Pad.ma, short for Public Access Digital Media Archive, is an Indian open video archive. All videos on pad.ma are freely available to watch or to download for non-commercial use, and the website is open source under the GPL, or General Public License.
Part of pad.ma’s mission is to provide an opportunity to view material that lives “outside the edits.” The available footage isn’t necessarily what you would see in films or other video sites, and is densely annotated with text and transcripts.
Pad.ma is a leader in contemporary digital archiving practices, and is currently at work on a new software framework and interface for the summer of 2010. According to the site architects, “the focus here is on annotation, cross-linking, downloading and the reuse of video material for pedagogy, research and reference.” This makes it a valuable resource not only as a free and open source video site, but as a developing library of unique and educational content.
We sat down with the folks behind pad.ma to discuss the site’s development, and the relationship between video, art, technology, and viewership online. In the process, we learned more about the conceptual background for their design, online video as it relates previous generations of the moving image, and what to expect from pad.ma in the future.
Open archives are an important topic at the upcoming Open Video Conference, and a number of developers and archivists from Pad.ma will be represented. By promoting free and open access, publishing content that may otherwise go unnoticed, and seeking new meanings and uses for material in a digital format, they are actively engaging with the process of building an archive for the 21st century. Read on for the interview.
How did Pad.ma come together?
Pad.ma: At present Pad.ma is cared for by 0×2620.org (0il21.org) from Berlin, the Alternative Law Forum from Bangalore and CAMP, based in Mumbai. ALF had been doing significant research around IP and Piracy, including authoring for example the Guide to Open Content Licenses (2004), prior to pad.ma‘s inception. Oil21.org, as it was known then, had been working with digital distribution and its effects: textz.com, v2v.cc and Pirate Cinema etc. for a while. When we all met again in Berlin during the G-8, in the summer of 2007, they had just finished the first version of 0xdb.org. In Bombay, CAMP’s (earlier chitrakarkhana.net) own autonomous video projects had emerged from a critique of conventional documentary form and politics, including looking at art more generally, as a “distributive” practice.
Another starting point came via Majlis, an established NGO working in Bombay in the cultural field, who had an on-going video archiving project, Godaam (Godown). Over many months of conversation with the NGO, with documentary filmmakers and other cultural practitioners, the idea of making the footage available in the public domain (which was not without resistance) was pushed forward. ALF and Oil21.org entered the conversation at this stage, and a formal collaboration towards an online video archive was born. Point of View, another NGO based in Mumbai also came on board.
In the winter of 2007 we all began working intensely on the code, on the content and on the licensing framework in Bombay. Pad.ma emerged from this process and its beta version was launched in February 2008, with over 100 hours of content, fully under the GPL (with a specific PGPL, pad.ma general public license for the content).
Can you talk a little bit about the use of text on your site? How does it fit in to the overall project for people who may be less familiar with interacting with video in this way, or contributing to it?
Pad.ma: In the days of silent cinema, there would often be a person, a ‘Banshee’, who would play the role of providing the sound/commentary/annotation to what was happening on screen. His version of course sometimes elaborated the image, sometimes it was in contradiction, but often brought to notice things that would sometimes be missing. We can also see a similar approach in the work of thinkers like Benjamin who saw their task as an excavation of the debris of modernity to bring to light the simultaneously enchanted and quotidian aspect of a proliferating commodity culture. In the era of the hyper image, one of the challenges is to think through what a banshee technology would consist of, of what writing practices could bring to light a different relation to the image and the idea of the archive. The challenge for us is how to intervene in an environment where the image is collapsed into an economy of information. I don’t think we have an answer yet, but Pad.ma is an experiment at trying to look at the practices of reading and writing as existing simultaneously with the image, and not merely as a supplement to the image.
What are the uses for this author or user generated text?
Pad.ma: The user on Pad.ma can locate precise moments in the video either through transcripts, description, keywords or annotations by academics, historians, filmmakers on the video, and also download this specific clip. The user can display, re-use and remix the videos. Multiple users can annotate on all the videos – and these annotations act as separate layers on the video, and explain, refute, contextualize the video. This allows space for many meanings to be remixed and altered, rather than only a singular, factually verified account. Pad.ma is a response to many of the anxieties around visual cultures and image politics as well; that if our experience of the media saturated world is not any longer about only seeing images, but constantly reading and seeing visuals, where reading becomes a process that is compliant to procedures of hegemonic power (technological, political, advertising and corporations). In the Indian context, Pad.ma is also a response to censorship or blocking (and to over-saturation of images that fills the void).
The text on pad.ma is of two primary types, apart from the basic metadata and clip descriptions that are provided on the info pages.
There is usually one layer of transcripts, where the motivation is to provide a transcript or translation (into English, or now other languages) of what is being said in the video. This is meant to provide deep text-based searchability across the archive, and an account of the video that is asubjective, or “true” to the video material. We are starting to experiment with voice-based transcription and other ways to speed up this process.
The second type, annotations, have a completely different function. The first layer of annotations is done by the original contributor, who may or may not be the author of the video. There are many cases for example of “found” materials, or materials such as film clips that are being annotated by scholars or enthusiasts. At the moment the bulk of textual annotation is done, as part of the archival process, by the original contributors. There is not much of a “comment” culture on pad.ma, and this is perhaps a factor of the interface, which requires you to mark a section of video, and be logged in, before you can annotate it. This is partial to more engaged, longer text pieces that the kinds of comments you might find on youtube or nico nico douga.
The most interesting types of user-contributions to the archive so far are text-pieces that span several, say a dozen, video clips. These are either invited contributions from people writing about a particular issue within the archive, or people writing elsewhere who can reference sections of video on pad.ma, by giving a link with timecode in-out, such as: http://pad.ma/Vfrbgdjr/00:09:19.320-00:09:58.919
Pad.ma “layers” all annotations and transcripts, so that when you play a video you come across a number of different textual entries. This means that we are not moving towards a canonical ‘description’ of the video material through text, but a multitude of voices speaking about and through it.
Recently, you conducted a workshop in Beirut to share your work with archival groups there. What do you see as Pad.ma’s role with these kinds of workshops, and how is it expanding?
Pad.ma: It has been a fascinating experience working with groups and archives in India over the past two or so years, and to get people’s perspectives on the possibilities they see for a text-annotated video archive. Workshops such as the one in Beirut are vital to bringing in more perspectives, as well as enlivening the debate around online archives in general, including issues of property and privacy. Going forward, we’re doing a workshop in Bangalore later this month that aims to work with content-makers, but for the first time we’re also having an intensive workshop sessions with coders to explore uses of the Pad.ma API and also re-imagine use-cases for a heavily annotated archive such as pad.ma. So, I think in some cases we provide a valuable service to people with content that may otherwise be lost due to years of sitting in cupboards, and on the other hand, we also hope to open up the possibilities of usage of this material. Some of the most interesting possibilities of the archive will emerge once people really get into using the website to do what they want – whether that means using the API to create a mash-up on their own site, or write new kinds of stories across video materials, or things we haven’t thought of yet.
What’s to come? Are you looking for any help, contributions, sources, technical work?
Pad.ma: From the onset, the software has been built as an open source project, and as such we’re extremely happy to have people use it, fork it, write patches, etc. The current code-base and bug-tracker lives at http://wiki.pad.ma/. The upcoming release is a complete re-write, being developed using the Django frame-work. You can follow development and get access to all the source at https://wiki.0×2620.org/wiki/pandora. If you are interested in development, please join the mailing list @ https://mailb.org/cgi-bin/mailman/listinfo/dev and you can generally find us on IRC @ #pad.ma on irc.freenode.net.
