WURFL and Database Copyright


When a phone's web browser visits your site, how can you tell what capabilities that phone has? How can you work out its screensize, whether it can play mp3s, or know if it supports a particular bit of JavaScript?

It ought to be possible using a mixture of UAProf, accept headers, and media queries. But it's not. The data are inconsistent and unreliable.

Out of this frustration, a number of databases have been developed to track the capabilities of as many devices as possible. For the longest time, the most popular and accurate was WURFL - the Wireless Universal Resource FiLe.

Developed over the years by a community of volunteers, and lead by Luca Passani, it was an excellent tool. Mostly accurate, highly useful, and generally well regarded.

As is common in the Open Source world, some people were dissatisfied with WURFL - both with the data and the lack of a commercial version - so they forked the project. This is perfectly normal practice. You take the Open Source information, and you do what you want with it - making it fit your needs better.
(The specifics of who wanted what and why are probably best left to historians.)

As I understand it, DeviceAtlas was originally based - partly - on WURFL. So too is newcomer OpenDDR.

Enter The Lawyers

In 2011, WURFL became ScientiaMobile - a company designed to profit from WURFL. Not unreasonable in my opinion. What they did, essentially, is fork WURFL and re-licence.

WURFL then became unsuitable for anyone wanting to contribute to or use future versions of it in a manner consistent with the original licence.

All of this is perfectly fine - it's what happens next which I find problematic.

ScientiaMobile claimed that OpenDDR was illegally using their copyrighted data:

On the 3th of January 2012 GitHub received a letter from ScientiaMobile requesting to remove from GitHub the repository OpenDDR-Resources. The letter claims that the repository contains material infringing Scientiamobile copyrighted work.

From: Open letter about the takedown by Scientiamobile on one of our GitHub repositories

As the letter goes on to say that the original licence for WURFL specifically stated:

This project is open-source and is intended for developers working with the WAP environment. All the information listed here has been collected by many different people from many different countries. You are allowed to use WURFL in any of your applications, free or commercial. The only thing required is to make public any modification to this file, following the original spirit and idea of the creators of this project.

(My emphasis added).

So, what have OpenDDR done wrong according to ScientiaMobile? Nothing that I can see. They state:

we allege that the two files entitled "oddrVocabulary.xml" and "DeviceDataSource.xml" available in that directory infringe upon my client's copyright in WURFL.xml

From the letter from Scientiamobile lawyer

Databases and Copyright

I'm not a lawyer (sorry mother!) so I find it a bit hard to understand exactly what's going on here.

There is currently a case before the European Court of Justice concerning whether a database of football fixtures can be covered by copyright.

The case has not yet concluded. But the opinion of the Advocate General is available.

As I understand it - and I may be wrong - there are three aspects to database copyright.

  1. Facts cannot be copyrighted.
  2. The structure of a database cannot be copyrighted unless it is sufficiently creative.
  3. "Sui generis" (a Latin phrase meaning "The lawyers are getting more expensive now!") It means that the work as a whole could be considered under copyright.

Undeniably, the contents of the WURFL database are facts. They are not subjective interpretations, nor are they creative works.

The structure of the database is an interesting one - there are only so many ways you can place this data into a file. One might be able to argue that the naming of the database tables and fields are creative and could be copyright. Although, as OpenDDR have done, one could change these in a derived version of the database.

Next, is the whole work subject to copyright? If a collection of football fixtures cannot be subject to copyright - how can a collection of device characteristics be?

Finally - even if the new version which ScientiaMobile have created is a copyrighted work, there can be no doubt that the original database was under a very permissive licence.

Taking Without Giving

The original WURFL was licensed as GPL. This means that if you took the code, modified it, and then distributed it - you had to publish those changed.

In the case of web services, the GPL doesn't consider offering a service through your website as distributing. Hence why WURFL decided to re-licence under the Affero GPL.

As Luca said at the time:

In a nutshell, AGPL is a strong copyleft, just like GPL, but with one big difference: using your WURFL application on a web server constitutes distributions and triggers the GPL copyleft provisions.

Now, is it wrong for companies like DeviceAtlas to take WURFL, improve it, and not contribute those changes back?

It may be a bit unfriendly, but there's nothing illegal about it. The original licence allowed for it. Even if it hadn't, it doesn't look like a database of facts can come under copyright.

Here's the kicker; OpenDDR were publishing all their changes to the WURFL database on GitHub!

So, as far as I can tell, OpenDDR are being accused of:

  • Taking a database covered under GPL.
  • Changing the data.
  • Publishing the changes.

What exactly is the problem with that? Nothing!

Conclusion

The original WURFL was created by a community of volunteers who, we can assume, contributed data with the understanding that it was to be licensed openly.

Other companies took the WURFL data - which was not copyrighted - and began to sell access to it. They also improved the quality of the data without contributing back.

This angered Luca, so he decided to close future access to companies unless they paid ScientiaMobile. I don't think that's the best thing to do, but it was his decision to make.

What he cannot do, in my opinion, is retroactively change the licence and then go after anyone who uses an older version of the database. Even if he could, the database itself is not subject to copyright.

I'll leave the last word to Luca:

Conflict of Interest

I'm not a fan of Luca Passani - and he's not a fan of me. We've both been at the opposite ends of technical disputes in the past. I like to think that I have always behaved professionally and courteously, but I do not feel that Luca has reciprocated that respect. This may affect my judgement in this blog post.

Support this blog

Enjoyed this blog post? You can say thanks to the author in the following ways:

Donate to charity
Give to charity.
Buy me a birthday present
Amazon Wishlist
Get me a coffee
Donate on Ko-Fi.

