Intro
Thank you for the possibility of being here. My name is Goetz, and I'm going to talk about very fresh ethnographic fieldwork.
Actually, I just come out of the field, and I've been hectically preparing this presentation in the last days. And I know what these presentations coming out of the field are. I'm sure that I'm going to be much too much in love with my field.
I just love my field at the moment. No distance.
I'm also going to be much too much interested in all sorts of details. They are amazingly significant for me, but maybe not for you.
And I haven't worked it through in any kind of theoretical way. I haven't worked through the literature that got evoked.
So it's a fresh presentation coming right out of fieldwork. And it has all the traits of that.
So I will focus on one of the themes that came out of it. I will focus on bootstrapping. On how bootstrapping is done, how bootstrapping works in this field, which is the Communications Design Group in San Francisco.
I will ask myself how bootstrapping works, and how the work of bootstrapping is done. And I'm playing here on a double meaning of "work", obviously.
But before I talk about all this, I want to just give you an introduction into what the Communications Design Group is, so that you get the context. And also because I'm so excited about it.
Communications Design Group
When I went to the CDG (Communications Design Group is "CDG") two months ago, I know almost nothing about it. The one thing that I knew is that this was a group founded by Alan Kay.
Most of you will know him, but some of you might not know him. So I'm going to quickly introduce him because he is an important figure, even though he's hardly ever there.
Alan Kay is mostly known for his work in Xerox PARC in the 1970s. Together with other members in the Learning Research Group in Xerox PARC and other people in Xerox PARC, he contributed significantly to the development of the GUI, but also to the architectures of hard and software of personal computers. [more]
And maybe most importantly, he was one of the people who invented a new form of programming, object-oriented programming, as famously come up in the early iterations of Smalltalk.
He is also known as one of the rare breeds of engineers who are more vocal than others, and who are almost kind of like media theorists.
Merging such diverse influence (and I just slashed this in because I like the picture) of Ivan Sutherland, Seymour Papert, or Marshall McLuhan, Kay envisioned, since the 1960s, computing as a medium for children of all ages. Where simplicity would not contradict invention and complexity, and where computing is a way to "amplify human reach and bring new ways of thinking to a faltering civilization". [more]
As such, Alan has become now, after a long career in the valley, a radical critic of digital cultures, providing scorching criticism of all sorts of things from object-oriented programming -- like the term that he invented, but with something that he now really regrets -- all the way to the way research is done nowadays in industry, as well as in academia.
With the CDG, Kay aims to provide a new generation of researchers the same working condition that he worked under in the 1960s and 70s, first through military funding, and then through Xerox PARC.
The idea for the CDG is simple. If great technologists and designers are given the highest amount of freedom that is possible, for a long time, they might be able to fundamentally rethink computational media. [more]
The CDG has two labs, one in LA and one in San Francisco, as well as further members and affiliates based in Boston, Munich, Berlin, and New York.
It is part of the German Logistical and Business Software Corporation, SAP, who employs most of the CDG's members. Yet the CDG operates largely independent of SAP's business agendas.
And some of the people in the lab are also formally affiliated with the Viewpoints Institute, which is a nonprofit organization that Alan Kay co-runs and has co-funded. [more]
Just like under the conditions of the military ARPA-IPTO funding in the 60s and PARC in the 1970s, researchers work (at least in theory) in a time frame that counts in decades and not in quarters. [more]
So I did my field work in San Francisco. The San Francisco lab has three research groups. I did most of my field work in the research group led by Bret Victor, which has roughly 10 members.
So let me start by telling you a bit about Bret Victor. Bret Victor is an inventor of media, that's a tweet by him, with a background in design as well as software and hardware engineering who has formerly worked at Apple. [more]
Just like Kay, Bret is an advocate for research neither driven by corporate nor academic logic, but by personal "principles".
To give you an idea what he means by "principles", I want to show you a short video of a talk that Bret Victor gave three years ago, and which made him rather famous. It's kind of like half a million views or something.
