Interview with Alexandre Leray, Eduard Martínez Piracés, Eric Schrijver, Femke Snelting and Stéphanie Vilayphiou of Open Source Publishing

Open Source Publishing is a design collective – or “caravan” – based in Brussels, but it includes people from all over the world. It was founded in 2006 and now it can even count on a second generation. OSP focuses on – just to mention a few things – free software culture, type design as a collaborative practice, tools’ intimacy.

What was the impetus to start OSP?

We cannot say that directly because we are, let’s say, OSP’s second generation, but we know the story. Femke Snelting and Harrison were working as graphic designers with Constant in Brussels, which is the association where we are right now. Constant is involved, among other things, with open licenses, therefore Femke and Harrison were developing several design projects oriented in free software culture: software’ releases, distribution parties and so on. At that time they were still using proprietary software. Then, in 2006, Adobe bought Macromedia, building in this way a big monopole.

These circumstances brought Femke and Harrison to question the tools they were using: why working with proprietary software when the content is coming from an opposite way of thinking? This is how they started to look for alternatives and began to organize “print parties”: public events in which content, layout and printing are produced live. Imposition is probably the trickiest part in the process, that’s why we have compiled a “recipe” in which we explain every step. An example is the recipe called How to print a booklet in 19 easy steps. Now we can do that in even less steps, because we have discovered new tools. Sometimes we fail in printing, but we’re not afraid of failure because we’re not merely interested in the result, but in the process, the recipe itself.

Recipe is not a casual word indeed: during print parties, the design recipe comes along with the actual preparation of food. We do that because we like the parallel between cooking and open content. It’s related to a metaphor by Richard Stallman often used to make people understand what free software and free culture is about. If you think of a food recipe, it becomes obvious that you are allowed to share it and modify it. If recipes were protected by copyright you wouldn’t have the right to use an ingredient if another one is missing, or to add anything. You couldn’t even share the recipe with you daughter.

You are not allowed to modify the proprietary software you’re using, so you can’t build upon. That produces an odd relation between you and your tools, because as a creator you have to know your tools deeply and be very intimate with them. This is true both for the physical realm and for the digital one. Design and art are not yet very intimate with the digital sphere. Working with open software helps designers and artists to get closer to their tools and to not be scared of understanding how those tools work.

Generally software is something that is completely finished and smooth. That was one of the points of departure: OSP started to question what is often taken for granted and considered neutral, transparent. We see tools not only as means but also as entities that embody cultural patterns and currents of thought.

Why did you choose the word “publishing” for your name?

We’re not publishers in the traditional sense, but we try to share a knowledge, to make it public. The word can be seen as an attempt. That is maybe confusing, but we like that.

What’s about you as the second generation? What did it bring you here?

(Eric Schrijver) I’ve known OSP for one year and half. I met OSP at the Libre Graphics Meeting and I got enthusiastic of how they were investigating the culture of free software. I appreciated their curiosity, their desire to understand how things work, and how programmers’ methodologies work for them as designers and artists. The original focus of these ideas was computer programming and suddenly designers were becoming involved, taking part in the community and attending meetings that were previously attended only by computer programmers. I thought: «Ok, I’m not alone in the world».

What’s your background?

I’m not really a programmer, I’ve always programmed a bit, but I never wanted to be a nerd (everybody laughs). Here there is something interesting about prejudices: in my work as artist and designer I was keeping my distance from programming. At a certain point I started wondering why I was doing that. Getting involved in programming was a way to confront my own prejudices. Why couldn’t you be artist and programmer at the same time? And what’s the relation between these two things? I think I came upon OSP just after getting interested in programming culture. I was fascinated by this culture also because computer programmers were making quite beautiful things and it messed with my idea of what programmers should be able to do. It turned out they were such smart and creative people, of course not all of them. I wanted to be part of this, but with my own background, as an artist and designer. And OSP was doing exactly this.

Is there anybody in Osp who has an official programming background?

Well, not very much. Some of us know how to program but we are mostly self-taught.

Do you believe FLOSS culture will be implemented on a large scales, by big companies or institutions? And why do you think this is not already happening?

Actually it is already quite implemented: if you buy a Mac, many software you use is open source, also Google runs a lot of open source software.