63 thoughts on “WURFL and Database Copyright

  1. As I understand it, you can copyright a database of facts based on the principle of "sweat of the brow". So if someone was to create a list of everyone involved in the Seven Years War, their year of birth and death date then these are all facts AND out of copyright. The data could be stored in Excel (ie not a clever structure) and could be copyrightable because of the effort of compiling the list. I made this up but you can imagine someone devoting a lifetime to putting this list together.

    Obviously the facts in the database in isolation are still not copyright. And you cannot steal someone elses work add two extra lines and say its yours .... or can you? See below.

    My knowledge of this comes dealing with Wikim/pedia pictures. The National Portrait Gallery claim that the "sweat of the brow" of taking a 300 year old painting and taking a photo of it means they can claim a copyright on the derived picture. Wikim/pedia don't agree as the effort is so small ..... there is no "sweat". The NPG have been litigous about this in the past .... but not pressed home their opinion. So it depends on the amount of sweat. Anti-perspirant? anyone.

    1. Good points, Roger. I think it becomes even more problematic because so many individuals contributed to the WURFL dataset. True, the team behind WURFL certainly put in a lot of effort in sanitising that data - but is that enough?

    2. Wikimedia allows the NPG photographs because there is no such thing as "sweat of brow" copyright in the US (see Bridgeman v. Corel), which is the primary jurisdiction that they care about. It has nothing to do with how 'much' sweat they think was involved.

  2. For the record, DeviceAtlas isn't based on WURFL but we did use data from it in the past, in addition to data from our many other data partners (listed on our site). DeviceAtlas uses a completely different data structure, API and pretty much everything else compared with WURFL. You can view our device property list here: http://deviceatlas.com/node/2410065

  3. There is clearly a long way to go concerning this debate. Commercial companies rarely embark on this kind of legal action without being sure of their position.

    As a newcomer we'd also like to mention 51Degrees.mobi did not use WURFL as a basis for our new products. A different data structure and vocabulary have been used.

    51Degrees.mobi provides a solution for those seeking alternative mobile device data and APIs free for commercial use, licensed under the Mozilla Public Licence. http://51degrees.mobi/Products/DeviceData.aspx.

  4. Eden, I am very tempted to answer point by point, but my legal counsel strongly advises against it. Suffice it to say that you reconstruction contains plenty of mistakes and inaccuracies.

    Everyone who has looked at Openddr agrees that Openddr copied WURFL hands down: data and vocabulary. Since I and Steve have dedicated several years of our lives (and used our own economic resources) creating and maintaining the project, I am not sure why anyone would find it surprising that we are protecting our investment. In fact, we tried to be as much as possible fair to WURFL users by not forcing anyone to upgrade overnight and leaving adopters plenty of time to decide whether to keep using WURFL or look for alternative solutions.

    OpenDDR is a different animal. Whoever is behind the initiatives (they are anonymous in case you have not noticed) is scientifically trying to work around IP protection laws to fork WURFL illegitimately and give a start-up in its infancy a hard life.

    In this context, ScientiaMobile's reaction is simply the least one can expect. Please remember that it's Intellectual Property (copyright mainly) that creates the foundation for FOSS and commercial software alike. ScientiaMobile is proudly exploring a new way to do commercial Open Source. A model where the advantage of open source code and open repository is passed on to WURFL licensees, for ease of integration and maintenance. Any developer just needs to think of the implications of working with closed-source software and systems to understand the huge advantage of this. The attempt to steal from a FOSS project because one doesn't like the license is ethically despicable, even before it is illegitimate.

    Finally, I am annoyed by your comment that I disrespected you, when in fact I have no idea of what you are talking about.

    Luca Passani
    CTO @ ScientiaMobile

    1. Luca,

      Thanks for your comment.

      No one disputes that OpenDDR copied WURFL.

      What I am saying is that your licence specifically allowed and encouraged them to do so.

      Was WURFL under GPL? Yes.
      Does GPL allow anyone to copy? Yes.
      You requested that anyone publish their changes. Did OpenDDR? Yes.

      So what exactly is the problem?

      Or do you claim that OpenDDR copied your AGPL code?

      If so, again, what is the problem with OpenDDR distributing as FOSS?

      Finally, while you may not remember it, you waged an extremely personal and nasty campaign against me and my colleagues when I worked for Vodafone UK's mobile web team.

      I understood and sympathised with your position on transcoding, but there was absolutely no need to whip up such a level of vitriol.

      Thanks once again for your comment - I look forward to answers to the above questions.

      Terence

      1. Come on. That was not a "personal" attack on you. Of course, it was an attack against Vodafone UK and Novarra, and a fully justified one: someone I know concisely summarized what the Voda-Novarra duo did as "pissing in the water everyone drinks" and you would be amazed at the amount of support I was getting behind the scenes...

        No more comments on the openddr thing. I think I summarized ScientiaMobile's position effectively in one of my tweets:

        - Can we use your old hammer to hang our picture to our wall?
        - Yes, you can.
        - Can we hit you on the head with it?
        - No, you can't.

        Luca

        1. So basically Luca still hasn't answered the simple question that Terence and so many others have posed:

          "If so, again, what is the problem with OpenDDR distributing [a WURFL derivative] as FOSS?"

          It seems completely clear that what OpenDDR are doing is technically legal, even if it is not "fair". It also seems that this outcome was entirely predictable given the WURFL license change.

          1. To be clear, I haven't addressed Eden's questions in detail because I have been advised not to do so, but we claim that what OpenDDR is doing is not legal.

            Luca

            1. This reminds me so much of the whole SCO vs Linux incident. SCO claimed they had copyright on parts of the code but refused to point out what part of the code.
              ScientiaMobile is doing the same. Claiming to have copyright on parts of the OpenDDR but refusing to tell why - which is an even more ridiculous stance.

              1. 1) ScientiaMobile has already explained that the vocabulary and the data were copied.