The extract that I'm gonna show you starts 35 minutes into the talk. So imagine just being wowed by a range of quite amazing demos, and then comes this moment.
"Why I have this principle, why I do this. When I see a violation of this principle, I don't think of that as an opportunity. [more]
"When I see creators constrained by their tools, their ideas compromised, I don't say, oh good, an opportunity to make a product, an opportunity to start a business, or an opportunity to do research or contribute to a field.
"I'm not excited by finding a problem to solve. I'm not in this for the joy of making things. Ideas are very precious to me. And when I see ideas dying, it hurts. I see a tragedy. To me it feels like a moral wrong. It feels like an injustice.
"And if I think there's anything I can do about it, I feel it's my responsibility to do so. Not opportunity, but responsibility.
"Now this is just my thing. I'm not asking you to believe in this the way that I do. My point here is that these words that I'm using, injustice, responsibility, moral wrong, these aren't the words we normally hear in a technical field.
"We do hear these words in association with social causes. So things like censorship, gender discrimination, environmental destruction, we all recognize these things as moral wrongs. Most of us wouldn't witness a civil rights violation and think, oh good, an opportunity. I hope not.
"Instead, we've been very fortunate to have had people throughout history who recognize these social wrongs, and saw it as their responsibility to address them. And so there's this activist lifestyle where a person dedicates themselves to fighting for a cause that they believe in.
"And the purpose of this talk is to tell you that this activist lifestyle is not just for social activism. As a technologist, you can recognize the wrong in the world. You can have a vision for what a better world could be. And you can dedicate yourself to fighting for a principle. Social activists typically fight by organizing, but you can fight by inventing."
So, after field work, two months of field work, I can confirm to you that this is a honest, heartfelt, and quite a genuine account how this guy really ticks.
And just to make sure that there is no misunderstanding here, when he talks about activism, then this is not anarchist techno-activism as we might have encountered.
It's for sure also not any other social cause, and it's also not the Silicon Valley mantra that technology will make the world a better place.
He means the activism of Doug Engelbart and Alan Kay. He means an activism that comes out of the hurt that something inside media is configured wrongly, and you gotta devote your life to it.
It's a kind of meta-activism, because this is, of course, also a talk, a performance, so he also wants other people to have similar principles, not his own, but similar principles when they work in technology.
Bret Victor's own principle is to develop immediate connections for creators to that which they are creating. For this to happen, programs need to connect in more direct and more dynamic ways to that what they actually do, enabling immediate access from the results of computing to processes of computing, and just importantly, the other way round.
Only when this is given, the right things are done by computers, and the "right things" are done by humans.
So what are the right things? It's a long topic, but maybe one example will explain it.
Have a look at most current forms of programming languages. They demand that the programmer first thinks the code before he trials it, turning him into an imperfect and stressed computer, instead of allowing him or her to take the role of a creative bricoleur, which we humans are much better at than being a computer.
Under the influence of the realm of possibilities that the CDG provides, Bret has extended this principle of immediate connections into the vision of a dynamic spatial medium, a spatial dynamic medium.
When programming leaves the rectangle of the screen and plays out in space and incites dynamic physical objects on them and in them, a new generation of immediate connections will become possible, enabling new forms of creating, of interacting, and of communicating alike.
So Bret leads a team of 10 younger researchers, all of them with a range of interesting topics in themselves, that talk to Bret's idea of dynamic spatial media in some or the other ways, but are often not directly confined by it.
Aside from Bret's group, as it is called in the lab, there are two other groups.
One of them is led by Vi Hart. Vi is a mathematician and musician. And here is a work by her. [more]
She has worked in the past at Khan Academy and is even a bigger star than Bret Victor on Twitter, YouTube, and Vimeo. Her videos have millions of views. Vi Hart shares also similar, yet maybe a bit more artistic devotions to a principle-driven form of research.
