11

You could argue that having at least baseline data is better than having no data at all. I was not here for the TIGER import back in 2007/8, but I suppose the reasoning for going ahead and doing this huge import was something like: ‘What we have here is a vast country where we’re unlikely to get decent contributor coverage anytime soon, if ever for the vast rural expanses between the metro areas. What we also have is a PD street network data source that has a lot of positional accuracy problems, but is at least mostly topologically OK, offers basic road classification and covers the entire nation. If we import this, we will have something to work with. Having a baseline street network available is likely to be more encouraging to both aspiring mappers and data consumers than having absolutely nothing.’

While I am super critical towards data imports, I think that importing TIGER has been the right decision, even though it may not have been perfectly executed. The only way we can improve the map here in the US is by attracting more mappers, plain and simple. We’re not doing a great job at this compared to Europe (http://resultmaps.neis-one.org/newestosm.php) but it’s not going all that bad either. And more importantly, we have a really dedicated core group of mappers who have a really good grip on what needs to be done. After the redaction, the repair started immediately and systematically. See http://wiki.openstreetmap.org/wiki/User:JohnANovak/Interstates, http://apmon.dev.openstreetmap.org/us_routing_grid_12_08_20.html, http://lima.schaaltreinen.nl/redaction/# and http://ni.kwsn.net/~toby/OSM/maps/redaction.html for some tools and resources implemented by the US community to assist in this effort.

So I’m optimistic. Yes, TIGER data sucks, but it’s mostly easy to clean up. There’s no quick fixes though, and you have to believe in the power of the community. We’re probably where most of Europe was in 2008 or so, depending in what you look at. Did they give up because it seemed like an insurmountable task?