                2) OpenDDR is an anonymous initiative created to damage both ScientiaMobile and WURFL. It is not fair to require that ScientiaMobile explains anything to them. We are kind people, but we are not stupid.

                1. OpenDDR is an open source, community driven project. I would think twice about attacking it. It /will/ make you look bad. The OSS community isn't exactly fans of DMCA takedown notices.

                  There might be a business idea and a even commercial company behind OpenDDR - but since they seem committed to keeping their product OPEN they get my vote.

                  I think it will be impossible for Scentia Mobile to try to kill off any open source effort to provide a DDR API. There is a OSS alternative for everything today and DDR APIs is no exception. Even if SM kill OpenDDR a new project will surface soon.

                  1. We don't want to kill other open-source DDR efforts. We just demand that they do not start by infringing our copyright (copyright laws applies also in the case of FOSS projects) .

                    1. This is preposterous. Luca this is fooling nobody. Quit dodging the central point. DDR is not infringing your copyright, they forked from YOUR liberal license.

                      Much as you'd like to put the genie back in the bottle you can't. The reality is that without OpenSource your project would not have experienced the popularity it did. Now having capitalised on OpenSource appeal you want to raise the drawbridge. Well you can't, it is TOO LATE. By all means retreat to your cathedral but the community already has a RIGHT to use the software you previously licensed.

                      If you improve your database you will be able to restrict access to the improvements, but you have the rest of the world to compete against hacking on your freely licensed original, and after the political disaster of this foolish takedown nobody will be beating a path to your door.

                      This is a REALLY dumb move and is only outdone by your insipid replies here. Do you think people in the community are idiots?

                    2. OpenDDR and their supporters are claiming they are not violating copyright law because they forked your project while it was under GPL. GPL is a contract under copyright law, and it's a contract that permits such forks. How do you respond?

  5. This is a bit off-topic, but I dare you to read this article with a 1024x600 netbook screen.

    Please fix your blog layout so that the two sidebars don't hog all the space, leaving almost nothing for the actual content. Thank you!

    1. Thanks - I've updated the layout so it looks a bit nicer. Will have a play with it over the weekend and see if I can make any further improvements.

  6. If anyone's in the market for practical solutions, the obvious solution is for OpenDDR to adopt the UAProfile "standard" vocabulary to function as the missing "W3C Advanced vocabulary" and remove the WURFL vocab and data.

    Since the OpenDDR front page gives great weight to "standards" and "not being sued", it would seem to be a good idea for it to use exclusively existing standards rather than "de facto" ones. That's what existing standards were designed for.

    Chris

  7. [I'll beat the layout issue and write here instead]

    @gus

    We claim that DDR is infringing our copyright. Since the OpenDDR gang hide themselves under a cloack of anonymity, I don't understand how anyone can demand that ScientiaMobile explains anything to them.
    If one is so confident of their position, why not come forward and stand by it?
    Why aren't you demanding that OpenDDR comes forward?

    The WURFL API is licensed under AGPL. An FSF and OSI approved license.
    But that's not all: legitimate users of our AGPL-licensed API can use our public snapshot. We tried to be super-fair to the community.

    If AGPL doesn't work for you, but the problem is money, drop us an email, explain and we will grant you a discounted (or even free-of-charge) commercial license.

    If the problem is that, according to someone, FOSS must necessarily mean free-of-charge, that's your problem. We disagree.
    RMS himself has confirmed that there is nothing wrong with selling GPL-exceptions.

    In general, maintaining WURFL requires hiring programming resources and paying their salaries.
    As I have explained multiple times, the default trajectory for WURFL with the old model was a slow and painful death, since nobody (well, very few) were paying a dime for it. ScientiaMobile is injecting new life into WURFL for the advantage of everyone who finds value in it (did I mention that nobody is forced to use WURFL?). The overwhelming majority of long-time WURFL users are cheering for the change and congratulating us. Of course, you can't make everyone happy.

    But there is one point in what you wrote that particularly bugs me:

    > If you improve your database you will be able to restrict
    > access to the improvements, but you have the rest of the world
    > to compete against hacking on your freely licensed original,
    > and after the political disaster of this foolish takedown nobody
    > will be beating a path to your door.

    Do you really stand by to what you just wrote? the implications of this is that one either belongs to the world of corporate software manufacturing (software = black box, guarded by lawyers) or one has no rights and everyone else is entitled to steal their work. Do you steal apples from stalls when you go to the market?

    ScientiaMobile is exploring an intermediate model where the power of open code and open data is preserved for commercial adopters, while an economic model is preserved for those who invest their time and money in delivering a good product. OpenDDR is anonymously trying to disrupt this in ways that (contrarily to what they claim) are not legitimate. We find that very arrogant and we will fight that as much as we can.

    Other comments have already pointed to ways one can create a FOSS DDR that does not steal from WURFL illegitimately. If OpenDDR was in good faith, it would know which direction to take...

    Luca Passani
    CTO @ ScientiaMobile

    1. > If you improve your database you will be able to restrict
      > access to the improvements, but you have the rest of the world
      > to compete against hacking on your freely licensed original,
      > and after the political disaster of this foolish takedown nobody
      > will be beating a path to your door.

      Do you really stand by to what you just wrote? the implications of this is that one either belongs to the world of corporate software manufacturing (software = black box, guarded by lawyers) or one has no rights and everyone else is entitled to steal their work. Do you steal apples from stalls when you go to the market?

      That's a dishonest response. You're completely ignoring the fact that your license terms allow this data to be used in the way it is being used. To use your own flawed analogy, OpenDDR is not stealing apples from your stall -- they're eating the apples you freely gave away.

      On the other hand, you're in the position of trying to take back something you released to the community in the past, because you feel people are abusing the rights you gave them. You say it isn't nice for others to use the product of your own work to build competing services. I say it isn't nice to tell people they're allowed to use your work and then turn around and sue them when they do.

