Location Inconsistency and Geotagging

Just today, you’ve probably read a handful of news and blog articles that discussed a place, whether that be a home, restaurant, or park. The author may have described the place in detail with a street address or vaguely with the neighborhood it’s located in. Either way, they’ve done their job to describe its location to you, a human. With the current growth of location-based applications, the description of locations on webpages must go a step further, to be accurately understood by code.

In 2012, a Pew Research study revealed that nearly 75% of mobile users use real-time, location-based applications and this number has likely risen. These applications — such as Waze, Yelp, and Google Maps — work with data that relies heavily on addresses, latitude, and longitude. Because of this level of organization, each of them are able to perform extremely useful, profitable tasks. Before this organization was done online, features like directions, maps, and local notifications were simply not possible.

In this post, we analyze current efforts that help define locations for code, the lack of consistency between them, and the opportunities that lie within a consistent solution.

How location is inconsistent

Tagging a webpage with a location has already been accomplished using many different solutions and is often called geotagging. Now when I say “many different solutions”, I don’t mean that in a good way, it’s actually the problem we have at hand.

Geotagging doesn’t mean writing an address or city somewhere in an article, but actually requires specifying details about the location as a key-value pair. Geotagging is often done this way using metadata tags, which are HTML tags that describe something (i.e. title, author, image) but are hidden when viewing a webpage. Metadata tags contain useful information to help online applications, such as search engines and social networks, accurately understand and present the content from the webpage.

Alternate solutions to geotagging include the HTML <address> tag or Google’s My Business tool but aren’t sufficient to be a universal solution for all websites.

Here are a few metadata solutions we found from popular, location-focused websites:

  • Facebook’s Open Graph
     In 2010, Facebook launched Open Graph, which gives websites a specific format to follow in order for the social network to decipher its contents and display nicely to users. This initial launch included a tagging format for location but was replaced shortly after with a more specific format for only business locations. Currently, Facebook still enforces this business approach but it’s not used on the web as much as it should be. Surprisingly, the code behind Facebook, Yelp, and Wikipedia don’t even use this Open Graph schema, or any schema, to describe exactly where a business is located.
  • Foursquare
     Foursquare uses metadata to describe the location of businesses but their schema was specifically created for their website only (playfoursquare:location:latitude, playfoursquare:location:longitude).
  • Groupon and Living Social
     Groupon and Living Social use a more generic metadata solution to store the latitude and longitude coordinates of a business or deal offered on a webpage. However, both use a different metadata schema.

What we see here is the implementation, or at least support, of metadata but none of them are consistent with each other — not one. With all the “hype” around hyperlocal in recent years, you’d expect there to be a consistent standard already, right? We’re surprised too.

Unfortunately, despite having a great local tagline, using geographic hashtags, or writing about local topics and businesses, websites are actually still missing out on a “code connection” with location-based applications. Without implementing geotagging and without a universal geotagging schema we are leaving these applications with no definition to follow, lowering the ability for innovation, and shoving our own web content away from “local” opportunity.

How location can be consistent

With the rise of interest in online local content, techies and web authors need to start working towards more efficient organization and specificity of location data. Since the big social and business networking websites can’t decide which one to use and there’s no known effort to make a consistent schema, Bloom has decided build the bandwagon.

Inspired by various sources, we’ve come up with a simple yet universal schema called Geo Metadata. The schema contains three metadata properties:

  • Latitude: geo:latitude
  • Longitude: geo:longitude
  • Address: geo:formatted_address

The properties can accurately store any location from any website and can be accurately read by an application with just a few lines of code. For example, the webpage doesn’t necessarily have to be an office page for a business, it can be a community page about an amber alert, an event page, or a news article talking about a traffic report.

With the introduction of this schema, we are now specifically advocating for it to be the standard, universal solution on the web. We have already implemented Geo Metadata on each article page on Bloom and are experimenting with its usage across our plugins. Learn more about installing Geo Metadata on your website.

Benefits of location consistency

We believe there is a lot of potential with the implementation of metadata. With the acceptance of a standardized schema such as Geo Metadata, it will bring opportunities into technology that are impossible to see right now. Think of how hashtags have made it easier to connect content on Twitter, Facebook and other social networks. With consistent data for locations, we could see the same type of improvement but on a much more global scale — no pun intended.

Here are a few ideas to get your brain bubbling:

  • Local search ranking. If you search for “Obama” on Google but live in Houston, it could prioritize a local campaign event webpage over a white house press release.
  • Local priority on social networks. Imagine if your Facebook news feed became locally driven and prioritized a recent, nearby link over a cat-in-a-box video. Think of how jealous that cat would feel.

In a perfect, consistent world, these ideas would be easy to accomplish. Real estate listings, events, businesses, and other types of content could easily be searched, filtered, and promoted together simply by their metadata location. Knowing how inconsistent location currently is online, this perfect world may take some time to get to but we have to start somewhere.


Originally published at www.bloom.li on September 22, 2015.