What’s about the publishing field?

There are some specialized companies working with LaTeX and software like this, particularly in the scientific field, but it’s a niche. Many journals are typeset using this software. Currently we’re experimenting with ConTeXt, which is another “flavour” of LaTeX.

Which kind of advantages does this software provide for scientific publishing?

LaTeX is purely made for scientists, because it’s really good at typesetting mathematics. ConTeXt is younger and it’s different because you are more free to modify the visual design.

Going back to the previous questions, there are many different levels to consider in order to understand why free and open source software is still not widely adopted. One of those is related to education: often big design schools don’t pay for licenses, Adobe offers the software because it’s conscious that designers come from there, in this way it consolidates the norm. The majority of professional designers uses these tools for years and doesn’t even think of something else. Then it becomes really hard to learn new tools when you’re so acquainted.

Then again, the tools of the trade can change quite fast: QuarkXpress was used across all over the publishing industry and Adobe InDesign replaced it rapidly. We should consider this as an example of how things which are taken for granted are not immune to change. Similarly, the desktop publishing revolution happened unexpectedly. This could happen with open source software too.

Another level concerns the development’s stage of some open software. For instance, Scribus – probably the most relevant open project for desktop publishing – is still quite problematic. On one hand there are features that you don’t typically find on proprietary software like Python’s APIs, which allow you to plug the software with something else and extend it. But on the other, some basic features that graphic designers expect are not there yet or they’re difficult to achieve.

Can you make some specific examples?

We were working on the catalogue for the Piet Zwart Networked Media Master final exhibition and we wanted to use spot colours in Scribus. We discovered that the feature exists but it’s still in an experimental stage. In the end we managed but it was quite exhausting. We needed to contact a guy from the Scribus developing team, who gave us a script written by someone else. It was very interesting for us to learn what a spot color literally is for the computer.

In the beginning we addressed the issue to the Scribus mailing list but no-one replied. So we thought it was simply because people who use and develop the software don’t need this feature, it’s not in their practice. They probably don’t design catalogues with spot colours, so they maybe don’t see the value of this possibility.

In fact, in the field of web design for instance, the people who write the tools are the ones who are using them. That is very common with open source because people write software partly for themselves and the best open source software is the one in which most programmers are interested. If you consider the field of desktop publishing, the amount of people who can program and at the same time are interested in design and printing is quite small. Therefore it’s hard to find people who wish to work on these kind of projects.

How will you share the solution you adopted for the catalogue?

We plan to write a post on OSP blog, but it’s always really hard to have the content ready to be shared. When you work on a project, there is a lot of stuff in it. Basically you know what belongs to what, but if you want to make the process usable to other people you have to document every single step. Ideally we would share everything but it takes too much time. Maybe this would become easier with new technologies… Generally when someone is interested in something specific, he or she asks by mail and we dig our archive to find it. It’s difficult to find a good balance between readiness and documentation. And our case is very different from regular publishing. There you get the end result and you should be happy with and there is a precise moment when you make your content public, while for us it’s more an ongoing thing.

You were talking about desktop publishing. It became a common practice for non-designers. Do you think many of them use open source software?

GIMP is used by many people. Probably more non-designers are enthusiastic to use this software than professionals. They probably use it because it’s free, but hopefully also because they want to join a community. In fact there is a huge Inkscape community. It’s also part of an environment: if you’re using Linux you don’t install Photoshop.

So do you think many non-professionals have an ideological attitude?

Maybe we are overly optimistic. But some are more political oriented than others. People are using Firefox and they’re aware they’re making a choice. It’s very surprising to see how it took a lot of ground.

Talking about publishing, do you know any big company using Scribus?

No. That’s because big companies are all about efficiency and productivity. From this point of view Scribus is not even an option. As far as we are concerned, we are not interested in being fast: for us it’s not about productivity, it’s more about design features.

Ok, let’s talk about that. On your website you say you «do not expect to find (or offer!) the same experience as the ones we are used to». What we’re used to? How do proprietary software, like Adobe’s ones, affect processes and results?

