We must work to understand what good citizenship and ethical behavior means in open data projects. The nature of communication, copying and competition in the space of open data is very complex. Yes, it’s not just about Google, but about raising the awareness of these issues among the people organizing open data projects, and especially the communities where we want to have an impact. The best idea I’ve heard this week (in a week of amazing ideas in Cambridge) was from Jeffrey Warren. We need a clear set of principles and ethics to guide the practice of open data initiatives in new communities. Open data collection should have: open and clear explanations of the purpose of data collection and the license of data; effort to find existing sources of data, rather than replicating and resurveying, and lobbying for the sharing of that data; effort to give the communities that collect data every opportunity to use that data in their own work, however they see fit; etc…
Participation in open data is both about the arcane topics of data licensing, and simply and most importantly being a good citizen. The impression of legal matters is usually all any non-lawyer grasps … it’s just too hard to understand the subtleties of legal text, and we rely on trusted intermediaries to explain what’s what. Google is trusted, and they’re misusing that trust with communities, NGOs, and governments. And Google has huge resources, they get a lot of attention. They’re representing their mapping offerings as free and open, when they’re ultimately controlled by their business concerns. Google has one of the slickest communication practices around, with subtle phrasing that gives an impression of cooperation, but with built in protection whenever someone digs into the details. This is deceptive.
There’s no problem with copying, it’s encouraged. Mapping Parties are not patented. There’s no license on ideas, because fair play is expected. Simply, being a good citizen, being a friend in this diverse space, means giving credit and giving back. Google has taken a concept, adapted it, and not attributed. Not until last year did they even use the phrase Mapping Party for Map Maker events (sometimes that got the attention it deserved). In places where there’s been exposure, Google didn’t “throw Mapping Parties”, because of course it would’ve been noticed as deriving from OSM. And Google is alergic to giving any notice to OSM, for example omitting any mention of the biggest open data project on the planet, which originated in the UK, from their glossy magazine on UK Open Data.
Competition is a very crude and inaccurate way of describing the relationships in the open data space, especially between OSM and Google, and is based on confusion of what OpenStreetMap is all about. OSM is essentially a commons, a data project where everyone can contribute, share and use. OSM is not a company, it’s not an NGO. It’s a collective where hackers, governments, universities, anarchists, slum dwellers, multinational corporations, and international organizations all contribute to a common endeavor to map the world. Every other major internet company contributes to OSM in some substantial way. It is Google only that has decided it’s in competition with OSM, not because of the designs of engineers and product people, but because doing so would undermine something Google thinks it can profit from.
Honestly, I would be thrilled to see Google adopt the use of OSM. The data of OSM is very different from the interfaces and the software. The openstreetmap.org site and tools definitely need improvement, but it’s not because osm.org wants to take away traffic from Google Maps. OSM’s frontpage needs to do something different: “The frontpage fails to communicate what the project does and why it does it effectively. The controls are confusing for new users (lots of random tabs and sidebar). In addition, it totally fails to connect new users to others around them and thus create a community.” Google could even be helping with this.
Never did I say that Google should stop making its products, but rather than it should stop exploitation. In more positive terms, Google must be a better citizen. I even said that a lot of what Google is doing in Africa is good for Africa. I use Google products all the time, live by Google Calendar, and think Summer of Code is awesome no matter how many spaces OpenStreetMap gets. They’ve contributed in great ways to humanitarian response. I have many friends at Google, who are not surprised by what I said, but perhaps surprised that I said it out loud on the Internet. But this is one place where Google’s Africa Strategy has fouled up directly in a space I care deeply about, fueled by business concerns, cloaked in open data clothes, but ultimately dangerous. As Ping said on Crisis Mappers (about a different OSM-Google issue, but equally valid here) …
It is okay to have issues with Google. Google is a big place; there are a lot of different people and different teams here, all trying to accomplish different things. It’s tough to assign a coherent intent to the entire company; we’re a bunch of people trying to work together even though we don’t always agree. No individual person is trying to be evil. Sometimes we screw up, though.
So yea, I’m pissed off, and can’t keep quiet. And actually, as some have asked, there is a lot more to these stories that you’ll only here about over a beer. If you only know me by this post, perhaps you’re surprised to hear that I’m usually known as a mediator, diplomatically open to all sides of an issue, generally a nice guy who just wants to map the world. And in case it wasn’t obviously clear, this all is my personal opinion. Another aspect of our open world is our multiple public representations, and I wasn’t speaking for the OSM Foundation, Humanitarian OSM Team, or GroundTruth. And yes, the OpenStreetMap community has a reputation for being “caustic”, which is based on a very small segment of our community, mostly the osm-talk list. Anyone who has ever attended State of the Map, gone to a pub meet, contributed to a HOT response, posted to nearly any other list, will have a different impression.
Finally, I regret to say that Unosat got caught in the cross-fire of this post. Unosat has an evolving cooperation with many entities, including HOT, where we have been given access to imagery in Japan and Liberia. While I find fault with their Google MapMaker cooperation, they aren’t giving Google any data, rather using Google data. They are well versed in the details of open data, of licensing, and are wise to explore opportunities equally. They’ve generously explored how they can work together with OSM, most recently in Japan and I look forward to more.