      1. "On the other hand, you’re in the position of trying to take back something you released to the community in the past, because you feel people are abusing the rights you gave them"

        People are not "abusing" - merely exercising the rights

  8. Luca, OpenDDR are claiming that they are using WURFL data under the terms of the license applied to the version of the data they are are using. You are claiming that they are violating your copyright.

    1. Are you asserting that they are misrepresenting which version of the data they used and are therefore violating the new licensing terms?
    2. Are you asserting that they are using the version they claim, but that they are violating the terms of the license applied to that version of the data?
    3. Are you asserting that the terms of the new WURFL license somehow apply retroactively to versions of the data that predate the license change?

      1. There is an expression in Italy: "Metterci la faccia", i.e. if you believe in what you say, stand up by it and show everyone your pretty face.
        This is exactly what you are not doing. Who are you? what is your connection to WURFL or OpenDDR?

        Granted, this is the Internet, a little intellectual challenge can legitimately come from anonymous users. But we are way past that stage here. We have a company, ScientiaMobile, which has invested massively to make WURFL a sustainable FOSS project, supported by the sale of GPL exceptions. We have an anonymous entity (OpenDDR) which is scientifically trying to sabotage that effort by misappropriating ScientiaMobile's intellectual property. At this stage, we are not going to play funny thought games any more.

        We want to know who OpenDDR is and get to the end of this. For the record, OpenDDR announced the availability of their product without mentioning WURFL on their website, while, at the same time, advertising that OpenDDR will 'screw' WURFL in conferences and on Twitter.

        Did they come to us and ask for permission about this operation and clarification of what they could and could not do with older versions of the repository? No, they did not. They just tried to discover loopholes in our licensing structure and exploit them. This is a fully-fledged cyber attack and we are taking it very seriously.

        So, the answer to your question is "no comment"

        Luca Passani
        CTO @ ScientiaMobile

        1. I work for a company that produces open source software (MIT-licensed, to be specific) and provides services using and built on that software. I have no connection with OpenDDR or WURFL, although I have colleagues who are currently using either WURFL or something similar (I'm not involved, so I'm not actually sure) in a mobile web application. Given what I see as unreasonable legal action taken against someone using WURFL data in compliance with the terms of the license (and ScientiaMobile has provided nothing to contradict this beyond baseless assertions, despite repeated requests for clarification), I'm hesitant to identify myself more specifically than that for fear of exposing my colleagues to similar legal action.

          As far as I can see, OpenDDR is not trying to sabotage WURFL. They are trying to provide an open alternative, which is exactly what WURFL used to be. The existence of a free alternative will no doubt impact ScientiaMobile's revenue, but they way to compete with that is to provide a better product or service.

          You claim that OpenDDR is "misappropriating ScientiaMobile’s intellectual property". I can only see three possibilities for this to be the case, which I have already enumerated, and I'd like to know which you believe to be the case.

          As I read the license terms, I don't see anything they have done wrong. There is no requirement to mention WURFL or ScientiaMobile, and doing so might actually open them to legal action regarding trademark infringement or something. There is no requirement to ask for permission from ScientiaMobile before using the data, and the term clearly state: "You are allowed to use WURFL in any of your applications, free or commercial. The only thing required is to make public any modification to this file, following the original spirit and idea of the creators of this project."

          In fact, there's a strong argument to be made that ScientiaMobile are in violation of these license terms. Surely prohibiting access to the data by means other than the provided APIs (which are unavailable for a number of popular programming languages) does not follow "the original spirit and idea of the creators of this project"? Given that much of the data "has been collected by many different people from many different countries", does ScientiaMobile even have the right to change the terms of the license without consulting all the contributors? (You need some kind of copyright assignment in order to do that, as I understand it. Since the contribution process isn't clearly documented, I don't know if this has been done.) Personally, I'd be extremely upset if I contributed to a project only to have my right to use it summarily revoked.

          Please note that I'm not trying to criticise ScientiaMobile for wanting to build a business on open source software development -- I'm criticising the manner in which it is being done. By sending takedown notices to competing projects, you are creating a hostile environment and discouraging others companies from opening their codebase.

        2. A wise man from our own culture has coined a similar but more relevant expression:

          "Believe in something, even if it's wrong!" -Glenn Beck

        3. And poor, weak ScientiaMobile and its lawyers are being exploited. What kind of monster actually believes that corporation means what it says in its contracts?

  9. What I find strange is how the license changes in the WURFL project have been carried out, regarding both the database and the code base. Unless the project has had a consistent policy of "no contributing without copyright assignment", any license changes would require that all contributors agree with the new license or the contributions of those who do not agree be removed from the relicensed work. I have seen no signs of such a massive undertaking rather than just an announcement of a license change with a lot of grumbling from contributors in response.

    As for the present matter, the facts as presented now look very damning toward ScientiaMobile given that OpenDDR publicly asserts that their work is based on a version of the database which undisputedly allows them to make a derivative work. Also, Luca Passani's comments here and on Slashdot give the impression that he is dodging the fundamental problems presented regarding his argument and seems more concerned about the competition OpenDDR poses instead of any legitimate copyright related issues. However, I do recognize that his counsel may have advised him to not speak too much and leave the best arguments for the court room.

  10. I can confirm that ScientiaMobile is in its full right to do what it did.

    Professionals in the software business don't normally have a problem with the idea that good software has to be paid for, since they are the first ones who know that software takes serious knowledge, serious research and serious effort.

    Apart from that, my debugger smells a loop here. We are not going to discuss aspects concerning copyright infringement in public. The takedown notice already provides plenty of information. Commercial entities that took from WURFL in the past have explained (some of them in this very thread!) how they changed their vocabulary, data, data structures and API to avoid infringing. This should provide OpenDDR plenty of inspiration about how to do things right.

