The commons

In September 2007 the German Federal Archive already made 113,000 images available on their own online digital archive. In total the Federal Archives keeps approximately 11 million still pictures, aerial photographs and posters from modern German history. The cooperation with Wikipedia is the next big step for the German Federal Archive in opening up the archive, as the vice president of the German Federal Archive Dr. Angelika Menne-Haritz said during the press conference.

Photo: Raimond Spekking, Creative Commons Attribution ShareAlike 3.0.

The photos are not of the highest resolution, about 800 pixels on the longest side. But, this is an enormous addition to the commons. According to Wikimedia, the repository of free content images, sound and other multimedia files on Wikipedia, the donation by the German Federal Archive of 100,000 images is the single largest one to Wikimedia Commons so far. This is even more than the archival project Flickr Commons makes available now in cooperation with 16 archival partners around the world.

Click here for the image gallery: http://www.bild.bundesarchiv.de/

Photo: Mitglieder des Deutschen Reichstag, German Federal Archive (1889). Author: Braatz, Julius. Creative Commons Attribution ShareAlike 3.0.

Creative Commons License

The images by the German Federal Archive are licensed Creative Commons Attribution ShareAlike 3.0 Germany License (CC-BY-SA). This means that you are free to share and remix the images under the condition that you give attribution and spread this with a similar or compatible license. The Federal Archive can do this because they own sufficient rights on the images to be able to grant this kind of license. To use such a free license for archival material is really exciting. Few archives work with Creative Commons licences. One of the rare examples is the McCord Museum and the Brabants Historisch Informatiecentrum. And, the archival project Flickr Commons works with “no known copyright restrictions”.

Persondata

The other part of the cooperation between the German Federal Archive and Wikipedia is a tool for linking people from a list compiled by the Federal Archive to the German Wikipedia Persondata and to the person authority file of the German National Library. Something German Wikipedia has already been doing since 2005. Around 27% of 100,000 photos is already done. The expectation is that because the cooperation is now public, the tempo will speed up. Moreover, the users will add new information to the images. You can find the To Do list here.

Conclusion

Though projectleader Creative Commons Germany, Markus says that this is only a small revolution for German notions, this could very well set an example for other archives to make their content publicly available and therefore grow bigger. It will be very interesting to see where we can find the photos and in which (rich) context. Because that will make a strong argument for archives to experiment with this.

The development of the project started with a kick-off meeting at Knowledgeland in Amsterdam, earlier this month. The aim-of-the-day was to map the (open source) digital video solutions that are available today and to get feedback. Eight experts in the digital video field where invited to this informal brainstorm session. After an introduction of Images for the Future and the Open Images project plan, the invited experts gave inspiring presentations of their current work. At the end of the day there was a general discussion about the project plan and the first steps that ought to be taken.

Please find a report of this day below. Open Images aims to launch a Beta release by the end of the year.

Bits on the Run (Jeroen Wijering)

After building the worlds most used Flash video player on the internet (JW FLV Media Player), Jeroen Wijering used his experience to co-found Bits on the Run. This is a “hassle-free” video hosting service for all types of video content (video-centric), that charges its users based on their usage of bandwidth. The service offers a broad set of features, including; hosting, statistics, streaming, encoding, subtitling, annotation, audio descriptions, skins, embedding, downloading, RSS, etc. The service also offers an API (that mimics the one Flickr uses).

Pad.ma (Jan Gerber & Sebastian Lütgert)

Jan Gerber and Sebastian Lütgert develop many digital video projects based around the idea that digital media have the potential to make the complete history of cinema accessible online. This sometimes means they move in a legal grey-area. 0xdb is a database of movies that have been shared via peer-to-peer networks. It uses the BitTorrent protocol to extract information about the movie, of which the database creates meaningful representations (timelines, previews, etc.). Pad.ma is a derivative of the 0xdb technology, but with some important differences. It is a collaboration with Indian NGO’s, activists, and documentary filmmakers and only contains legal content, since all material is open content. It serves as a decentral searchable footage archive – with manually edited meta data – and offers downloads of the content using peer-to-peer technology. Pad.ma is released as open source software under GNU GPL and uses the Ogg Theora open standard video format.

Tribler (Freek Zindel)

Tribler is a peer-to-peer file sharing application with a social component, developed by TU Delft and funded by the EU. Sharing and finding (both legal and illegal) content via BitTorrent is enhanced with recommendations, tagging, and anti-freeloading mechanisms. On top of this Tribler offers additional functionality for video, including; instant previewing, video-on-demand and live streaming (all via peer-to-peer technology). At any given time there are 4.000 people online using the application, and there are 150.000 torrents in the system. Tribler is released as open source software under LGPL.

SURFmedia (Alexander Blanc)

SURFnet recently launched a complete video distribution platform called SURFmedia, offering live streaming and video-on-demand for the higher educational sector on the SURFnet 1 GB network. Because of its user group SURFmedia is a closed system, and was hence developed from scratch. At the end of this year the core of the platform will be released as open source software, under the name SURFmediaCore. There will also be some pilot projects with third parties who will develop their own front-end on top of this core.

General Discussion (all)

During the general discussion the experts express some concerns about the size and goals of Open Images. They stress that the project shouldn’t try to reinvent the wheel, but should rather make use of the (open source) video solutions and (open content) platforms that are out there. The experts advise to seek co-operation with and/or to build upon existing projects with a similar focus. Mapping the possibilities for this is indeed our primary goal for this kick-off meeting.