First, when you go to Adobe webstores, the vocabulary they use is purely marketing oriented: «you’re gonna be creative if you buy our software». Then Adobe tends to divide operation in compartments. You’re an illustrator? You’ll use Adobe Illustrator. You’re an amateur? You won’t use Photoshop, but Photoshop Elements. This way it’s like putting people in boxes. Why wouldn’t I be an amateur and use Photoshop? Eveyone is amateur and does use Photoshop!

Working with the Adobe Creative Suite determines the workflow: Illustrator, Photoshop and Indesign. You can’t go outside the circle. On the contrary, open software are pluggable and if you think you can plug everything you realize the possibilities are endless. For instance connecting GIMP to a website and genereting images on the fly… If you look at GNU/Linux philosophy, it says: «we have programs that do one thing and do it well». You combine all these minumum tasks so that you have complex structures, in this way you can make very powerful stuff.

Adobe software generally employ the metaphor of the canvas for the process of image creation. This makes sense on paper because sheets are actually canvases. But in the computer’s context it makes very little sense, because the computer is very good in repetitive tasks, patterns, algorithms. These kind of things are very natural for a computer, but they’re completely lacking from the metaphor of canvas. This metaphor basically hides much of the power of these new and superfast computers.

A good example could be the difference between Illustrator and Inkscape. In Inkscape you can view the source code of your document. You can switch from the visual representation to the textual representation, go back and forth and modify. We like the idea that the format is visible and readable at the same time but this is not something you find in Adobe software.

In which way your approach has a reflection on the final result and on the users?

There are several aspects, we are informed by this culture, so we end up affecting the content through design. It happened with the catalogue we were talking about: there are some patents’ pictures on it. Patents are about control on ideas but paradoxically the images of these are in the public domain. So we put all these patents’ pictures all over the catalogue. Our approach hopefully contributes to build a visually richer culture.

Another aspect is about collaboration: for example, you can input a text which comes directly from the internet in ConTeXt, that means that you can have several people all over the world working collaboratevely and remotely. It’s again about methodologies and recipes. A lot of these things are fueled by new ways of working together. If you have the idea your work is not going to be changed or reused, you’re not going to be interested in open licenses. The software and the way it’s used are strictly interrelated. Open work and design is a way of thinking, largely inspired by ideas about working collaboratively and remotely. Basically, these ideas came up in programming culture, and it’s nice to see how they’re not restricted to this field.

Do you think the fact you work on programming has a reflection on visual results?

Programming doesn’t necessarely mean that you end up with a specific kind of visual output. There can be many of them, and they don’t necessarely have to look like it was made on the computer. The few studios that are known for mixing programming and design end up with some kind of imagery that we can define “computational aesthetic”. We’re not against that, we are just not interested in building a consistent visual style. As long as you’re using your computer to design print or digital objects, you’re making digital design anyway, even if the software simulates the print output.

Are you saying that everyone who is using computer is somehow forced to produce computational aesthetic?

Yes, it’s a good way to put it. This is a relevant point, in fact several works of ours don’t generally look like what you expect from programming.

Let’s talk about readers’ and users’ involvement. In your blog posts and comments are at the same level. It seems like a dialogue more than a vertical communication…

(second generation OSP) The blog was made before we were here, but I wouldn’t see that as a coincidence. The blog is meant to show the processes, not the final products. It’s a way to communicate differently to people.

How do you get responses from them?

Many conversations took place through the blog’s commenting system. This is one means, but there are others. We are subscribed to the mailing lists of the software we use, and we talk a lot with software developers. Print parties are also a way to interact with people and start some new discussions.

Can you tell a bit more about print parties?

We can talk about the last one. Our starting point was the wikipedia page of the Atomium monument in Brussels. What was intriguing about that was the fact that people can’t put pictures of the monument on the page, because it’s an architectural work and it’s protected by copyright. So at that time there was a scanning of the Atomium taken from the two euro coin. We wanted to layout the page and the discussion as well, because it was about pictures’ rights and about the company who manages these rights. We wanted to do this in ConTeXt, because we discovered this software two days earlier. Eventually the printer failed.

Generally print parties don’t have a strict timing, but we were in an opening so we had one hour to do everything. Sometimes we don’t have the recipe ready but we have an idea of what to do, so we develop the recipe live. People ask many questions and try to understand.

