Gowalla Leads in Openness, to Offer Place Data Download This Year

Daniel Luxemburg
Mar. 25 2011, 12:00AM EDT

Gowalla, the location-based social network, is beginning to establish itself as the most generously interoperable of its check-in–offering peers. At a SWSX panel last week about integrated location data, Gowalla co-founder and CTO Scott Raymond indicated that developers would soon even be able to download data en masse from Gowalla for their own manipulation and analysis. When asked by the moderator, Programmable Web Executive Editor Adam DuVander, "Can I download Gowalla's data," Raymond responded that providing the option was "definitely on the roadmap."

Offering this sort of full access would be no small thing for a company whose value is tied to the location information it accumulates from its users, but Gowalla already offers features that demonstrate its tendency towards that sort of openness. Gowalla fully acknowledges that its users are interested in other check-in platforms and facilitates participation on those other networks. Since the release of Gowalla 3 in December, users have been able to link their Gowalla accounts to ones on Twitter, Facebook, Foursquare, and Tumblr. Allowing users to post their activity to social streams is nothing new, but for the two other check-in services Gowalla provides integration with—Facebook and Foursquare—users are actually checked-in to corresponding venues on those competing services.

In order to accomplish this, Gowalla's own location listings have to be associated with those in other databases. According to Raymond, Facebook handles this on its own by crawling Gowalla's listings and creating a corresponding location page when a user checks-in if a matching one doesn't exist (although this seems to result in some duplication). Foursquare is a different story, and requires Gowalla to map a Foursquare identifier to its own listings. Raymond explained that the company goes as far as to handle this matching manually by keeping a queue of ambiguous cases for human resolution.

Gowalla already makes these hard-won Foursquare IDs available to developers through its own API. Moreover, when DuVander asked whether Gowalla's teased data export capability would "include the knowledge you have of other services and their IDs", Radymond said there was "No question" the company would so so.

Gowalla integration pins

This is all in contrast with the approach in this area recently announced by Foursquare, Gowalla's larger and more visible competitor. Earlier this week we posted about Foursquare's announcement of the company's "Venues Project", at attempt at "building a comprehensive Venue Harmonization Map". The effort is a recognition that there is an abundance of location-focused information available but that much of it isn't organized well for cross-referencing between sources. The announcement post on the company blog explained that the "goal of foursquare’s Venue Harmonization Map is to translate between these databases, making it easier to create mash-ups."

While Foursquare's Venues Project brings in new types of information about locations from publishers focused on creating valuable content, it doesn't do much in the way of providing linkages to other existing players in the online location space. In fact, much about Foursquare's attitude towards external integration looks somewhat less open—or at least more self-focused—when compared with Gowalla's. In our post about Foursquare's Venue project, we mentioned that the service offered a special "Platformer" badge during SWSXi to reward users who checked in to Foursquare using three different third-part clients. It's good for the developers of those applications, but also a wise move by Foursquare in terms of supporting the growth of an ecosystem around its own API.

Gowalla takes it a couple steps further by offering rewards that support other company's platforms, not just their own. Users receive a Gowalla pin for their first cross-platform check-in on each of their linked accounts (pictured).

But the differences are hardly just superficial. Foursquare's documentation page for its Venues Project includes very specific rules about what developers may and may not do with the company's data in each of the three general use-cases it imagines:

Use foursquare as your location database

If you’re an application that lets users tag photos or other things with places, or a game that uses real world places, or a service for browsing trending or interesting places, you’re free to rely wholly on the foursquare API, provided you are linking to foursquare, providing attribution, or offering the ability to check in on foursquare. It is okay to keep caches of foursquare data as long as they are refreshed regularly. You cannot combine this with other location databases or export your own location database that violates these terms.

Correlate foursquare IDs and data with your existing database

If you’re a magazine, listings web site, or something similar that wants to find the foursquare venue IDs corresponding to your own places so that you can add “Add to foursquare” buttons, links to foursquare, or other foursquare widgets on your site, you can freely use our API to collect identifiers or display foursquare information related to places you already know about, provided that foursquare receives attribution for any of its data. You may not use the API to to add new places to your database or alter location details for places in your database.

Incorporate foursquare data into another location database

If you’re a service that wishes to combine foursquare data with your own or third party location databases such that you learn about new places or alter place metadata based on foursquare, please email api@foursquare.com, and we’ll work something out. The same goes for collecting anything beyond a simple cache of our API results, for example longitudinal data about checkins to locations, or redistributing foursquare data via your own API. (Emphasis in the original)

That's a far cry from the full exportability proposed by Gowalla. And, even though the smaller company allows Facebook to index its data, Facebook's rules suggest it's not inclined to return the favor. According to Facebook's "Statement of Rights and Responsibilities":

3.2 You will not collect users' content or information, or otherwise access Facebook, using automated means (such as harvesting bots, robots, spiders, or scrapers) without our permission. [...]

9.2.8 We can require you to delete user data if you use it in a way that we determine is inconsistent with users’ expectations.

9.2.9 We can limit your access to data. [...]

9.14 We do not guarantee that Platform will always be free.

To be fair, the terms for Gowalla making its data available for developers to download may come with any number of strings attached as well. And for all the cross-platform integration Gowalla seems inclined to support, it's not bringing in new data sources the way Foursquare is, just building closer links between the existing big players in the space. Gowalla's own user population is smaller than those of any of the services it provides connections to, so openness is also more clearly in Gowalla's strategic interest because it provides the company with wider relative exposure than would be the case for integrations in the other direction.

However, the spirit of the approach is still markedly different. On last weeks panel, Raymond said that he wouldn't want to trade Gowalla's data set for one seeded with a more thorough listing of places because "our place database really reflects our community." Leigh Dodds also put it well in a post over at the Kasabi blog that looked in part at our coverage of Foursquare's venue harmonization work:

Sharing identifiers and linking between datasets is useful not just because it helps any individual dataset owner become the “Rosetta Stone” for their specific domain. It’s useful because we live in a Long Tail world.

As a data provider, no matter how much energy you put into curation to make your data more comprehensive there will always be some additional external data, some additional context, that can add value. That value may be incremental to the majority of users, but it will be important to someone.

This attitude leaves a lot more room for a wide range of players to thrive in the location space than one that assumes it's possible or desirable to make a one-stop-shop for venue information.

Daniel Luxemburg I'm Daniel and I'm a web developer and social media consultant based in Brooklyn with a particular interest in how open APIs are making it easier to innovate on the web. I started contributing to ProgrammableWeb in November '10.

Comments