    Luca Passani
    CTO @ScientiaMobile

    1. > I can confirm that ScientiaMobile is in its full right to do what it did.

      On what grounds? You keep making assertions without providing anything to back them up, which is why we're finding them hard to believe.

      > Professionals in the software business don’t normally have a problem with the idea that good software has to be paid for, since they are the first ones who know that software takes serious knowledge, serious research and serious effort.

      That's not the issue under discussion, and it reflects poorly on you that you are implying that people using permissively-licensed software under the stated terms of the license are somehow immoral freeloaders.

      I'm a professional in the software business. I'm happy to pay a fair price for quality software, but I'm also happy to use free software. I generally prefer free software, but only because it tends to have developers that are more responsive to bug reports or feature requests, and because I often have the opportunity to improve the software myself and contribute my improvements back to the community. So no, I do not have a problem with "the idea that good software has to be paid for", as long as you aren't using that phrase to denigrate the vast number of extremely good software projects that are available for free. I do, however, have a problem when people who use permissively-licensed code under the stated terms of the license are sent takedown notices and accused of copyright violation.

      > We are not going to discuss aspects concerning copyright infringement in public.

      Why not? If I were interested in using old WURFL data in one of my own endeavours, I'd like to know where I stand and whether I should expect legal action against me. ScientiaMobile made the code and data public under a permissive license, and I believe it is entirely reasonable to be told what the perceived violations are so that others know where they stand. In fact, I believe that you have an undeniable responsibility to be clear on what kinds of uses you consider to be in violation.

      > The takedown notice already provides plenty of information.

      No, it does not.

      1. My client's work that was infringed - the WURFL.xml - can be found at the following URL: http://www.ScientiaMobile.com/communitywurfl/wurfl.zip. You also ask that we identify the work infringing upon the copyright in my client's work. We have done so in item 2 below.
      2. The material we allege is infringing WURFL.xml is found in the directory entitled "Resources" available at https://github.com/OpenDDR-org/OpenDDR-Resources. In particular, we allege that the two files entitled "oddrVocabulary.xml" and "DeviceDataSource.xml" available in that directory infringe upon my client's copyright in WURFL.xml and it is these two files that we request you cease distributing through your website.

      The claim is that two files in the OpenDDR repository violate the copyright in "WURFL.xml", and the version linked to in the takedown notice is the August 2011 version, with a different license from the July 2011 version OpenDDR claims to have used.

      This license states "You are required to make modifications to this file public and be fair to other developers and companies that have shared their changes with you and with all other WURFL adopters." OpenDDR have made their modifications public, as required by both this version of the license and the July 2011 version.

      The only other possible restriction is very ambiguous: "The data is meant for use with the WURFL API available on the official WURFL website at http://wurfl.sourceforge.net." I'd read that to mean "this is the intended use, and we're unlikely to accept patches that conflict with this", but it could be meant as a very poorly worded variant of "you are prohibited from using this in any other context". Is this the violation you're claiming?

      In summary: The takedown notice claims unspecific violation on grounds that are tenuous at best, against a version of the WURFL data different from the one OpenDDR claim to have used, which is licensed under slightly different terms.

      > Commercial entities that took from WURFL in the past have explained (some of them in this very thread!) how they changed their vocabulary, data, data structures and API to avoid infringing.

      Where is the documentation about how to do this? Where does it state in the license which bits are excluded from the "only thing required is to make public any modification to this file" clause? When I see a license applied to a file, I expect the entire file to be licensed under the same terms unless there are specific exclusions noted.

      Luca, you are not being reasonable. No matter what heinous crimes OpenDDR may or may not have committed against you or anyone else, your refusal to explain what uses you consider to be in violation of your July 2011 (and August 2011, for that matter) WURFL.zip licence terms is not making you any friends. Even the people who, like myself, are inclined to support any commercial enterprise that releases open source versions of their products are being alienated by your apparent disregard for the terms of your own license and your obvious disregard for our concerns about your recent actions.

      1. A small correction: I believe OpenDDR says they are using the database of April 2011:
        One of the sources we looked at was the content of the file wurfl.xml, in its version dated April 24th 2011,

        In fact, they explicitly state that no data from the July 2011 archive is used:
        no data coming from July 2011 Snapshot has been used in OpenDDR, while April 2011 version has been used instead.

        Unless OpenDDR is lying about this, I can see no basis for accusations on grounds of either copyright or licensing related issues.

        1. Oh, and in case it got forgotten in the progress of this debate, the April 2011 and July 2011 snapshots apparently have slightly different licenses. According to the open letter published by OpenDDR, the April 2011 license reads as follows (hopefully I did the quote correctly this time):

          Dear Developer, the WURFL file contains information about the
          capabilities of mobile devices, capabilities and features.
          The main scope of this file is to collect as
          many information as we can about all the existing wireless devices
          that access mobile pages so that developers will be able to build better
          applications and better services for the users.
          This project is open-source and is intended for developers working
          with the WAP environment. All the information listed here has been
          collected by many different people from many different
          countries. You are allowed to use WURFL in any of your applications,
          free or commercial. The only thing required is to make public any
          modification to this file, following the original
          spirit and idea of the creators of this project. This will help WURFL
          to grow better and better every day. The use of WURFL is at your own risk,
          there is no warranty that all agents and
          capabilities are correct. All the information, as stated, has been gathered
          by many users from many places around the world and as such there is
          no guarantee about the reliability of this
          information. Post any noteworthy modification and comment
          to the file on the WMLProgramming mailing list on Yahoo Groups:
          http://tech.groups.yahoo.com/group/wmlprogramming/ Alternatively,
          you can mail your modification to Luca Passani,
          the WURFL maintainer, at passani at eunet dot no

          Most notably, the license pertaining to the version OpenDDR claims to have used does not even contain the passage you suspect to possibly be the basis of the accusation, as stated by the open letter.

