Update 2016-11-15: Today Adam Hyde has published an update on his blog. Although Hyde doesn’t mention us, he seems to admit that he hasn’t developed BookJS and apparently not most or all of the other software he declared he had developed. Unfortunately, Hyde seems to exchange his claim of having developed BookJS with a new claim: Now Hyde complains that his non-developer contributions aren’t mentioned enough. such a claim seems slightly odd from someone who’s Foundation only recently threatened with legal consequences should we insist on Hyde not having developed a particular software and who has traveled large parts of the planet for the past four years, speaking at conferences and receiving funding from major actors such as the Public Library of Science (PLOS), the Shuttleworth Foundation or the Gordon and Betty Moore Foundation, all while his listeners were made believe that he had participated in the development of software interesting to them. However, we are happy about Hyde’s new line of communication. As for his contributions to BookJS: Hyde did not develop it, nor did he manage the development of it, and he wasn’t the “idea man” behind it either. He and his organization BookSprints, were however the first and likely entity who ended up created real-world books using the software before it became unusable. In that capacity, he did a lot of real-world testing, that neither BookType nor Fidus Writer could provide. If Hyde agrees to no longer claim that he developed BookJS or that he was the idea-man behind it, we do not mind him claiming he did the first and only real-world production tests with BookJS. We hope that this case is closed once and for all and that this case serves as a warning to all those who want to fund open source tools so that funding does not go to waste on people and projects that aren’t likely to produce what is expected.
This statement is to clarify the situation surrounding the software “BookJS”. This statement is made due to claims made by Adam Hyde at conferences and on his blog. We have previously asked Mr. Hyde and the organizations he has set up, such as BookSprints or the Coko Foundation, to stop propagating inaccurate descriptions of the history behind BookJS and had hoped that the issue would be resolved this way, also because BookJS has not been in a working state for several years. Unfortunately, Hyde has chosen to continue spreading incorrect versions of what has happened, and the Coko Foundation has threatened us with legal action should we email them about this a second time.
To be absolutely clear: Mr. Hyde did not at any time develop BookJS nor can he it in any reasonable way be said that he directed the development of it. He also did not come up with the idea behind BookJS. We do not know if Mr. Hyde was involved with the development of other software he claims to have developed.
What happened?
In short: Johannes Wilm was contracted by the Czech NGO Sourcefabric to work on their product BookType. Hyde also had some kind of contractual relationship with Sourcefabric during this period, the details of which we do not know, but it was not in the area of software development. Wilm did initially not know the roles of the various actors at Sourcefabric, and therefore followed through with an review BookType’s current state which Hyde asked him to do. In Wilm’s findings, he looked critically at the existing PDF export filter that was part of BookType at the time. Hyde responded, mentioning that he had seen the product wkhtmltopdf, which allows to create PDFs from HTML-files, and Hyde pointed Wilm to a W3C draft specification of a technology called “CSS Regions”. Hyde also commented that the technologies possibly could be connected. Wilm instead came up with using the browser’s print-to-PDF functionality and a lot of JavaScript to create PDFs. In the initial version of BookJS, Wilm also used the aforementioned CSS Regions, even though this technology was still in the development stage and it was not sure whether CSS Regions would have a future. Hyde had explained that he was sure that CSS Regions would be adopted by all major browsers, and Wilm therefore dropped plans to create a version without CSS Regions. In early 2014, after Wilm’s contractual relationship with Sourcefabric had ended, it turned out that the team behind the Google Chromium browser, which Hyde and Sourcefabric had been used to generate PDFs, decided to remove CSS Regions, and BookJS became unusable.
In his blog post, Hyde mentions a meeting he had in Amsterdam prior to Wilm’s contract with Sourcefabric where he claims to have seen a book made using a browser, but the link he provides to backup his claim shows that he had merely seen a demonstration of wkhtmltopdf in Amsterdam.
Hyde initially liked the idea of Wilm developing BookJS, but within a few weeks he tried to convince Wilm to work on other things. Wilm protested, saying that BookJS was not yet usable, and other people within Sourcefabric intervened, which led to Wilm being able to continue work on BookJS until he had a fully working version. Hyde used BookJS himself for his BookSprint business, and while he came up with many feature requests, he never participated in development.
What does it have to do with Fidus Writer?
Given that BookJS was open source with an AGPL-compatible license, it was open for others to use in their respective products. BookJS was used within Fidus Writer, until BookJS became unusable. Wilm has since been one of the main developers of Fidus Writer. Wilm spent a lot more than an average working week on programming BookJS during his time with Sourcefabric. After BookJS stopped working, Wilm programmed paginate-for-print, a basic version of BookJS that works without CSS Regions. Vivliostyle has programmed a much more professional formatter using JavaScript.
Why does this matter?
There is very little funding in the sector of technologies for publishing in the area of publishing, and even less for open source projects. Everything depends on being able to show what one has achieved in the past and how one has a track record of coming up with innovative solutions.
Plagiarism in the area of open source software is no different than plagiarism in fields like art of scientific research. For the is survival of the free and open source software community it is vital that it is stopped, both because it opens the doors to the direct non-respect of licenses as it leads to resources for further development ending up in the wrong hands.
Also, it is important to make clear what it means to develop. Wilm has spent more than a year working on BookJS. There were no contributions from Hyde. It is unclear on what basis Hyde tries to claim that he developed BookJS, but just like paintings are drawn by painters and books are written by writers, software is developed by people writing code (developers), and not someone going to conferences claiming that they did the work, when they clearly didn’t.
How to go on from here?
We wish the best for Hyde, Booksprints and the Coko Foundation. We are happy to let them use our software under the terms of the licenses of the software, just like anyone else. All we are asking is for Hyde to stop claiming he developed software he hasn’t developed. Should any of his organizations have applied for funding based on claims of Hyde having developed BookJS, we expect them to give said funding back and to not apply for more funding from other sources based on these claims.