Viet Nguyen sipped his coffee as he scanned the weather outside his window in Madrid. There were no ominous clouds. It had been a bad week and he wanted to climb. Then the email came.
On the morning of Feb. 24, Nguyen received a cease and desist letter from onXmaps, Inc. (onX), the parent company of Mountain Project (MP). The notice alleged that OpenBeta, Nguyen’s not-for-profit project which provides open source climbing data, had infringed copyrighted works held by the business. A few days later, onX filed a Digital Millennium Copyright Act (DMCA) takedown notice to GitHub to remove a data repository uploaded by Nguyen, effectively shuttering the project. Then MP disabled Nguyen’s account, which had been active since 2012.
“Ironically, the climbing community rallies behind the Access Fund to fight for open access to our public lands. Yet as a climber one has limited options when it comes to getting access to our own climbing data other than being able to browse [the] MP website,” Nguyen wrote in an open letter posted to MP.
In response to the notice, Nguyen contacted the Electronic Frontier Foundation (EFF), a leading digital rights nonprofit. On March 25, the EFF, representing Nyugen, sent an open letter to onX asserting that the “claims are meritless” because he’s using factual data—route information such as names, location, height, etc.—and that he can do so under fair use laws because it is noncommercial and educational. They shared additional details in a blog post.
“The people who FA the route, [the developers who] make the topo and the description, they share the information that has been passed down since climbing started and continues to today. It would be a travesty to see that locked into one platform by one company who would have ultimate say on what we can do with that information,” Nguyen says. “Can we make that a more collaborative effort?”
Nguyen started climbing ten years ago, around the same time he began a new job in the enterprise open-source software industry. “The idea of sharing your work and the open culture of climbing, the two just clicked together,” he says. Nguyen started OpenBeta, a not-for-profit educational initiative to promote open-source software in the climbing community.
“I spoke with many people who made a mid-career transition into coding, and they go, ‘I want to build a climbing app!’ But there’s a blank slate, nothing to build on top of.” Nguyen compiled a dataset, created tutorials, and shared coding examples to help programming gumbies practice machine learning and data science tools like pandas, an open source library in Python.
The crux is that Nguyen populated the dataset with information from MP.
Mountain Project is the most comprehensive database for climbing information in the U.S. In 2015, the site featured more than 125,000 routes and had over 3 million users for the year. By 2020, the number doubled to more than a quarter million. “Thanks to the help of thousands of contributors, we’ve built something amazing,” notes Co-Founder Nick Wilder in the thread Mountain Project’s Next Chapter. Almost all of the content has been uploaded by users since the website was launched 16 years ago. OpenBeta was an attempt to give back to the climbing community by building on the crowdsourced data.
onX made their position clear when Chris Hamilton, VP of Product Management, shared, “[o]ur approach to licensing will remain consistent as it has been in the past. When users submit their adventures, they do so knowing that they are sharing it with the MP community, and not for redistribution.” Users didn’t expressly agree to share their data with OpenBeta.
That’s where Nguyen ran into problems. Or is it? It’s complicated.
What Are Facts, Anyway?
onX asserts that Nguyen infringed copyrighted works. The EFF argues that Nguyen is only using facts, which “are not protected by copyright.”
Jeff Smoot, a climber, guidebook author, and attorney who is not involved with the dispute, views inherent route data—things you can see by looking at the line itself—as fair game. This includes information such as location, type of route, number of bolts, etc.
Within a guidebook, or on MP, the history of a route’s development, anecdotes from the first ascent, or beta shared in the description would be considered proprietary. And, Smoot adds, for liability reasons, it’s unwise to publish climbing route information without conducting independent research.
Mountain Project says that they do “not knowingly publish (or permit individuals to publish) copyrighted materials without permission,” and that contributors are responsible for the accuracy of the data they upload. The website doesn’t require a citation for new route submissions though, and area and route pages rarely link to sources.
For Nguyen’s part, he says he compiled, cleaned, and uploaded factual data to GitHub. He contends he did not copy and paste the representation of information from MP, or as EFF states, “any material authored by Adventure Projects or onX.” In an email, onX asserts that Nguyen, included “user ‘star’ ratings, descriptions, and instructional need-to-know information,” which they say “includes significant editorial, opinionated, and experiential information.”
Fair Use or Foul Play?
In the United States, Fair Use law is a defense to copyright infringement. The doctrine allows the use of copyrighted material without permission in some cases, such as for educational purposes, news reporting, criticism, commentary, parody, and research.
