How can designers and programmers work more harmoniously? How can the tools being created better meet the needs of users? There is a need for designers to have a greater role in the production of the tools that they use, aside from just reporting bugs, requesting features or designing logos for open source projects. This is where the Libre Graphics Research Unit comes in. The Libre Graphics Research Unit (LGRU) is a traveling lab where new ideas for creative tools are developed. The unit has grand aims, looking to bring aspects of open source software development to artistic practices. The programme, sponsored by many organisations in Europe, is split into four interconnected threads:
- Networked graphics
- Piksels and lines
- Abstracting craft
The first meeting, Networked Graphics, took place in Rotterdam from 7-10 December, 2011 and was Hosted by WORM. This second meeting, Co-Position, for which I was present, took place at venues across Brussels from 22-25 February 2012. Co-Position is described by LGRU as:
[...] an attempt to re-imagine lay-out from scratch. We will analyse the history of lay-out (from moveable type to compositing engines) in order to better understand how relations between workflow, material and media have been coded into our tools. We will look at emerging software for doing lay-out differently, but most importantly we want to sketch ideas for tools that combine elements of canvas editing, dynamic lay-out, networked lay-out, web-to-print and Print on Demand.
The meeting saw the coming together of many international artists, theorists and developers for four days of work around this subject. As some of the sessions of the meeting took place simultaneously I'm unable to give a full synopsis of the event. Instead, what is presented below are some of the key issues raised at the meeting.
The subject of copyright cannot be avoided when discussing digital art and collaborative practices. There is a definite need to foster a safe and welcoming environment for artists and designers to produce, share and remix their work. Licensing of artwork under Copyleft licences - such as Creative Commons - helps to create this environment.
In his presentation, entitled "Libre Workflows - A Tragedy In 3 Acts", Aymeric Mansoux was quick to point out that Creative Commons licences do not cover the source of the artwork. To put it into context, a JPG is covered by a Creative Commons licence but is the XCF/PSD file? Mansoux also considered what is actually a finished piece of artwork? In a remix culture is an artwork ever finished? Mansoux refers to this quote from Michael Szpakowski for further elaboration:
I've found it helpful to think of any artwork, be it literary, visual art or music as a kind of fuzzy four dimensional manifold. So the "complete" artwork is the sum of all its instances in time, and all epiphenomena. The entire artwork, seen this way, is a real and precisely enumerable sum, a concrete, not imaginary, set, which could be knowable in its entirety by something long lived and far seeing enough.
From their home town of Porto, Portugal, Ana Carvalho and Ricardo Lafuente produce Libre Graphics Magazine with ginger coons who is based in Toronto, Canada. For the production of the magazine they use Git, with their repository being hosted on Gitorious. As a tool for sharing files between collaborators Git is very useful. However, they explained that they feel they are not making effective use of all that Git has to offer. Part of this comes from the complexity of using Git. There are more than 140 commands in Git, each with their own unique function. These are usually entered via the command-line, but there are a number of programs with a Graphical User Interface (GUI) available. Programs with a GUI are usually favoured over command-line programs as they remove some of the complexity. Carvalho and Lafuente have found, however, that many of of these GUI programs simply replace commands with buttons, which doesn't remove any of the complexity in using Git. What is needed is an easy to use specialised tool for the production of art.
Visual Versioning Work Session
In this work session, presented by Ana Carvalho and Eric Schrijver, the work group imagined how to adapt existing version control tools to meet the needs of artists and designers. The session began by taking a look at how people currently implement version control. A common practice is to manually make backups, renaming files to differentiate between stages. This can be an effective way of making different versions, but it doesn't address other issues such as making comparisons or merging changes. The ineffectiveness of these manual methods is soon very apparent. The work group was introduced to the Open Source Publishing (OSP) Visual Repository viewer, which begins to respond to some problems with current version control systems by providing thumbnails of files in a repository.
Using this as a basis we began to look at other functions that the OSP Visual Repository viewer should have, such as the ability to compare graphical files in different ways and to revert back to previous versions or merge versions. Although there was no time to produce working code we did seek to address the complex task of merging and comparing not only the ouput file but also the working files (svg/xcf/psd).
Every good work of software starts by scratching a developer's personal itch.
This quote from The Cathedral and the Bazaar by Eric Steve Raymond could not be more accurate in describing the motivations behind the development of Laidout, developed by Tom Lechner, a comic artist from Portland, Oregon. Perhaps one of the most impressive software demonstrations of LGRU, Laidout is a program for laying out artwork on pages with any number of folds, which don't even have to be rectangular.
SVG from scratch
In an attempt to devise new tags that can be added to the SVG specification, Michael Murtaugh and Stephanie Villayphiou presented a work session that looked at the different ways language is interpreted by both humans and computers. To address this the work group took part in a task that saw them act as an interpreter of commands. With nothing more than a list of tags used in SVG files the work group would attempt to construct shapes.
The results varied from person to person and highlighted an important question: How can computers interpret ambiguity
Using the Richard A Bolt "Put that there" demonstration, Murtaugh showed how human-computer interaction is still based around using very clear, unambiguous commands that can be easily interpreted by computers. In SVG only the most basic of shapes - rectangles, circles and lines - are represented. But, as the work group participants asked, could there be tags to represent more complex shapes, such as a horse?
On the final day of the meeting I took part in a roundtable discussion, chaired by Angela Plohman and featuring myself, Stephanie Vilayphiou, Camille Bissuel and Ana Carvalho. The discussion first went over all that we had achieved over the four days at the meeting, and then the discussion focused on how and why we share our artwork. Expanding on the earlier quote from Szpakowski, how can we make sharing all of our artwork - including the early stages and inspirations behind it - an easier and integrated part of making artwork? In addition to sharing our final, "finished" artworks do we want to also share our processes and ideas behind the artwork? More importantly, can software easily aid this?
Other topics debated in the discussion revolved around opening up our artwork and processes to others. By opening up the development process of our artwork do we do so to invite collaborators and contributions or just observers? The Blender Open projects, for example, are highly regarded as an example of the work that can be made using open source software. The files used to make these projects are are released upon completion of the project, but the development process remains closed to the team of artists and developers. Would opening up this process to contributors add any value or could having too many ideas dilute the original vision of the project.
Although no conclusions around these topics were made, it was nonetheless important for everyone at the meeting to think critically about their practice
The Future of Layout
A concern of mine is that research is not always acted up on and exciting possibilities exist only as theory. However, I feel that the approach of Libre Graphics Research Unit, which combines research and practice, will ensure that the work undertaken at the meetings is implemented. It is actively working with developers and users to try and create solutions.
At the Co-Poistion meeting not one final product was made, but the initial vision for the future of layout was formed.
The next meeting, Piksels and Lines, takes place in Bergen, Norway and is organised by Piksel.