It is for her demanded by a personal need to create. She and other people in her group share this view and think that people in the CDG have to create. If they don't, they die or go insane.
Vi leads this approach into the interstices of geometry and music, and more recently, in conjunction with the new possibilities that the CDG provided her with, into virtual reality in general, and in particular, into the aesthetics and technology of spherical 3D video.
The third research group inside the lab is led by Dan Ingalls. Dan is the lifelong collaborator of Alan Kay, who pioneered as a programmer Bit Block Transfer, and is credited with programming large parts of Smalltalk, Squeak, and other languages that develop ideas out of Smalltalk further. [more]
Most recently, a language called Lively, which has an idea of liveness which resembles Bret's idea of immediate connection, of the creator to that which he creates, or she creates, and combines this with a wiki-based model of full real-time cooperation in the very process of programming. So whatever I do, I can directly go very deep in whatever you do. [more]
All three groups share one big office space. The space is a large loft with several smaller breakout rooms full of early prototypes, blackboards, whiteboards, art, specialized and non-specialized computers, machines such as laser cutters and large poster printers, an excellent library, and even a small Bucky dome, which has recently been turned into a rectangle-free zone. [more]
The lab is placed right in the middle of startup land in San Francisco, which Soma is. The idea behind that was originally to become a "virus" that might insert a different logic into the contemporary tech entrepreneurial culture of the valley.
In reality, there is little contact with those who are around them, all the app developers (who are really behaving like the cliches, at least what I've met in the Silicon Valley TV series, Startup Cultural Representations).
But the differences to them seem to be too big to be addressed. Most of the visitors that come in the lab, and there are a lot, usually one a day, sometimes two, are therefore not from these startups.
They are either friends (and they're called "friends") who are often working as developers in tech companies such as Oculus, Google, or Dropbox, and/or members of the universities you'd expect, MIT, Berkeley, Stanford, and so on and so on.
Knowledge is shared freely and with all guests, as all research in the CDG (and that's important) is under Creative Commons license. [more]
At the same time, the CDG has no website as a lab. This is partly not necessary because most researchers have individual or often highly effective ways of engaging with the public, but a CDG website would also contradict the idea of spatial dynamic medium.
A third reason is that they want to prevent to become the next big thing, and therefore aim to draw as little as possible unwanted attention to their safe space.
Indeed, the name "Communications Design Group" was chosen because it was so generic, and thus so easy to be forgotten. That's the logo that you're seeing at the moment that tells you a lot.
So, that was an introduction into CDG, and now I'm going to talk a little bit more on how the topic of bootstrapping emerged.
Bootstrapping
When I arrived at the Communications Design Group, I had very little idea about this. My initial impulse was fueled by curiosity about what Alan Kay and the people he is working with are up to in the year 2015.
Out of this curiosity, I generated some vague research ideas. Ideally, so I thought, I could generate new questions to be addressed at past moments in computing history, such as PARC in the 70s, by looking at the contemporary work of the Communications Design Group and vice versa.
This could (so the initial idea) not only teach me more about both labs, but also about the respective different contexts, that is about two very different historical moments in the emergence of digital cultures.
A second impulse was just as vague. If the researchers at the CDG, so I speculated, rethink computing from scratch, this might be an ideal place to learn and to deepen my own understanding of computational media. Just go there and learn.
Aside from these two vague impulses, I only knew what I did not want to do: another lab study.
The latter were done at great extents in the 80s and 90s. It seemed to me that the observations and analysis of Satchman, Rheinberger, Knositina, Latour, Pickering, Helmreich, Turkle, and so on and so on did not need another researcher who would prove once again how practices, relations, power structures, objects, subjectivities, prevalent representations, all contexts preform the search for the epistemic object, and through that, the outcome of research itself. [more]
Obviously, that has changed. I will now deliver to you the beginning of a lab study.
