From: Kai Krueger [mailto:[email protected]] Sent: 15 April 2011 02:17 To: [email protected] Subject: Re: [Talk-GB] Maxspeed tagging for the UK
Richard Fairhurst wrote: > > Why are we doing this? > > In OSM we optimise for the mapper, not the data consumer. That means we > tag exceptions, not majorities. > +1 / -1 (Yes and no) It has to be a compromise with which both sides can live with, mappers and application developers. If a tagging schema is too complicated for mappers to add, but easy to use for data consumers, it is of little use, as the data won't get added. But vice versa, if the data is easy to input, but too complicated for data consumers to ever feasibly use, then that is also fairly useless. It will just remain dead data, filling up a database. Both sides need to be adequately represented in the thoughts of what good tagging schema's are. In most parts of the world, mappers are the limiting resource, so a good compromise will sway towards being easy for the mapper, but it still has to be a compromise. Unfortunately if we thought anything other than the simplest way for mappers then OpenStreetMap would never have got off the drawing board. I agree in an ideal world the data would be perfect for entry AND use. But we don't live in a perfect world so we must look at it differently. Data users need more tools to help them sanitise and make the data more applicable to their needs, especially if those users have limited skills or software/hardware platform, but it's post processing tools and alternative data formats that will help overcome the inherent disadvantages of not realistically being able to ask mappers to do more complicated data entry. If I am not mistaken, you your self have said that you would rather use Ordinance Survey data then OpenStreetMap data, despite being an absolute OSM enthusiast. And if I remember correctly, this was not only due to licensing, but also because of ease of use? But that aside, this particular issue has actually been the mappers who want to note down national speed limit tagging and has not been a "request" from data consumers, as far as I can tell. Mappers wanted to distinguish between "no explicit speed limit/ national speed limit" and "no one has surveyed the speed limit, so I need to go there to survey it". At least that is what I remember from the very long discussions on just this topic on talk-de. This is why assuming something on the absence of data is much more reliable than assuming the data is correct. As more gets mapped the data becomes less complete because different mappers map different things in different ways for any given area. Relying upon all roads to carry the appropriate tag is far more dangerous an assumption that assuming something where no tag exists. Cheers Andy
_______________________________________________ Talk-GB mailing list [email protected] http://lists.openstreetmap.org/listinfo/talk-gb
