Spatially Adjusted

Tag: SQL Server

  • Microsoft Geo

    I still see projects now and them that are spatial. I think of the US Building Footprints project and how they had to give away the data and couldn’t monetize such a project. Bing Maps went through so many name changes that we can’t even recall them all. Heck Microsoft bought Nokia but only the phones. They didn’t buy HERE (Navteq) which could have been a great coup for them.

    Visual Studio Code… Someone needs to check some files into Git.

    I have to admit, I’ve been a user of BBEdit since about 1994, but I’ve found myself using Visual Studio Code much more. If I search my blog posts over the years, there are posts littered with Visual Studio hate. But now I find VS Code to be my go to code editor and not only for programming but also editing GeoJSON files.

    But this really has me thinking. Microsoft and Geo really has died. I’m not saying that SQL Server isn’t used for spatial queries. Or that occasionally I see Bing Maps used in apps. But really they have become such an also ran that I really couldn’t even recall the last time I used Bing Maps API, let along SQL Server (I actually do recall and it was SQL Azure back in 2016). For a company that really has reinvented itself, they have fumbled what little they had in spatial away.

    I’m sure I have a screenshot of Bing Maps, but I didn’t search for very long.

    While at Cityzenith, we dealt with CityNext, which had much sway in the Smart City space, but so like depth. I think it was just an excuse to get their name on Smart City conferences.

    I have to tip my hat to Microsoft for many things, but in our space, they really have become at best a follower, at worst an also ran.

    At least old Gil is trying…

  • Geocoding with SQL Server

    Link – Geocoding With SQL Server

    With the release of the Google Maps API, I wanted to try out some mapping. This ended up being the fairfaxinfo.com project. As nice as the Google API is, it does nothing to help with the hardest part of mapping: getting addresses translated into longitude/latitude. The only free service I could find was www.geocoder.us, but I didn’t want to rely on it being up/free forever. The code is available in PERL with a Berkley DB, but I would much rather chew on tinfoil than try and read someone else’s perl code. So, after much headscratching, fist pounding, key banging, debugging, eye wateringly boring (but thorough) census documentation, and in general driving everyone around me crazy, I have finally managed to use the Tiger/Line census data and SQL Server to geocode addresses.

    I found this during a totally unrelated search for Oracle database backup (kind of makes you wonder how Google works sometimes). Anyway it is a pretty good method of geocoding addresses using Tiger/Line census data and SQL Server.