UKH

Noticed this on a tech forum (HN).

As far as I understand it, MountainProject [mountainproject.com], is a US climbing website with lots of route info posted by users - similar to UKClimbing or CamptoCamp). OpenBeta [openbeta.io] is a site that pulls information from the MountainProject servers (via an open API I presume), and allows users to create new statistics, lists etc out of the data.

Mountainproject has sent OpenBeta a cease and desist letter claiming that they are infringing on MountainProject's copyright (on user posted data). OpenBeta is being supported by the Electronic Frontier Foundation (one of the main US open source advocacy and lobbying groups).

https://www.eff.org/deeplinks/2021/03/free-climbing-rock-climbers-open-data....

I don't have a particular axe to grind here, but I am wondering who owns the data, what legal fair use is, and what people think reasonable use of the data would be? I can see that on the one hand, there may be a lot of effort that goes into curating the data that is entered by individual contributors. On the other hand, they may own their individual contributions. I'm not sure where the balance lies.

Thoughts?

Steve

 

Reminds me of that story about Trivial Pursuit - apparently lots of the questions were taken out of one book of facts, but the writer of the book had slipped in a couple of untrue ones to try and catch out people plagiarising his work, and sure enough one of them turned up as a Trivial Pursuit question. 

He took it to successively higher courts, but in the end lost as the judge ruled you can't copyright facts.

Not sure if that applies here, but it's a cool story anyway.

It's an interesting project, I suspect they don't have a leg to stand on in copyright terms. Not least that they try to claim that all of their (openbeta) content is licensed under Creative Commons, which could be true for the presentation, but not for the content they're stripping. Those users have given a licence to Mountain Project to use the content, no-one else. The users still own the copyright to their contributions to MP. OpenBeta also imply that they think they can use the data if it's "Not for profit" which is nonsense.

I don't actually understand what's useful about it, but it would seem ridiculous as a user to think you contributions and licence can simply be pinched by an unrelated website. It seems very naiive to imagine that Mountain Project would do anything other than try to protect the data they've collected.

That said, from what I can see it doesn't look like the way the data is used is in competition with MP. Rather it gives a way to collect and process the data. Doing that legally might not be simple, but it could be something MP could work with rather than against.

My understanding is that this wasn't done using an API. It wouldn't really make sense to create an API for others to use and then object to someone using it. Most likely is that the data was 'scraped' which is a means by which you can download other people's databases using software without their knowledge or consent.

Assuming this is what happened, then I don't think the OpenBeta.oi has a case or at least they wouldn't under EU law (and hence I am pretty sure current UK law). This is due to The Copyright and Rights in Databases Regulations 1997 (boring link here - https://www.legislation.gov.uk/uksi/1997/3032/contents/made ) which assigns copyright to the effort of creating the database in addition to the data itself. Hence even if you have mainly user-contributed data, you retain copyright over the database structure which I suspect scraping it and using it without consent would breach.

Having said that, this law is relatively untested and there are certainly ambiguities. A list of route names is likely to be the same wherever it appears so you can't really claim that as a 'database'. However, anyone who works with databases knows that they will have their own DNA owing to the way the data is structured and it is usually pretty easy to identify data that has been scraped from your database.

Whatever the situation under US law, I think the defence that OpenBeta.oi are using is a poor one and obviously isn't even recognising the existence of Database Law. The idea that because data is uploaded by users means it is open season for all the data ignores many other factors. A decent percentage of the data on UKC Logbook was originally uploaded by us in order to provide the initial framework. We also spend lots of time moderating and maintaining the data along with many excellent volunteer helpers. We also pay for the database hosting and the design and building of it. That isn't to say that UKC claims ownership of the data in UKC logbook, however, it is also not for someone else to claim it and use it for their own purposes even if they tag it as 'open source'.

Alan

Exactly. It also ignores the point (as Alan has alluded to) that MP provided the platform and serviced it at considerable cost to collate the information which OpenBeta are now skimming. Let someone else do the work and pay the bills, is this the new way?

But you can! Look at Lillian Virginia Mountweazel a fake entry (and corresponding court case) that coined the term Mountweazel for an added entry to catch plagiarists  (see https://en.m.wikipedia.org/wiki/Fictitious_entry and section called Terminology

Same goes for maps - cartographers will often put deliberate errors in their maps to see who's copied them.