Who were the people in that case?

We were in a opening of a common atalier, so there were friends, artists and people in the field of publishing and graphic design. But often OSP speaks to a public of developers in the free software community, like at the Libre Graphic Meeting. It’s a funny development, because we have a relation with a different group of people, while in general artist and designers are interested in attracting people from their own field. We try to be the missing link.

Let’s say you are contacted by a large publishing house and it asks you to implement in their entire workflow free and open source software? How would you do that?

That would be a very different job compared to our designs—more like consultants focussing on efficiency…

But there are some big publishing houses with an ideological point of view, which maybe don’t know how to improve their workflow. In this case how would you take this challenge?

As we said before, LaTeX is a good example of open software used in large scale. And it’s quite surprising that some publishers don’t already use it. It seems they have forgotten about their own free software’s history. LaTeX has the basis for their publishing industry, because it makes the workflow faster.

For us free sotware becomes interesting when there’s some kind of friction between content and tool used, between licenses and the content licensed. That seems not to be the case, it’s not our area of expertise.

It’s becoming more and more difficult for designers to define their area of expertise. How would you define yours?

We don’t claim to be experts in something. We try to explore different territories and take the risks of using something that we don’t know. That’s inherent to the way OSP works. Our people are dispersed, they speak different languages, have different ages and so on. There’s a lot of blur among areas where people feel competent. If you create the right balance between waht you feel competent in and what you feel out of, you can develop very interesting works. Often it happens that you don’t know anything about, but it’s interesting to nibble the edges of the area in which you feel comfortable. For us it’s important to have people with different level of comfort in each project. You can’t do that alone, you need a small community.

Do you think there’s a relation between this approach and what me may call the neutralization of professionality, in which anyone can do anything?

We are not part of this sort of uber-amateurism. There’s an extreme intensity and seriousness in how we work, but this doesn’t mean that it comes from expertise. It’s something else than making a difference between professional and amateur. In the discourse around amateurism versus professionality there is an easy split made between rough and precise, for free and for money, trained and untrained. These divisions are not very constructive.

We are constantly re-specializing. We know about things that other might not know. We actually created a quite focused area. But it might not fit with the metrics of what a designer should be as projected by companies, schools or even employment offices. Maybe we’re taking part in redefining that. And hopefully it will be part of a redefinition of what a developer or a mathematician is.

There are a lot of disciplines we come in contact with, and these come in contact with us. People often ask: «Are you scientist or humanist? Programmer or designer?» So we started to say we’re not part of the “or”, but more likely into the “and”. We try to link stuff that are usually separated. We don’t like to exclude one or another.

Of course, code and programming is a kind of vogue nowadays. The most fashionable studios do code fireworks and use new media as a mean of expression. The new designer is the one that can code. In fact it’s becoming really hard to get your way in design if you’re not involved in new media. Maybe because more money is invested in web design. There’s this economy issue and there’s hipness. Part of the jobs we get is because people are starting to recognize that coding has a value, five years ago it was not the case.

Right now people appreciate the ability to juggle technology in a clever way, which is part of the glamour of OSP. But our work is never about virtuosity. It might look like it from the outside because if you don’t know how to program any code looks complex. Our work is about flows in technology and engagement with software culture. We run low-tech sotwares on six years old computers. If you realize that, then you have a completely different image of design and technology. Furthermore, we try not to be too fascinated by the code and by what it produces. Instead we try to understand how things works, how to reproduce a feature and how to refine it. Anyway this misunderstanding probably helps us.

How do you position yourselves in the debate around digital innovations and printed matter?

The paradigm of print is being redefined, the way things are published is changing and the same goes for the editing part. The digital realm is affecting printed matter. The two should be seen as complementary. From our point of view, the connection among how the content is produced, how it comes to exist and the license under which it’s released, it’s crucial. If the material – whether an image, a text or a combination of the two – is released as a source, it’s possible to output it in different formats with different devices, not only at the same time but also in time. Books that are out of fashion can be picked up later again and exist in different formats or in different configurations. The existence of different versions has free license as a natural consequence. In this way free license becomes part of the object’s materiality. New relations between content and form will be highlighted, because some choices are not default anymore. As a design era, it’s really exciting. You need to ask the right question.