It's a lab study that hopefully will entail some, in the future, some elements of the other two approaches too. And of course, I'm doing research now, and not in the 80s and 90s, and in the area of invention of computational media, and not in biology, mathematics, physics, or the life sciences.
At least in the CDG, a lot of what lab studies have surfaced (not all, but a lot) is seen as given. Bret Victor, whom you just saw in the video, has turned a Latour paper into a comic. [more]
(This was one of the first things I saw. I mean, damn, that's kind of intimidating to do research...) And put it up in the lab.
Sherry Turkle has a prominent place in the library of the lab, and Lucy Suchman is circulating on the email list. And these are more than just anecdotes. [1] [2]
Research with the aim of inventing media has used the processes in which the lab shapes the research consciously since a long time.
You see on the top, you see some images of Alan Kay, and in the middle you see a little nice thing that Bret Victor did with his girlfriend.
They went around San Francisco and they put the 1962 Engelbart paper and they put it on the walls and tried to distribute knowledge this way. [more]
Doug Engelbart said for his research group at the SRI in the 1960s famously the "interesting recursive assignment of developing tools and techniques to make it more effective at carrying out its assignments. Its tangible product is a developing augmentation system to provide increased capability for developing and studying augmentation system." [more]
Obviously Engelbart talks here about bootstrapping. In bootstrapping, the fact that practices, relations, networks, and objects influence the outcome of the lab is not something hidden, or overlooked, or to be uncovered, but at least in part, consciously enforced. It becomes the recursive process, indeed the very method of inventing media. [more]
Bootstrapping is thus media theory put into practice. The assumption that media determine our situation is made useful in a recursive process. If you would set up a race, we would be the hares and Engelbart the hedgehog. However, in research it is sometimes necessary to be the hare, and that's what I invite you to do with me now in the next part called "auto-scaffolding".
Auto-scaffolding
When I spoke with one of the members of the Lively team (that's this programming environment that takes other ideas of Smalltalk than usually are taken out of Smalltalk), he told me that of course they would love it if other people would use their new programming environments to program other pieces of software in the lab. But their colleagues don't, and he understands that.
The Lively programming environment is an experimental research system. If you want to build a prototype, you do not want to use other people's experimental research systems. You want to use most of the time the simplest and most banal tools possible, exactly because you want to concentrate on the prototype and not on the challenges of someone else's prototype.
Indeed, even Lively (that is the programming environment) which is too experimental to be used by other colleagues in the lab, itself uses JavaScript, a language they dislike on many levels, but gets their work done.
So the fullest bootstrap, when the virtual machine for a language is written in itself, is suddenly achieved anyway. And if it is, it happens after many years of developing and in retrospect, and when the original language has stabilized.
So what does bootstrapping mean then? At least in the CDG and on a day-to-day level, bootstrapping is not something that plays out seamlessly in an ever-augmenting recursion of tool building.
(And by the way, Engelbart also didn't mean that. It's kind of ambivalent about that, but if you read it around, then he surely didn't mean that. )
Could we try it with a less ambitious technical metaphor, assembling larger systems out of smaller, simpler parts? In the case of booting, this happens in a few seconds, and in the case of bootstrapping a new medium, it happens over decades.
This too does not describe it. One of the reasons for this is that, at least in the CDG, prototypes work decidedly against usefulness. That's the mantra, "against usefulness".
To be very clear, this being against usefulness is not the being against usefulness of artists, who kind of do their autonomy. It's also not the grumbling hate of engineers, who hate users, stupid users, who don't understand their products.
The paradoxical reason for the need to avoid usefulness is the experience that usefulness can stop the overall longer process of bootstrapping.
Yes, at the end of the process of bootstrapping, products need to be shipped. But before this happened, usefulness is feared, at least for many researchers in the CDG, as a trap.
Once you produce useful tools, they reify. Usefulness reacts to present users, and present users are not the future users that the lab is working towards.
Useful prototypes stop being pointers at something larger, more long-term than what the prototype achieves. Useful prototypes become a means to an end, demand too much attention, become solutions, and maybe even are in danger of being turned into opportunities. And you remember, that's not what this is about.