As our lives increasingly function online, fair use impacts us more and more. Last year, the Senate Subcommittee on Intellectual Property held a hearing titled, “How Does the DMCA Contemplate Limitations and Exceptions Like Fair Use?” Sherwin Siy, the lead public policy manager for the Wikimedia Foundation, shared in his testimony: “Fair uses aren’t rare exceptions… [it] ensures that our everyday activities aren’t constantly infringing copyrights.”
According to the EFF, OpenBeta is a noncommercial project that is educational in nature. The initiative uses material “to contribute to the general knowledge of the climbing community… [and which] help[s] others to learn about machine learning, making climbing maps, and otherwise using software to generate new insights.” The EFF also notes that the usage of the user-generated data is transformative, a type of case that is “likely to be fair use.”
Contentious Content: Who Owns the Data?
You might be surprised to know that content submitted to Mountain Project is licensed to the company for use more or less as they please. Forever. A user agrees to let the company:
“Use Your Content in a number of different ways, including copying it, publicly performing or displaying it, reformatting it, modifying it, translating it, incorporating it into advertisements and other works, creating derivative works from it, promoting it, distributing it… As such, you hereby irrevocably grant us world-wide, non-exclusive, royalty-free, sublicensable, transferable rights to use Your Content for any purpose.”
This is the tradeoff a contributor makes when sharing information on the site. More content is useful for both the community and the company, though in effect, the data becomes part of MP’s product. The cease and desist letter provides insight into the company’s thinking. The notice has not been publicly disclosed, however the EFF shares that onX declared they “own all rights and interests” to the user-generated work. In a post on March 25, onX clarified, “Mountain Project is a perpetual licensee of the data, but not its owner.”
While content is uploaded to MP by the community, contributors do not get to choose how the data is used or where it goes. Should the company decide to put the content behind a paywall—they’ve said they are unlikely to do so, multiple times—there is little recourse users could take. To remove content, users can flag items for copyright infringement and contributors can delete photos, routes, and areas they’ve uploaded. This doesn’t guarantee they will be permanently deleted from the site. Some users allege that routes they’ve submitted and then deleted re-appeared under “Orphaned User,” who lists 6,282 contributions. The terms of service explicitly state, “once published, it may not be possible to withdraw or remove it.”
In the future, MP is likely to release a paid app with premium features. This is how onX’s apps work: they offer a tiered service with a free option, a premium tier priced at $29.99 per year, and $99.99 for the Elite version. One primary difference between MP and onX’s other products is that onX didn’t crowdsource the data they monetize.
Mountain Project isn’t strictly a labor of love, of course, and the employees and owners have been compensated for their efforts. In 2015, Wilder stated that the company had seven employees and was breaking even, with all revenue coming from ads. Since then, the user base and quantity of data has grown. In 2018, onX raised over $20 million in investments, led by growth equity investor Summit Partners, which likely provided the funds for the MP acquisition. As a user, you didn’t agree to share your content with REI, who purchased MP in 2015, or onX. But Terms are terms.
In his open letter, Nguyen suggests that if Mountain Project unbarred their climbing data it “would enable endless possibilities for innovations and more importantly greater improvements to the existing data set.” He encourages a Share-alike copyright license which doesn’t operate in the traditional sense, but rather makes sure the work stays in an open collaborative spirit for all to build on and add to.
He continues, “for example, someone could use the data [to] build a better search engine or another person could make maps of all [the] 5.7 trads in Red Rocks and in the process improve accuracy for [the] crag’s GPS and approach trails. In the end onX would continue to benefit from all of these improvements. It’s a win-win.” If MP was structured like an open source project, the net-net is a continual building up of the data and functionality for everyone.
Mountain Project has benefitted from open source software. As of 2018, the site was using Mapbox GL JS to enhance their maps. In an ironic twist, Mapbox announced at the end of 2020 that it would no longer be open source.
Another option is to provide access to the data through an API (Application Programming Interface).This would allow developers to use a limited amount of MP’s data to build and iterate upon, though it would not offer as many opportunities as a project like OpenBeta—users would be restricted to only what the company provides. MP had an API, but the service was quietly closed to new users after the acquisition by onX.
Climbing is a sport that is built on the sharing of information and access to land. Since 2005, Mountain Project has been a community resource that supports these aims, helping climbers explore new areas, share beta, and make connections the world over. As a hub for crowdsourced information, the company demonstrates both the value and limitations for contributors to the site.
For Nguyen, OpenBeta is a passion project built on the collaborative spirit of climbing. “As a member of the community, I believe we should have the freedom to study, improve and re-share data so that the community can benefit from our work,” he says in a blog post.
For now, the climbing data repository on GitHub is still blocked. Access to data, like climbing areas, is ever precarious.