-------- Original Message -------- Subject: RE: Google Maps/Earth as sources Date: Tue, 7 Aug 2012 07:42:32 +0000 From: Brendan Whyte <[log in to unmask]> To: mapsL <[log in to unmask]> I drew the boundaries for one section of the India-Bangladesh border used by Googleearth. They should be good to a similar scale as the source material I used: 1:63,360 or 1:50,000. They match pretty darn well with the satellite imagery (in terms of rice field bunds), but beyond the section I worked on, who knows what the source material or its scale is. The Burma-China boundary is pretty good in places, but in others it's been badly digitised as line segments, rather than traced from the 1:50,000 maps that accompanied the 1960 boundary treaty. The treaty's 1:5,000 maps of several complicated sections have not been used either by Google. Brendan Whyte National Library of Australia > -------- Original Message -------- > Subject: Google Maps/Earth as sources > Date: Mon, 6 Aug 2012 12:37:48 -0400 > From: Fry, Michael <[log in to unmask]> > To: Maps, Air Photo, GIS Forum - Map Librarianship > <[log in to unmask]> > > > > Some of this has been discussed here before (search the archives for a > 2008 thread entitled "Google Earth miss-matches") but I thought I'd try > to get an update... > > How do you all regard Google Maps, Google Earth, Bing Maps, etc. as > sources for accurate boundary-related data? Aside from several > high-profile mistakes (e.g., http://goo.gl/nP6ev and > http://goo.gl/0gKfH), how dependable do you think Google, Bing, etc. are? > > I understand that Google gets more than half of its boundary files from > the State Dept's Office of the Geographer, so there certainly are places > where boundary data comes from a highly reputable source. But it's not > clear to me how reliable and accurate the depicted boundaries are at > large scales. My primary concern is that users can zoom way in--to > scales much larger than most anything they'd ever see in print--and may > draw conclusions about the boundaries that the underlying data, if not > also the legal documentation, don't actually support. > > Am I right to think that boundaries, at the very least, should be viewed > skeptically, particularly at large scales? > > Thanks. > mf > > -- > Michael Fry > Senior Map Librarian > National Geographic Society > 1145 17th St. N.W. > Washington, D.C. 20036 > 202.857.7098 <tel:202.857.7098> > [log in to unmask] <mailto:[log in to unmask]> >