So bootstrapping is seldomly a simple, straightforward form of recursiveness, or at least not in the CDG.
But at the same time, there are myriad ways how recursiveness of bootstrapping unfolds. One form you might call "auto-scaffolding".
The researchers in the CDG build themselves scaffolds which help them to achieve, to come a little closer, to a yet unclear vision of what a future medium exactly is.
It seems to me as if auto-scaffolding is achieved by mixing three fundamental ingredients. Researchers build A, representations, B, tailor-made tools, and C, pointers.
First, representations. They are iterated visions of what the medium could and should become. They talk about the medium, explain it as good as possible, constantly search anew for description. [more]
The second ingredients are pointers.
This is a pointer that Robert Ochshorn made. [more]
Pointers are early prototypes that point in the way they work towards a possible property of the medium. They point towards this property not only for others, but also for the researcher itself. They might see something through the pointer that she has not been able to see before.
A third ingredient are tailor-made tools which help you in the process of early prototyping.
Auto-scaffolding uses all three techniques and often mixes and swaps them. Indeed, many things that are produced in the lab can take the role of pointers and representation.
For example, here, this representation, once you point a laser to it, all sorts of very, very strange things will happen, which are spatial. [more]
Other forms like, for example, here, Paula Te's simulation of spatial behavior, is at the same time a tool and a representation. [more]
It is consciously spatial and not, for example, situated in VR. Which would be also a possibility because you could model all this relatively easy nowadays in VR, but that's not what would represent the right thing.
Or even a tool and a pointer and a representation at the same time.
Auto-scaffolding can lead into objects of multiple levels of partial recursion. One example is the system Hypercard in the World. [more]
Hypercard in the World consists of a series of paired beamers and cameras (would be somewhere here, and there) placed in several places of the lab, and of the software and database that runs these pairs.
The beamer-camera pairs catch laser light on the projecting space, making the projecting space interactive. In some way, this can be imagined almost like a series of connected interactive whiteboard.
However, the system is much more than that. While an interactive whiteboard is a reified finished useful product, Hypercard in the World is not only fully malleable as a system itself, it also allows members of Bret's group to build all sorts of pointers towards spatial media in general, and pointers towards a medium where code lives not anymore inside the rectangle screen, but on and in the object in particular.
Even more so, the code and the data of Hypercard in the World is partially projected on the code wall itself. [more]
So what you're seeing here is a Hypercard form that presents its own code, which is Hypercard.
This is not only a pointer towards how code could and should be placed in the world, it is at the same time an internal recursion of Hypercard in the World, though an imperfect one, which maybe can better be described as a pointer towards a recursion.
But even this pointer towards a recursion allows the researchers at the CDG to explore further possibilities by pointing to them. They can explore how code, once it is out there on the wall, takes on properties of immediate connections that especially Bret is working towards.
When you move your laser over this wall, you can see the changes in code and data live in front of your eyes.
The lab is full of such experiments of recursion, once you give up the expectation that this recursion has to be straightforward, total, and seamless.
On the other hand, you can also probably see that new forms of recursions are constantly developed. Even bootstrapping is bootstrapped.
However, I want to make a cut here and interrupt the argument, the reason being is that such forms of bootstrapping become only visible to me once I started to look at how bootstrapping actually works, and also what the work of bootstrapping is, how it plays out in time, and how different people cooperate in doing so.
This seems important to me because work is not coming out of the semantic field of cybernetics that bootstrapping comes out of. It introduces further layers.
The layers of work are manifold. I personally like to indulge here in the possibilities of the English language that most of you English native speakers might take for granted.
Work enables me to blend out, for now, labor. The difference between work and labor does not exist in the German term "Arbeit". I used "work" both in the ways of functioning (something "works", like a process), and in the way of working (a person does "work").
