This research is organised around different axes. It will look into formal narratives generated by algorithms, question the form of the book as an object and authorial product, explore the concept of a ‘decolonial publishing house’ and experiment with the influence of ‘forest baths’ on the writing of code and the communication with trees.
Concretely, this translates into the following current questions.
## Algorithms as Authors
- consider our geo-political and body-political position when designing, building, researching and theorizing about computing
- embrace the 'decolonial' option: thinking through what it means to design and build the systems that we propose to produce with and for those situated at the periphery of the world system. This involves thinking with trees as entities that have been under cared for and are literally on fire due largely also to the effects of colonialism and subsequently global climate change. How can we respectfully engage with trees when creating a publishing house?
- unfold coding practises: embrace this 'decolonial option', understand their structures, the histories and contexts they are embedded in and the radical processes they execute. What if algorithms were committed to a more equal world and a healthier planet? What if algorithms tried to form a symbiosis with trees?
## About trees
- does it still make sense to print books on paper? And what is the ecological footstep of creating and reading a digital book? Can a ‘decolonial’ publishing house exist? Is it possible to organise an equitable book publishing activity in which trees have a say in content and form? Is there a way to publish books in a way that is respectful of trees and nature? What form would this take?
- how can we find materials that translate in a metaphorical way the reasoning of the algorithm-author. The research will focus as much on data retrieved from analog data such as temperature, light, moisture related to specific trees, as on today's data architectures, colonial herbaria, legislative texts, philosophy, literature. The idea is to question the representation of trees in texts and data.
- since programming code - the dialogue with algorithms - is often a very draining activity that can generate a lot of stress, we would like to lead experiments to see if and how conscious visits to the forest could influence the writing of code.
## Book as form
- when algorithms produce a book, they go beyond its established form. Index ‘pages’ can run over a thousand different pdfs, page numbers become a technical artefact for a digital book. Furthermore, computer scripts can generate an infinite amount of text, but given some graphical lay-out elements in the code, they can also generate an unlimited amount of pdfs, of which each one can be different. They are capable of generating so much text that it becomes noise, too big to grasp, impossible to read in a lifetime and potentially useless.
- if we consider that the narratives of the algorithms are important to teach us about their functioning, then what they produce matters. What is printed also matters. We ask ourselves what kind of decision making methodologies we can invent to deal with their abundance and make their work legible for human beings.
## Contents of the book
- Does reading lose its value when you can choose one million copies of a slightly different version of a book?
- Do we show the infinity of the generated copies? If yes, how?
- Does each book need the capacity to exist in infinite variations? Can we decide upon a 'static' version of a book? Is this interesting?
- How can the organisation of books be presented in a fair inclusive way What 'fair' categories/characteristics can we think of for a first 'index' web page?
- Do we keep the plain text as a style throughout all editions of the publishing house, reflecting the materiality of code and logging? Or do we also create 'books' that look like more classic / elaborately laid-out books. A more structured layout seems to invite a different way of coding that is written with the layout in mind, for example by splitting logs into parts. But also, the code to produce the layout becomes part of the code, making the scripts less simple and more clearly designed. (Thought: make a special stdout which writes html, rather than plaintext? Could we then use something like ANSI codes to set the styling? )
- Now the pdf does not exist on the server. It only exists in the RAM memory of the reader’s computer, and it only exists on your computer if you decide to download it. This is a response to the field of literature, where the book is considered as something fixed. This immediate download takes a step into the distribution itself. We could include the sender in the pdf!
- The format of the pdf is good, because otherwise it would become something else: a media art/a webpage. It would be more about the interactivity and the experience. Now we speak back with an object that is produced by the publishing industry and tradition. We take on the dress and habits of that context: we start as a website and freeze it; what is generated is static; it is a quality that it is a pdf; a way to connect to the website and vice versa.
- How would you cite this book?
- Is each generated book a unique object? It can be easily copied and redistributed. Should we talk about unique objects? NFTs (non-fungible tokens, a way to integrate artworks into Blockchain, wasteful by design...) are unique objects.
- The book can be present in different places: on the website, as a pdf, as a book during the walk, shared in the neighbourhood. They can become topics for a workshop: as tools and to open up algorithms to a wider audience.
- Is it an idea to open up this platform as a service? This rises other questions: what about curatorship? A publishing house is assuring quality, it is more than a Print-on-demand service.
- What about formats of the books? Now they switch between A4 and US A4. Will they always be A4?
- Can we imagine linking the generation of the pdfs to a POD-platform like Lulu?
## Technical aspects of generating books
- Now the graphic design of La Distancia Levenshtein lee a Cortázar is using html and weasyprint, but weasyprint does not support javascript. All variations of the book have to be generated by the layer generating the scripts. This doesn’t have to be a problem, but it does offer a limitation. Weasyprint is implemented in Python, but it is also implementing it's own browser/render engine. The development of the library is limited.
- Gijs has a clear prreference to work with paged.js because it supports javascript. On the other side it needs a browser running on the server (100MB). Another downside of pagedjs is that it is running in node, that means an extra layer / technology to maintain.
- What is the api of the publishing platform? What is infrastructure of the publishing house? On what server? How to deal with safety? How to make sure it keeps running? How to avoid server overload? Harmful robots? Or multiple users at the same time? Make a waiting row? What is the scale of this project? For now, there is the proposal to run the server a year to see what happens.
- What about the ecological impact of the 'infinite' generation of books? How do we calculate that? Include or exclude it?
