TheFactory an overview of editorial toolchains

Situated and Collective Factories of Web-to-Print

Alternative editorial practices, code cultures and web-to-print

This website presents a selection of editorial projects that place code and programming at the centre of printed publication production, redefining the toolchains of editorial production.

The publishing process involves political choices, cultural positioning, and a vision of how knowledge and printed forms should circulate. It is within this space of inquiry that The Factory sits, a non-exhaustive state of the art of printed productions made with web technologies.

Producing printed artefacts : books, journals, catalogues, posters, from web languages : HTML, CSS, JavaScript. This is what we call web-to-print. This approach breaks with the dominant production workflow built around proprietary software such as Adobe InDesign. It is rooted in a culture of open source software, modularity and interoperability¹, and brings together communities of designers, developers, researchers and publishers who share a common conviction : the tools we use are not neutral, and reclaiming them is an editorial act in its own right.

A long history, a persistent tension

Digital publishing did not emerge from a sudden rupture. It is a sedimentation of techniques and practices spanning more than half a century, from the first typesetting systems such as TeX (1978) and LaTeX (1984), to page description languages such as PostScript (1982) and PDF (1993), through to contemporary web-to-print tools². This genealogy reveals something essential : a permanent, unresolved tension between two logics of making.

On one side, proprietary tools oriented towards visual composition : PageMaker (1985), QuarkXPress (1987), InDesign (1999), built on a WYSIWYG logic that privileges immediate rendering over content structure. On the other, tools built on the separation of structure and form : semantic markup languages, stylesheets, version control systems. The latter inherit a philosophy rooted in Unix, a multitasking operating system developed at AT&T’s Bell Laboratories from 1969, founded on modularity and user sovereignty : each programme does one thing precisely and can articulate freely with others. Web-to-print firmly inscribes itself within this legacy.

This tension is not merely technical, it is political. The capitalist concentration of the publishing industry since the 1980s has produced structures whose profitability demands have marginalised challenging publications and niche works³. Adobe’s dominance in desktop publishing reproduces the same logic of lock-in at the level of production tools. Adopting a publishing workflow based on open source software and web technologies thus becomes a gesture both against editorial rent and against software rent and for technical and editorial sovereignty, for the collective ownership of production tools.

This political dimension is not new. Editorial avant-gardes : fanzines, artist magazines, experimental publications, have long anticipated these hybridisations between print and digital⁴. What the web brings is less a rupture than a common, open infrastructure for practices that already existed at the margins.

A profession in transition

Graphic workflows bring together actors with very diverse competencies : designers, developers, copy editors, printers, whose interaction is still too often shaped by industry pressure rather than a holistic and collaborative approach. Opening up and sharing knowledge so that each link in the chain can reclaim it : thinking about connectivity within these workflows becomes an additional mindset to integrate into the designer’s toolkit.

This shift is above all cultural. Practices built on tools drawn from web development, Markdown or AsciiDoc for semantic writing, Git for version control and collaboration, HTML and CSS for layout, do not merely propose a substitution of tools. They propose a paradigm shift in how we think about publishing : no longer as a finished object to be produced in a final layout stage, but as one artefact among others emerging from a single process. « The difficulty of grasping this change is cultural before it is technical.»⁵

This paradigm shift also implies a reconfiguration of roles. When a designer writes their own CSS stylesheets to compose a publication, they are no longer merely a software user, they are a tool builder. A generation of designers has thus adopted code as a graphic medium in its own right, questioning tools, repurposing them, building new ones.⁶ Open source tools offer a plasticity that proprietary tools make impossible : intervening in the source code means understanding how a tool works and therefore being able to adapt it, correct it, share it.

A culture of the commons : opening knowledge, sharing tools

These practices are part of a broader culture of open making. The maker movement, born from the convergence of open source digital culture and more traditional know-how, encourages the sharing of ideas and collaboration between expert and amateur practitioners alike. Becoming aware of technical objects not as a consumer or user but as a maker means both revaluing the technical object and reducing an alienation that is not caused by the machine, but by ignorance of its nature.⁷ Design outside the secrecy of industry, accessible, thought in common, these are the values found in web-to-print practices.

The notion of mutual aid reinforces this framework. Against the social Darwinism that makes competition the sole driver of development, cooperation and mutual support prove to be at least equally powerful adaptive strategies⁸. The existence of a shared objective is one of the factors that influences mutual aid within a group : one might imagine the actors of graphic workflows through the lens of a shared goal, calling for cooperation across disciplines. The communities around Paged.js, OSP (Open Source Publishing), and editorial hackathons operate precisely according to these logics, code sharing, collective documentation, contributions to open libraries. A distributed knowledge, in perpetual construction, owned by no one.

The notion of permacomputing adds an ecological dimension to this picture. By questioning the continuous growth of computational demands and dependence on massive, energy-intensive infrastructures, it « embraces limits and constraints as something positive in computational culture »⁹ : lightweight systems and open, durable formats are favourable conditions for creativity, not limitations. HTML and CSS are textual, lightweight formats, readable by a human without a dedicated application. A Markdown file transformed into a layout by a CSS stylesheet can be opened, modified and republished in twenty years without loss, something no InDesign file can guarantee. The constraint forces the separation of structure and form : it compels one to think of the publication as a semantic object before a visual one.

