :) Which sort of brings us full circle to the original posts about Wolfram and their thoughts on relying on OSM :)
yay!
> * We're currently revising these pages to make it even clearer what comes from OSM.
>
> > Actually, I think it's the other way around. The biggest issue with OSM for the use cases that concern me right now is that its usability to locate an address is pretty limited. Improving that is the single thing ("step one") that would make it better (again, for my use cases). So having streets with correct names and at least some numbering data helps a lot.
>
> Oh I agree but what I'm saying is that the above case is immediately fixable whereas the latter cases are implicitly fixable, with some more unknown work. See what I mean?
> Ah, yes. Agreed as to the fact that fixing these latter cases means more work!
exactly
so all I'm saying is - fix the easier ones first!
>
> Cheers
> b
>
> --
> Ben Last
> 0423 475 673
> Development Manager (HyperWeb)
> NearMap Pty Ltd
>
Steve
stevecoast.com
More information about the talk
mailing list