          1. I may have been unclear. The reference to that passage is in the context of the version referenced in the takedown notice, not the version OpenDDR claims to have used. I was struggling to find /any/ grounds for complaint in the takedown.

        2. Whoops, thanks for catching that. For some reason I was using July/August instead of April/July. Please adjust accordingly when reading my previous comment.

  11. @everyone

    We got many pings asking us to respond to OpenDDR's posting. We are not going to do it. ScientiaMobile has already expressed its position multiple times. Not only do we not have an obligation to answer them here, but we would be naive if we did in such a hostile environment.

    Having said this, I would like to make a point clear: we are not OK being represented as a big evil corporation, which is what some postings on Slashdot implied.
    We are neither big nor evil. 10 years of work with WURFL and its community vouch for that. WURFL users have recognized the passion and the dedication that I and Steve have placed in delivering WURFL and they understand that the previous model was not sustainable, hence the support for the ScientiaMobile move that we have received all the way from its inception.

    Far from being a big evil corporation, ScientiaMobile is a start-up that tries to make Richard Stallman's dream come true: work for a company that produces Software Libre. Of course, going from dream to reality is not for the faint of heart, but we try...

    Now, Let's try to be constructive here. I want to talk about ScientiaMobile and I want to talk about how ScientiaMobile could do something that the larger community of developers would appreciate, while keeping our business model alive. Of course, I would like to hear honest reasonable opinions. Good, fast and free of charge is not an option.

    So, here are the questions for you:

    In your opinion, what would a fair business model be for ScientiaMobile and WURFL?

    is there something that we can do that would be welcome by the community while not totally destructive of our business model?

    How can we achieve our objective of a commercial FOSS offering that will preserve the value of open code and open data to licensees, without having to fear that someone will copy and squander our investment?

    Before you answer, please consider this: running a 6 people company in the US means having a revenue of *at least* $500k per year. There is no way one can make $500k on voluntary donations.

    Opinions that are both honest and reasonable will be taken in serious consideration.

    Thank you

    Luca Passani
    CTO @ScientiaMobile

    1. > In your opinion, what would a fair business model be for ScientiaMobile and WURFL?

      I don't have a problem with your current business model, although I think you'll find it hard to compete with a high quality community project that provides a similar service free of charge. On the other hand, you probably have the right skills and experience to pull it off. Having standard options for hosting companies might be a good idea.

      > is there something that we can do that would be welcome by the community while not totally destructive of our business model?

      The restricted set of languages supported is problematic. There are a number of popular web frameworks in Python, Ruby, etc. that are currently excluded under the licensing terms that prohibit using the data apart from the provided APIs. I think having some officially sanctioned APIs in these languages, even if they're community-contributed, would help with this.

      > How can we achieve our objective of a commercial FOSS offering that will preserve the value of open code and open data to licensees, without having to fear that someone will copy and squander our investment?

      My own company maintains an open source framework and set of tools, and provides services and custom applications built on top of that. In our case, the revenue stream is from building and operating applications and services on behalf of others rather than selling licenses to the public. I doubt WURFL is suited to this sort of model.

      1. Thank you for your comment, Jerith.

        You are right that business model which works for other FOSS products does not work for WURFL. Been there. Done that.

        As far as third-parties API go, there was a subtly crucial technical problem: third-parties API tend to be simpler than the standard API, hence people ended up adding unnecessary/bogus data in WURFL (think 7 identical Android devices where only the locale substring changes).
        Anyway, we are not really blocking third party WURFL APIs. People with a non-standard WURFL API can bundle it with their own product, add an older version of WURFL (the April 2011 version, for example) and advise users that they can get an updated file from the WURFL site. At that point, how their users will cope with WURFL licensing is not their problem.
        For example, commercial licensees of WURFL will also get a customer vault with weekly snapshots all licensed more liberally for the use by that customer.
        Of course, I am not saying that this situation is ideal for other FOSS projects, but it it is certainly not a showstopper for those who still want to use WURFL as a DDR module in a larger open-source artifact.

        Did I mention that we handled out discounted and free-of-charge commercial licenses to non-profit, non-commercial and governmental entities?

        Anyway, point taken about the need to enlarge the number of supported languages.

        Thank you

        Luca

    2. Hi Luca,

      I think it is totally reasonable for you to try and create a business out of this - if that's what you want to do.

      You ask an interesting question:

      How can we achieve our objective of a commercial FOSS offering that will preserve the value of open code and open data to licensees, without having to fear that someone will copy and squander our investment?

      The answer is simple. You can't.

      The database is under the GPL - although is now under the AGPL I believe - that means anyone can copy it and resell it (as long as they publish the changes & comply with the licence).

      So, anyone can take your data - and reuse it for free - this means that you can only compete on value for money through superior customer service and professionalism.

      Essentially, because your data is freely available, you have to compete on service. That means a slick website, guaranteed SLAs, versions in as many different programming languages, working with CDNs like Akamai, show a long term partnership plan to continue getting data etc.

      Make it so that it is easier to use Scientia than to use the competition. While people like me are quite happy to manually compile stuff, code around API limitations, etc - most businesses want an "out-of-the-box" solution which doesn't require them to think, and comes with support.

      You can't stop other people from (legally) copying the WURFL database. So you need to decide whether Scientia will be able to recoup the investment. It may not be possible. You may have to be satisfied with being one of many DDR providers - even if it isn't profitable.

      Terence

      1. As far as third-parties API go, there was a subtly crucial technical problem: third-parties API tend to be simpler than the standard API, hence people ended up adding unnecessary/bogus data in WURFL (think 7 identical Android devices where only the locale substring changes).