Making a publication, editing a factory

There is in web-to-print practices a double movement that the notion of publishing factory (fabrique d’édition) conceptualises precisely. This concept, developed by Antoine Fauchié in his thesis supervised by Marcello Vitali-Rosati, is rooted in the theoretical framework of editorialization, « the set of dynamics that produce and structure digital space »¹⁰ and extends its stakes towards the concrete question of making. For Antoine Fauchié, « the publishing factory is a phenomenon : it is the interweaving and reflexivity of work on the text and the constitution of a technical process enabling that work. The choice, arrangement and even development of editorial tools are carried out at the same time as the text is selected, structured, corrected, formatted and published »¹¹. The layout is code, the code is layout. This phenomenon goes beyond the notion of a publishing workflow, too linear, or a system, too closed. It is an articulation in which the production process is constituted at the same time as the editorial act is realised. Publishing factories are constituted in situations where technique is questioned and critiqued, it is a desire to do things differently, a will to challenge established or imposed archetypes, opening up perspectives towards new epistemological models.¹²

The question of tools is here central and deeply political. Julie Blanc precisely documents the problems posed by Adobe InDesign : a subscription model that raises the question of ownership of designers’ working tools, access cuts such as in Venezuela in 2019, the Pantone library made payable in 2022 following a commercial dispute, and more recently the default activation of an option to analyse creations in order to train Adobe’s artificial intelligence models.¹³ As graphic designer Nicolas Taffin, cited by Blanc, puts it : « it is the moment when the tool is no longer the property of the craftsman that he becomes a worker »¹⁴. This progressive proletarianisation of the graphic designer is precisely what web-to-print practices seek to contest.

Formats play a fundamental role in this dynamic. Antoine Fauchié demonstrates that they are not neutral : they model meaning, they condition thought.¹⁵ Choosing open and interoperable formats, Markdown, HTML, CSS, is not a purely technical choice, it is a decision that commits to a certain vision of what editorial production should be. Julie Blanc further specifies that CSS was designed from the outset to adapt to a wide range of output devices, both screen and print, but that its possibilities for print were long overlooked by graphic designers.¹⁶ It is this long-underestimated potential that web-to-print practices are reactivating today.

This coherence between the object of study and the form of the research is exemplary in both theses. Julie Blanc’s thesis was composed with Paged.js and is available online as a website. Antoine Fauchié’s thesis itself constitutes a publishing factory : composed with the same tools it describes, available as a website and PDF, with its sources publicly accessible. This performative dimension is not incidental, it means the research embodies the values it defends.

Paged.js, a JavaScript library distributed under the MIT licence allowing anyone to freely use, modify and redistribute it by crediting its authors, is at the heart of this dynamic. Its open licence is a political condition of sharing : it allows anyone to fork the code, develop extensions, and integrate it into their own workflow without requesting authorisation. This is what makes possible what Julie Blanc conceptualises as « collective instrumental geneses » (genèses instrumentales collectives) : these processes designate « the elaboration of instruments by and for the community of practice, and thus new resources for the activity of subjects »¹⁷. Building a tool in common is building a community. That said, Julie Blanc and Nolwenn Maudet note that while many designers consume open source, far fewer contribute to it¹⁸, active participation remains the work of an engaged minority, even if it is precisely through this process that knowledge is transmitted and community is formed.

These practices moreover emerge from a context worth naming : they are predominantly the work of Western designers with continuous access to electricity and the internet, fluent in the English of technical documentation. Antoine Fauchié acknowledges this explicitly¹⁹, just as Julie Blanc highlights the paradoxes of graphic design education caught between the devaluation of technical knowledge and the multiplication of postures.²⁰ These limits do not disqualify the practices, but they invite vigilance regarding the conditions of their emergence and who they include or exclude.

C&F Éditions is a concrete example : the publishing house has developed a complete workflow based on Paged.js and AsciiDoc for books printed in CMYK offset. The goal is not to find a cheaper alternative to InDesign, it is to have a shared working environment for text and layout that facilitates the flow of information and the intersection of interventions between different trades. It is horizontality that is sought.

A space to name

The documented projects are part of structured production processes and professional contexts : print runs of several hundred to several thousand copies, industrial printing. The explicit exclusion of fanzine or micro-publishing practices is a deliberate choice : to demonstrate that these methods work at scale.

The listed projects cover a broad spectrum of production workflows. Publications composed with Paged.js and Markdown, publications from wiki-to-print workflows, a collective pad-to-print translation, a magazine composed with html2print and OSPKit, screen-printed posters generated from CSV data via Python and SVGNest.

The nomenclature proposed by this site, CSV-to-print, Wiki-to-print, Pad-to-print, CMS-to-print, is not merely descriptive : it contributes to structuring a shared vocabulary for these practices. These projects moreover share four characteristics that might be borrowed from Antoine Fauchié : modularity of technical processes, openness of source code, originality of editorial approaches, and hybridisation between print tradition and digital technologies.²¹

This website therefore adopts an analytical perspective aimed at the technical examination of these editorial workflows and their associated ecosystems. By documenting tools, methods and infrastructural configurations, the site seeks to constitute a shared reference resource for research and practice in computational and code-based editorial production.

This network of varied elements, humans, software, formats, printing machines, with aligned interests within a production process refers to the actor-network theory of Bruno Latour, Michel Callon and Madeleine Akrich²² : no element is hierarchically superior to others, and it is the analysis of the process that matters, more than the result itself. The production workflows documented on this site are precisely that : assemblages in which adjustment and negotiation between actants are at the heart of the process.

To present oneself as a « non-exhaustive state of the art » is not rhetorical modesty. It is an assumed epistemic position. Web-to-print is a constantly evolving field, geographically distributed, rooted in practices that are often undervisibilised by academic or institutional circuits. To say that the state of the art is non-exhaustive is to acknowledge that there is always more to see, an invitation to contribute rather than a closure. A commons that everyone can enrich, owned by no one exclusively, and whose value grows with the number of those who participate in it.