EPUB – which is an open format – is becoming the standard for ebooks, but it allows very little room for designers to play with…

It was the same situation in the web ten years ago, even now you can mess up a website’s layout, just by zooming the text. EPUB is still young, it will evolve and people will design differently for it. For instance, line breaks in printed books are mostly set automatically, but the designer checks all of them anyway. Re-encoding the book to a digital object without taking care of line breaks is like throwing away something from the book itself. These kinds of issue blink very interesting questions about form and content and about the materiality of the book. We should try to embrace EPUB’s limitations because that’s the place where fun stuff often happen. At the same time we should investigate how we can manifest a sort of digital materiality and find what variables we can play with.

Maybe the traditional typographic knowledge will be entrusted to programs and algorithms, while the designer will take care of other elements that can be put on the page somehow, considering the flexibility of the screen, changing sizes and so on. It’ll allow different things, it’s not about placement’s control but about control of the relation among elements. It’s a choreography job.
It’s interesting to hear this discussion about HTML, CSS, standards: there’s a mix of serendipity and planning.

I suppose that even in the early stages of the web there was already a strong experimentation around it, not only techinically oriented but also artistically. Why do you think this is not happening for EPUB?

It’s probably because EPUB is basically offline. In addiction, the fact that you have to buy a device to make experiments tells you a lot. And even if you have an e-reader, you can’t view the source code and start working on it. In HTML, the editor and the viewer are generally on the same machine. How are you literally going to make experiments through two different devices? It’s much less exciting because you cannot immediately edit and view. Furthermore, many ebooks have closed source reading software.

In HTML you have the combination of a free browser, an open standard and a simple way to view the source. These three things together are powerful. You just need a text editor to do experiments. EPUB standard might be open, but the connection between coding, viewing and sharing is completely dispersed.

What’s about the hybrid solutions, as with QR codes and fiducial markers?

(Stéphanie Vilayphiou) I saw some QR codes on tv, which was kind of nonsense because you had to photograph the TV to go on a website.

(Eric Schrijver) I made a QR code for an exhibition, it was pink and glittering.

Did it work anyway?

Well, at least for some of the visitors it did.

Maybe the possibilities to experiment lie more in the way to spread a content, than in the design of the layout.

FLOSS manuals is a great example of how hybridizations can work effectively. The project is down to earth, it’s modest. It’s about how people can write together, and there is an actual relationship between the phisical and the digital and vice versa: you can translate the book and print it, you can correct it, you can customize it. This way to use digital space as a way to produce a shared output from your own desk is quite amazing. This is much more interesting than digitally augmented books, which are mostly a marketing strategy for publishers. It’s already there, we don’t have to look far to see where the phisical and the digital come together. We don’t really need to do fireworks, maybe some typographic enhancements (laugh).

In fact a relevant part of your work is about type design.

OSP is working on typefaces because they’re a great example of what free software is.Typefaces are complex entities, many things in one: text, drawing, software, math, aesthetics. Making type design is like crystallizing all these things in one. Typefaces are rarely under free license even if they’re freeware, this means you can’t modify a typeface even if you bought it.

In Hackerspace, John Haltiwanger and Pierre Marchand are working on an interesting project about books and typography. They are constructing a book scanner, a sort of DIY Google Books scanner. They say: «If Google can do it, we can do it». So they’re building a wooden construction that allows you to process phisical books. It has to be suitable for high volume scanning and it needs something to automatically turn the pages. Pierre and John are using OCR (Optical Charachter Recognition) not only to extract the text, but also to extract the font. In this way after the book is scanned you would be able to typeset it again with the original layout and typeface, or you could just use the typeface for another content and so on. In the perspective of the relationship between digital and printing this is a very interesting project. It makes the book reflowable but it preserves, at the same time, some of its physical characteristics.

OCR was developed by IBM in the eighties and made open source because of a lack of interest. Then Google got it and started to develop it actively. Now the technology is still open source, but Google does not release training data, the data through which the software is trained to recognise characters in a better way. Here you can see where the open source philosophy only half works, in fact there are many luminaries of FLOSS culture who work for Google. We will never have even a small percentage of the power of Google, but at least we will get an idea of what is going on there.