        Firstly, this is a data management problem rather than an API problem. As the gatekeepers of the data, you're well placed to handle it. If a particular API that is causing problems, you can take it up with the maintainers or something. A blanket ban on any APIs not provided by ScientiaMobile is overkill.

        Anyway, we are not really blocking third party WURFL APIs. People with a non-standard WURFL API can bundle it with their own product, add an older version of WURFL (the April 2011 version, for example) and advise users that they can get an updated file from the WURFL site. At that point, how their users will cope with WURFL licensing is not their problem.

        This adds barriers to using WURFL in an application, and it isn't obvious what the options are. I'd be hesitant to build and maintain (for example) a Django module that required users to choose between a very old version of the data or a commercial license. Besides, the licensing options specifically state that they are "per platform", and there's no option for "using my own API".

        Of course, I am not saying that this situation is ideal for other FOSS projects, but it it is certainly not a showstopper for those who still want to use WURFL as a DDR module in a larger open-source artifact.

        It's a showstopper if the project in question doesn't use any of the supported languages and wants to provide access to data newer than April 2011.

        Did I mention that we handled out discounted and free-of-charge commercial licenses to non-profit, non-commercial and governmental entities?

        It might be worth mentioning that on your product page. The only options there are "open source" and "commercial".

        1. Thank you again for the comments. If one is only interested in licensing the data, then they can chose any platform. This is not going to change the licensing price anyway (there's no discount in only wanting the data. The spirit is always the same of not incentivizing 'diverging' implementations). btw, if one is the creator of a Perl/Python/Ruby WURFL API and is willing to do whatever it takes to implement the same logic that is in the standard API, please feel free to ping me offline, because we are interested in talking to you.

          wrt the new licensing being a show-stopper, I see plenty of software products/projects that will direct you to third-party sites to download this or that component, which is not directly bundable, but can be obtained elsewhere.

          Finally, I added a FAQ entry where we mention that non-profit/governmental may obtain discounted/gratis licenses, provided that an explanation is provided.

          Thanks

          Luca

  12. Thank you Terence. Only a couple of comments. The ScientiaMobile standard WURFL APIs come with a dual licensing model AGPL and ScientiaMobile custom license.

    The repository comes with our own custom licenses (one for the public snapshot and one for the licensed version) which is not AGPL, albeit users of the AGPL version (who in compliance with the terms of AGPL, of course) are allowed to use the public snapshot.

    > The answer is simple. You can’t.

    well, we think we can. And this is exactly what we are trying to do. The implication of what you write is that one necessarily has to choose side: either go commercial black box or blithely accept that others will freely take and confine yourself to the world of services.
    What we are trying to do (let me write it in US English...)...what we WILL do, is show the world that there is a third way in the middle and that one can legitimately sell open code and open data to entities for which openness represents a great value proposition.

    Thank you

    Luca

  13. Another point to muddy the waters - did the contributors to the WURFL data set sign over copyright to Luca/ScientiaMobile?

    If not, then Luca has no right to relicense the whole lot under a new licence without contacting every contributor.

    Of course, new contributions may be under a new license, but the old ones cannot be.

      1. I don't understand why you keep claiming 'hostility', when all we really want are answers.

        As it stands, there is *still* confusion over whether or not you/Scientia are claiming that OpenDDR have abused the licence by using the old WURFL data under the old licence, or if they're flat out lying about which data set they're using.

        If it's the latter, then there's no doubt that you're completely in the right.

        All we want are answers to what seem to us to be pretty obvious, and easy to answer, questions.

        *far* from hostility.

        1. Daniel, are you seriously asking why I claim hostility?

          In this case, I will let a selection of tweets *you* directed to me over the past few days speak for themselves:

          "I've just discovered the current shitwar between @OpenDDR & @luca_passani over 'illegal activities'. Absolutely incredible stuff."

          "@luca_passani Er, it's clearly about protecting Scientia, not about protecting the (wurfl) data itself @openddr"

          "So @luca_passani says he can't comment because his lawyers said not to. Absolute bullshit. If you can't plainly explain the .... /@OpenDDR"

          "... reason for being so aggressive against a free, seemingly legit operation like @OpenDDR, then stop being so vocal in public @luca_passani"

          "There's little I hate more, than hiding behind 'the lawyers say shush'. Ugh. This while situation is a complete farce @luca_passani"

          "Wow - the WURFL/@scientiamobile/@openddr pissing match hit slashdot. And @luca_passani seems to be taking some flak for the whole debacle"

          "The Wurfl/@scientiamobile/@OpenDDR debacle just took an interesting turn, with @luca_passani refusing to discuss the infringing data /@edent"

          "I don't see how @luca_passani can spin this positively. He&@scientiamobile are clearly doing the wrong thing. Being open&honest is important"

          "For those wondering what I'm on about: http://shkspr.mobi/blog/index.php/2012/01/wurfl-and-database-copyright/#comment-15448 @luca_passani responds to unhappy people regarding @scientiamobile/@OpenDDR"

          And all of this, after I had amply explained how openddr is an anonymous initiative aimed at harming WURFL and ScientiaMobile.

          BTW, one day you will need to explain to me what I did to you. Did you also work for VodafoneUK or Novarra at some point?

          Luca

          1. *sigh*

            None of what I said was hostile.
            And, while it certainly came out of frustration at your complete unwillingness to answer the questions that so many people wanted you to address directly, it was almost entirely the result of disbelief about the whole situation.

            Nothing about this situation makes sense to me, a current outsider, and you're not helping that fact by claiming I (or others) am being openly hostile towards you.

            Even now, you're avoiding the core questions that everyone in this thread has asked you to answer clearly.

