English | Fran\xE7ais |

FilmForge

This was the first stage proposal of the FilmForge concept. It looks like http://wiki.koumbit.net/FilmForge will be the new spot for project development. If you want to get involved please go there! - schock

Core Idea

Free/Open Film project management / user community. Like Sourceforge (http://www.sourceforge.net) but for filmmaking.

Short Description

FilmForge would be a community of filmmakers working with a common set of project management tools, for each stage of film production, that enabled and encouraged project initiators to think about and set their own parameters for how open or closed each step of production would be. For example, FilmForge could include:

script tool. This would be a wiki; project initiators could determine levels of access to collaboration, anywhere from no collaborators (the script is already complete) to invited collaborators (project initiators invites particular individuals and they get script editing permission) to registered users to completely open (anyone can access and modify the script). Once the script is 'done' (to be determined by either project initiator or all participants), it can be 'frozen' (or forked) and the project can move to the next stage: storyboard. Examples: for a fully functional, open source preproduction package see http://www.celtx.com. For a simple script to format wikis like screenplays, see http://tinyurl.com/hzond.

storyboard tool. Based on the script, project collaborators (again, open/closed determined by project initiator) use this tool to generate storyboard from the script.

shot (media) acquisition tool. Based on storyboard, participants submit media (footage, stills, audio, etc.) to the project's media bin / pool.

logging tool. (maybe combined with acquisition). This allows participants to view, select, tag, and rate media that has been submitted to the project. With video clips, needs to be able to allow browser based submission of in and out points. Ideally this is done directly within the clip viewer (see example of this capability at http://jumpcut.com; http://eyespot.com) but can also be a simple webform for entering timecode in/out.

editing tool. Again, remember that project initiators can open or close the editing stage as much as they want. Maybe they have a particular editor they want to work with; maybe they want it completely open to public. Maybe only some segments of the project are open, while others are closed. In any case, the idea is that whoever is editing can drag media files from the media bin/pool into project timeline. (See http://eyespot.com for kind of clunky, but working, implementation of this idea; http://jumpcut.com for a slightly better version - and rabble tells me it's built on top of FOSS Gstreamer: http://gstreamer.freedesktop.org/).

transcript and subtitling tool. Global.Example: see http://www.dotsub.com) or Subtitles Workshop which seems to be a freeware see http://www.urusoft.net/downloads.php?lang=1

distribution tool In addition to distro of the films (p2p), a tool to facilitate large numbers of people participating in arranging screenings, sales. (example: see http://crm2.metamute.org/)

note on web based vs. user side from a chat between dongwon and sasha. write up in clear format

instead of or in addition to shared edits via a browser, people could edit locally and the EDL gets exported to a shared database?
sure... how about this? didn't i tell you? we can edit via a browser with small size of files. when all are done, the final cut can go both some web distri channels and web-based DVD-making server.
which make final cut with higher size of file for dvd based on editing code.
well, it could also be some kind of plug-in for the editing software that will be packaged with Ubuntu Studio
so we don't have to reinvent the wheel, with any editing software
if we have same time code btw small and big size of files when we upload both.
but it could be both 
yes, it's important to figure out how to track files across the project
oh, sure... i mean plug-in.
ok... getting more and more ideas. but how/who implement them is the question here, too.
for example, people can choose to edit via web browser (in which case they are editing using small files, not the full size source material as you say)
or, they can edit on their machine locally
and either way, the EDLs get sent to the project database, which lives on the server
so, it's kind of like the EDL is a wiki that a lot of people can write to
that's just a tiny (text only) file
then, when people want to view it,
they can either just download the EDL and watch using local media files
or maybe the server can generate a low quality (flash?) version to be seen on the site
what is tricky is rendering, unless the software is powerful enough to do all the effects rendering in real time?
also, if people are 'subscribed' to a project then they can automatically download any new source material that gets added?
final cut pro create rss, right?
as well.
and to save the server, all the people who are part of the project are sharing the source media files via a P2P client

Theory

Yochai Benkler usefully investigates the conditions under which commons or peer production can succeed. He thinks that it requires widely distributed production power, a high base of knowledge/skill, and sufficient modularity and granularity of production process. Modularity means that the process can be broken down into chunks that can be swapped in and out. Granularity means that these chunks can be small enough and manageable enough so that large numbers of people can take a small chunk (while a smaller number of people will deal with larger chunks).