That explains how I ended up in East Croydon.

I've had a look at openbeta and interestingly they mainly include factual information in their dataset (route name, grade, lat+lon etc.) which I suspect plays to their favour, as this means their database is closer to a list of publicly available facts (as they're not including route descriptions or similar).

On the other hand they also include a crag identifier which I think was a poor choice on their part. Sorting climbs in to crags and buttresses is obviously not trivial and you could argue it was a creative work done by mp (depending on how the site works, Im not sure).

My grandfather apparently had a stint doing this for the navy, with the rather fabulous (though possibly apocryphal) title of 'map spoiler' - though I gather it wasn't so much to catch out plagiarism as recognition of the fact that you can't copyright reality so small discrepancies had to be introduced to make the maps demonstrably distinctive

This is precisely the point. It is the effort involved in creating and maintaining a database that counts here not the individual bits of data.

Climbers need to be careful assigning specific data to a certain publisher even the erroneous hidden inaccuracies mentioned above. If you suggest these errors belong to a certain guidebook/publisher then you are trying to suggest that the correct data does as well. This is a dead-end for guidebook writers since the best guidebooks will use and reference all available sources and try and improve and enhance the information base.

For example, a certain guidebook includes a route description with a 'left' where it should be a 'right'. If a new book by another publisher reproduces that error then that is not great practice but it doesn't show wholesale copying, it just shows that the other book was used as a reference. It also can't be plagiarism if the new book includes a list of its reference sources. Sadly UK guidebooks have not been great over the years at doing this but, even without the reference sources, it isn't something we should get too overheated about. If climbers really want guidebook writers to go back to the original source material every time, and climb every route between editions, then they will be waiting a long time for new books and they will be rubbish books.

The issue here is all about the effort and resources that are required to create a dataset. That is  what Openbeta.oi appears to have used without first getting permission. This is wrong on many levels not least simple good manners.

Alan

It's a difficult one and quite nuanced as there are a lot of similarities with the first rejection of google search and google news services. 

If I can use google to see the top of the headline for each news item, they are essentially stealing content from the news agencies. Likewise, any website that gets indexed by google. 

The same goes for the Google book project which people have found incredibly useful but was being blocked at some point. 

Personally, if the tool is non-commercial and is being used to direct people back to Mountain Project but is adding a layer of additional search capability etc. then I don't see a problem. 

However, OpenBeta.io is just providing a caching search system in front of mountain project. They show a couple of example tools from what I can see but don't actually directly reproduce the end data (at least not that I could see). 

Personally, I think climb information should be considered 'open'. It's nearly always been supplied in good faith to be shared (I don't know many people who have licensed the information from their first ascents) and most outlets, guide books, websites, etc. all reuse this information in some way. I'm fairly sure Mountain Project managed to get some information copied from guide books at some point (who themselves copied information from other guide books, other climbers or directly from the first ascensionist). 

Do we want to be in a situation where some 'controls' the information provided freely and in good faith by the climbing community?

Tim

p.s. Perhaps the correct answer is that Mountain Project should share some of the raw source material they have gathered with a website such as OpenBeta in order that the community can create the tools it wants/needs

I think it's worth bearing in mind that these are opening shots in a dispute.  Both sides are probably guilty of overstating their case.  At the moment these are just claims and we don't yet know whether they can be substantiated.

EFF philosophy seems to start from the position that copyright is often controlled by powerful interests who use it against the interest of consumers.  That may sometimes be true, but it can often also be the case that copyright is owned by small operators. This is particularly true since the internet made it possible to publish without going through the traditional gatekeepers. The open source philosophy that everying should be freely available nearly killed off the music industry, and it was the small independent labels and musicians who came off worse (and who still receive only derisory payments from Spotify and other legitimate streaming sites).  The large music publishers are still doing well.

I see a parallel with climbing data.  It was obviously in music consumers' interests not to have to pay to listen to music, and it is obviously in climbers' interests to have access to information about climbs.  However if that undermines the efforts of those who provide that resource, whether its music or information, then everyone loses.  This clearly goes beyond "sharing beta" which EFF claims support its case, morally if not in law. A balance has to be found.  

I've wondered about this sort of thing on occasion.  It wouldn't be especially difficult to stick some restful webservices in front of the mountain project DB (or other such websites, if you get my drift).

Isn't going to happen though.