The experts are also sceptical about the assumption within the project plan that there will be a substantial interest from the general public to invest time and effort in the creative reuse of the open content. They ask how “interesting” the available material actually is and stress the need for a careful selection. Furthermore they suggest that the project should focus more on showcasing and spreading the material, because to them this implicitly seems to be the primary goal of the project. The experts advise to start with a (thematic) selection of the material on Open Images and to distribute this selection through as many outlets as possible, to first see what happens. This implicates that the project team at first needs to invest in developing a good interface for showcasing and distributing the open content, and exchanging material with other outlets. Technically facilitating the creative reuse of the open content on Open Images can – if requested – always be added within a secondary (or parallel) development trajectory.

Blender is an open source and open content 3d modeling software package that uses a pre-financed model, in which customers can preorder a movie, from which the profits are being used to create the said product. Other ways of income include foundation community, offering documentation as to how to work with the Blender software, fundings and commercial sponsoring. The main goal of the Blender Project is technological innovation. Ton Roosendaal emphasized that open content doesn’t mean that you don’t have to pay for the goods, but rather that it could be used in a free way. That is, to use that content in any way you see fit.

Steal This Film is another open source project, with the emphasis on community based funding. Steal This Film, both part I and II, are a documentary style film using clips from other movies and distributed on the BitTorrent network, using one of the largest trackers. Jamie King claimed that peer-to-peer distribution exceeds all other methods, because it is ‘…the most significant data transfer in the world’. There are hardly any costs in distributing via peer to peer, neither for the producer nor the consumer. Therefore, it’s the most easy way for a consumer to get a product.

Felix Stalder started of the panel, by acknowledging the models used by Ton Roosendaal and Jamie King and added a third model in using advertisement combined with open source and freely distributed content. This does not mean that all creative industry will be affected, but only one type of business model. This being the traditional consumer-producer model, in which the consumer directly purchases the product from the producer.

After this, the panel moved to the topic of quality, content and distribution. Jamie King stated throughout the panel that quality is not the single most important aspect of your product. Distribution is equally, if not more, important. If only you put it out in an easily obtainable manner, people will be inclined to look at your product. Jon Phillips contributed by claiming that content is equally important, after a discussion on the costs involving peer-to-peer distribution. It must be acknowledged that there are inevitable costs in running networks, he stated. Jamie King quickly responded by saying that this is included in the ADSL-connection fee.

After Peter Kaufman asked him to comment on the current situation of piracy in China, Jon Phillips responded that the illegal sales of pirated material on the black market had diminished and that the focus was now on streaming HD content via broadband. Jamie King referred to Stage 6, a similar initiative but without the advertisement, which went down over a month ago due to bankruptcy. Jon Phillips replied, and said that this initiative could be feasible because of this use of advertisements.

The last major topic that was raised concerned the narrative of the Big Buck Bunny film, just released by Blender. Anthony McCan asked why open source productions can’t steer clear from the use of violence, as violence evokes violence. Ton Roosendaal replied that this was the best way to show the technical abilities capable by the Blender rendering software and that creative freedom in the way these technological abilities are displayed is highly regarded.

The music and film industry are facing similar issues but have yet to find a grip on the situation; the content is more often than not available for free through peer to peer networks therefore a large part of the incentive to go to a shop and buy a cd or film has vanished. As we are digitizing vast amounts of audio-visual cultural heritage we are facing the same questions: what models can be developed that fulfill the need for broad accessibility for the public while securing a solid return on investment for owners of the material (authors, producers, directors, etc).

Some, like Chris Anderson in his soon to be released new book’ Free’ build an entire economic theory based on the notion that free’ will be the leading model for media due to the vanishing marginal costs of distribution via the internet. The new model that rises from the ashes will be a model where the content or service is free, at least for the user. Google of course is a great example of a company that has turned ‘free’ to it’s advantage; the service is free to users while advertisers are the paying customers. At the core a beautiful system as the more you use the service the more revenue it generates for the service provider. Keeping the attention of the viewer is key in the ‘economy of abundance’, so you better make sure the service you develop is so appealing that users get hooked on it. In fact, if this becomes the case, there may be an opportunity to upsell them from freeloaders to paying customers by adding a an additional layer of services or privileges. This freemium model (term coined by venture capitalist Fred Wilson) has quickly become the leading model for web 2.0 companies like Flicrk and Linkedin. Interesting fact is that the rule of thumb is that the 1% of paying users supports the rest.

The crux of developing business models in this economy of abundance, where content is free, seems to be to tap into values that people are willing to pay for. And those values may not be the same as in the old days where content was king. Kevin Kelly calls them ‘generatives’:

‘’A generative value is a quality or attribute that must be generated, grown, cultivated, nurtured. A generative thing can not be copied, cloned, faked, replicated, counterfeited, or reproduced. It is generated uniquely, in place, over time. In the digital arena, generative qualities add value to free copies, and therefore are something that can be sold.” Think ‘trust’ or ‘personalisation’. In his blog ‘Better than free’ Kevin distiguishes eight of them.

So how does this translate to our audiovisual digitization adventure? Will the specific characteristics of cultural heritage lend itself to open content models like advertisements (google just released a beta service of video advertisements: Adsense for Video ), Freemium services or even community supported businessmodels?

We are hosting a workshop on this topic during the Economies of the Commons conference on saturday April 12 2008 in Amsterdam to investigate the options.