Can the Motion Picture production process meet these conditions? I would argue that it can, and that it is possible for a vast, distributed network of people to produce feature length films that are successful by all criterion of filmmaking.

This process can theoretically be used to produce many different kinds of material. However, there are real differences between genres in modularity. For example, individual actors are not modular, unless they are CGI (in which case the code for the character can be passed around to different people who all create different shots, then reassemble later). A story that closely follows one individual actor can't be distributed in the same way. To begin with, documentaries and perhaps animated films are the genres that most obviously lend themselves to this kind of process.

A documentary is ideal b/c it includes a large number of fine grained tasks that may not be present in other genres, in part because B-Roll may be drawn from such a wide range of sources. It fits perfectly to have a huge list of B-Roll clips, stills, scanned documents, and so on, each of which could be found and submitted to the project media base by a different individual.

It also is the perfect kind of process to do films/documentaries that examine some aspect of globalization, where you need to pull in footage from around the world.

I would guess that we would probably see the following pattern across projects:

* A core group of 5-50 people will probably do most of the work * A much larger network of 500-5,000 people might contribute some small element * Need a good system of peer review for each stage (to weed out the crap)

Example

For example, a documentary on human migration and migrant rights is one ideal subject.

A nationwide network of video makers, programmers, and activists, and May 1st last year triggered us to take an initial step of putting out a call for independent documentary makers to contribute footage of mobilizations (here's the call: http://www.nettime.org/Lists-Archives/nettime-l-0604/msg00078.html)

So, we now have a huge pile of mobilization footage. (here's the footage: http://www.videobomb.com/tags/show/noborders).

This is cool, but of course it's only one part of being able to tell a compelling and deep story about immigrant rights. To take it to the next level, we can try, as much as possible, to experiment with this distributed production process.

[task: watch submitted footage and mark selects]

Same goes for audio tracks.

Transcipt and subtitles will practically write themselves.

What's the distribution model?

Key function: 'filmforge' Here's the key function of the process: tracking and organizing. Think of the role that sourceforge.net plays in the development of all kinds of FOSS projects.

Technical Requirements

* Everything is FOSS * Need to examine and select CMS to build on. Likely candidates: Drupal, Plone. * Flash or not to Flash? * Video codec prefer ogg vorbis, but either will need to convert people's files or support as many as possible. * If we need to host media, can probably use internet archive

Budget

* There is some funding (in the 10,000 dollar range) ready and waiting for developer.

Process Chart

* TBD

Prior Art/Existing elements

• comparison of existing video sites: http://www.dvguru.com/2006/04/07/ten-video-sharing-services-compared/
• editing in the browser: http://www.jumpcut.com; http://eyespot.com; http://www.motionbox.com/
• editing streams (!?!): http://octune.org/Partage/
• Internet Archive Open Source Movies
• See http://en.wikipedia.org/wiki/Open_source_movie_production. There are a number of 'open source' movies that were produced entirely with open source software and/or released under an open content license. For example:
• An open source movie about open source video: http://www.masternewmedia.org/news/2006/04/21/grassroots_collaborative_online_video_editing.htm
• A Swarm of Angels, a movie to be released under the Creative Commons CC-NC-SA (Creative Commons Attribution-NonCommercial-ShareAlike) license
• God's Companion, a full length 3D movie made using open source software
• Elephants Dream, a 3D short film made using open source software, and released under the Creative Commons CC-BY license
• Boy Who Never Slept, a full length dramatic comedy released under the CC-NC-SA (Creative Commons Attribution-NonCommercial-ShareAlike) 2.5 license

But these are not really what I'm talking about, because they are all produced by small teams of people without mass distributed participation in any element (script, shooting, post). Swarm of Angels comes closest perhaps.

* Various Indymedia documentaries, made with large numbers of people (typically 50-150) contributing footage and smaller numbers (5-15) contributing editing and postproduction tasks. These are also usually subtitled in many languages by a distributed network of people. Most of these are shot without a coherent script beforehand, and document protests and mass mobilizations. Examples: This is What Democracy Looks Like, Kilometer Zero, Zona Rossa, the Miami Model. These are closer to the idea I'm proposing b/c they spread production labor more widely. However, they typically so far have failed to involve large numbers of people in post production, and except for shooting haven't been able to really reach that stage beyond a core project team (haven't reached fine grained modularity that could bring in a thousand contributors).

drupal video tools

• Drupal Video module: http://drupal.org/project/video
• Drupal Video module demo: http://www.varesano.net/video-demo/
• OpenPackage Software (openpackage.biz) has begun development of an OSS (open source software) Flash video module for Drupal.
• 'YouTube' module for Drupal: http://www.tunaspecial.com/?p=162
• OpenPackage Software (openpackage.biz) has begun development of an OSS (open source software) Flash video module for Drupal.