            As for your last line - you have never done anything to me. I have nothing against you.
            I *do* have something against dishonesty, or avoiding answering questions (potentially making you appear to be dishonest)

            And finally, (as I presume you already knew), I was a core developer on DeviceAtlas for a number of years, until summer of last year, when I left to pursue other challenges.
            It's because of my history with device detection that I'm so interested in what's going on here. I may not be in that space anymore, but I find the whole problem (and potential solutions) fascinating.

            Seeing a small, open source, free player being DMCA'd out of existence made me sit up and take notice. Your (and Scientia's) stance has made me pay attention to it more closely.

            If OpenDDR were doing something illegal, then fine.
            But everything you've presented here by way of explanation implies that you/Scientia are the ones who are doing something shady, and you flat out refuse to address this criticism head on, relying instead on complaining about seemingly hostile comments, apparently directed at you personally.

            Daniel

            1. I now see why 'big evil corporations' will serve everyone a dry public 'no comment' and that's it:

              > Seeing a small, open source, free player

              small, open-source, free players are not anonymous and, above all, do not open anonymous LLCs in Delaware.

              Finally, I take offence at your suggestion that I am being dishonest.

              Luca

              1. Luca, you're taking offence because it's easier to do that than to answer our questions.

                I said that by dancing around the questions, it could give the appearance of dishonesty.

                Nothing I have said up to now has been hostile, or insulting. Everything I have said has come from disbelief, a lack of understanding, and the frustration borne from your absolute refusal to address the issues that we are all asking you to.

  14. Ha, Luca is just a mercenary kid!

    He deleted all my posts on WURFL forums after I asked him about the "previous" license (before 2.1). He also requested my "linkedin profile", "email address not from a free mail service, and my FOSS project name). If I do it, then he tells me what the license was used before. Luca, you're going wrong way!

    P.S. People who want to use WURFL and do not want to pay to Luca, you can just use WURFL 2.1. It available for all FOSS projects without any limitations (its licenses could not be changed)

    1. As we have explained multiple times, we are not going to talk or disclose anything to anonymous requestors anymore.

      The commercial move was the only way to make WURFL survive. Also, I am not sure how one can call a company that tries to establish an honest business mercenary.

      We hardly delete any posts that's not spam on our forum, but we made an exception in your case because you were out to provoke, rather than engaging in constructive discussion.

      Luca

      1. bla-bla-bla, just give me some money, money, money =)))

        You deleted my post because I explained you that you're wrong. Understand open-source applications cannot hide information about the license! And of course they cannot require people to “linkedin profile”, “email address not from a free mail service, and "my FOSS project name" before getting more information about license.

  15. Regardless of the ins and outs of the law in this case (use vs republishing rights for the data file, for instance), I don't see that there's any upside for OpenDDR continuing down the path of using WURFL fields.

    A Device Database must be kept up to date, or it's going to do more harm than good. I think it's underestimated how bad at detection a device database can get after a few months of not being updated, given the number of devices being released.

    Many of the fields in WURFL are specific to WURFL: i.e. the data in the field has no equivalent anywhere else on the Internet. So, given that OpenDDR claims it's working off a very old data file, how exactly are those fields going to be populated for new devices?

    It seems to me that there's no practical, economic or sustainable way to keep OpenDDR up-to-date on these "extended" fields other than its users populating it independently with data derived from a licenced version of WURFL: which of course would not be allowed. I simply can't see a sustainable way forward for OpenDDR in its current form that doesn't involve repeated infringement of WURFL licenced data over time.

    Chris

    1. Chris,

      You can use WURFL 2.1 (not 2.2 or 2.3) which was available for all FOSS projects without any limitations (its licenses could not be changed). So start with it and continue populating it yourself (or with the help of your users).

      1. Antonie,

        Misunderstanding: I don't need to set up a DDR, I already run one, and I'm not connected with OpenDDR.

        But your advice to me there kind of makes my point: for any data going into a database, you have three options: do it yourself, acquire the data from a third party, or get your users to do it.

        If OpenDDR (or any other DDR) commits itself to using WURFL fields and supporting them independently of Scientia, then I think it will have an impossible time getting legal data for those unique fields either internally, because so many of them only originate from Scientia.

        Which means that at some point you're either going to have a systematic flow of data from Scientia to OpenDDR for those fields, or OpenDDR's updates will be restricted to just easily available data: thus making half of its extended schema useless.

        Everyone seems to obsessed by the origins of OpenDDR I haven't seen any legitimate questions or answers about how effective it is, how sustainable it is going forward, or indeed about the credentials and history of the people running it. Those practical things matter.

        Chris

  16. OpenDDR's URL has been inactive for quite a long while (suspended) but their API code was accepted into Apache's "DeviceMap".

    The WURFL data and schema thing was never resolved: they never removed it. Their Twitter feed claims "the status is definitely active updated data sets have been always released monthly since first release!", and the Github would bear that out.

    However, I suspect that they're only maintaining the WURFL fields that could be plausibly maintained independently of WURFL data: which means effectively deprecting a large part of the WURFL schema.

    1. Apache's "DeviceMap" is at the incubating stage. Presumably Apache will risk the same issues Github experienced if they don't clearly resolve the legal issue.

      If they do address the legal issue it's very hard for a community of people to maintain accurate and consistent device data without screen scrapping from web sites like GSM Arena. No serious business would want to pin their mobile architecture to something that is so fragile and could be removed at any moment. That's why good DDRs cost a little bit of money as a professional team needs to be employed. DDR data is very different to maintaining source code.

      I would like to see Apache create a reference implementation for device detection which could be extended to allow other DDRs to be plugged in. This way you could choose the DDR. I don't think this is a direction the DeviceMap team want to go down.

      If you need an open source DDR my company 51degrees.mobi provide one with a paid for option to get more features such as weekly automatic updates.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.