(activist) media-related FOSS

• [http://pubforge.org PubForge: Open Source Software for Public Broadcasters]
• [http://ourvideocms.sourceforge.net/ Ourvideo CMS(soureforge.net)]
• [http://en.wikibooks.org/wiki/Video Video section from Wikibooks]
• [http://www.dynebolic.org Dynebolic]: Bootable linux CD for media activists. Awesome tools for capture, editing, streaming

people

Here we collect the bios of the people interested in the project:

* TheAnarcat, part of [http://koumbit.org Koumbit], a worker collective based in montreal that develops drupal modules and software. Relevant projects:

• http://media.mcgill.ca/en : deployment of the video module
• http://radiovision.ca http://cibl1015.com: online radio sites
• many more...

The way I see this, we could already install a simple drupal with some default (cool) theme in 5.0, with the [http://drupal.org/project/liquid liquid engine] (for wiki collaboration), and the video module, for ... videos.

For me, the big things missing in drupal for this are:

* realtime (text) collaboration "a la" gobby (http://darcs.0x539.de/trac/obby/cgi-bin/trac.cgi) * online video montage (jumpcut.com etc), not sure this is relevant: montage can be done offsite * thumbnailing in the video.module (maybe it already exists, should be fairly simple to do, provided the relevant software is available on the server)

Also, I think some of the requirements of this project could be funneled "outside" of drupal to projects that are more relevant, very quickly, i think about Dyne:bolic as a likely target, since it is a very good platform for multimedia stuff... -- TheAnarcat - 30 Jan 2007

* In, Dongjun (Feb. 2007)

Basically, he thinks that information and knowledge should be shared, and he lkes linux and free software. He also does web programming with php + javascript, and can manage the server. He likes to ride a bicycle. Since last fall, he starts to become vegetarian. Nowdays he is falling into socialist SF novels. He feels sorry that he can't speak english nor do image-retouching works, that's what he try to practice in this year(2007).

He has worked in website-building team for a variety of trade unions in Nodong.net(Labornet in s. korea, http://nodong.net) for 2 years. He is particularly interested in the issues like "information divide" and ip-justice. While he is a member of cultural department of Korean IT Industry Labors Union(http://it.nodong.net), put his efforts to protect IT worker's human rights. He is also a member of Media Cultural Action against neoliberal globalization, taking care of tech aspect of internet live broadcasting.

since last year(2006), he has been mastering the drupal, now gives folks the word to use it. Recently he develop one of modules for programmng of the internet broadcasting, which has some bugs at the moment. As for filmforge, he may want to work as an assistant developer.

* Kim, jieun (Feb, 2007)

very soon to be updated.

* jo, dongwon (Feb. 2007)

jo, dongwon is an indedendent researcher and alternative media activist. He was the policy research director at MEDIACT(http://www.mediact.org), a South Korean public media center last 4 years(2002-2006). His roles included researching the independent/alternative and public media movements; the objective of which is the establishment of local media centers, community radio station, public access channels, progressive media literacy movements and audiovisual media studies. Now he is one of volunteer activists of Media culture Action against neoliberal globalization(http://gomediaction.net), while collaborating with KIFV(Korean Independent Film/Video association), Jinbonet(korean progressive network), Nodongnet(labornet), and Cutural Action etc. And he is in PhD course in Cultural Studies, ChungAng University.

since last year(2006), like dongjun, he has been using the drupal, now he manages a couple of drupal-based websites. As for filmforge, he may want to work as an PM(project manager) for korean participation and development of korean-specific filmforge system by developing extra modules(if needed) or organizing translators etc. And he can coordinate to directly beta-test/use it through a pilot project such as ip-justice collaborative film-making project under being planned.

other

NOTE: expanding p2p to the physical realm: http://blog.p2pfoundation.net/?p=434

Additional video sites: http://dabble.com/

-- SashaCostanzaChock - 10 Oct 2006