Again, this is not possible with the German term "Arbeit", where "Arbeit" is only tied to human work. When I look at a person working, I mix the practices in the plural, i.e. routines, with the practice in the singular, i.e. the one of specific event that plays out in time when something is getting done by modeling through.
Again, we have two words here in German, Praxis and Praktiken, and in English, we have one. And I can merge that, which makes things easier.
But instead of now going into a frenzy of theorizing about all of this, I want to use the remaining five minutes, I think, to stay close to the field.
Deep work
Bootstrapping does not only need work. It needs deep work.
I learned about deep work in a discussion that one of the members of Bret's team had initiated. His idea was to watch together a video of a presentation of an indie game developer called Jonathan Blow, who gives in his talk a personal account of his own experiences in pursuing long-term projects. And these experiences he calls "deep work". [more]
Without going too deep into the notion of deep work, a quick summary would say that deep work for Blow is a form of work that changes your life, that lasts for decades, that makes you an absolute specialist, one of the best in the world, and that makes you very, very lonely. So much of this talk is about being lonely.
Deep work is indeed the consequence of the principled approach that Bret Victor has described in the video you've seen before. In this discussion, all the lab members really related strongly to the notion of deep work and the necessity of this to be protected. Only when you engage in deep work, bootstrapping will happen.
Only after I listened to this conversation, I started to understand something that I was wondering for a while before. When you enter this lab and you come in and you try to say hello, you are greeted with absolute silence, something deeply irritating for a German. And also when you leave in the evening, no one would say hello.
And also during lunch, very often there are moments of silences. And also when you smile, no one would smile back. It feels like a really, at the beginning, like a rough place.
And only after this conversation and people confirmed that, I understood that this ability to not say hello, to not greet, to not smile, to not talk, is a deep, deep trust and a common achievement of the lab. And it's there to not interrupt each other in deep work.
It's collaboration for them, and one of the deepest forms.
And yet, both deep work and the new forms of togetherness are so essential for bootstrapping media. Indeed, you might argue that bootstrapping a new digital medium is in itself a reaction to the pain of deep work by developing new models for new forms of togetherness.
Much of the work in the lab is an attempt to get the creator out of the prison of such deep work. and into an open space (that's the ideal) where the computational medium becomes a colleague and a supporter. Not only for thinking, but also for communicating those thoughts which are, with current media, not communicable.
Especially the work of Bret's group is driven by the idea that media has to become more effective at bringing thought outside of the body, so that we can use such external imagination to play with them, to communicate them, to interact with them, instead of being stuck with thought in an already overcrowded brain. (And this overcrowd is so often being described to me, this problem of having this overcrowded brain, especially when you program.)
As Bernard is here, media for the lab are indeed truly tertiary retentions. And the longer you dream all day long about these tertiary retentions yet to be invented, talk is indeed a sad, painful, ineffective, hurtful form of communication.
And I can still feel that, and I'm still disabled in talking, coming back with it. I can still hardly make conversations, and I also notice that giving my talk is difficult, because there is this great idea to go beyond speech in this lab.
The work of autoscaffolding that I've described in the last part is therefore situated in a tension of the needs of bootstrapping, the need of deep work, and the need of collaboration.
A second set of processes organize looser form of collaborations for the aim of bootstrapping. One particularly important and interesting form is "riffing".
Riffing occurs when one member of the lab builds an early prototype and another one reacts to it with another prototype, and so on, and so on. Sometimes, but not all the times, this is very, very funny.
Some of the members in the lab compare riffing with the "yes and" routine in improv, because you always add to it. You don't criticize, but you keep the flow going by adding another prototype.
However, such forms of riffing by prototype need a medium for time delay, as new prototypes which seemingly react spontaneously to other ones usually need at least a few, couple of hours to be built. Riffing is therefore often done via email, a very important form of conversation in the lab, even though most people are co-present.
Email has a further advantage that the riffing process is documented. Indeed, the documentation of process is so important that the more important projects in the lab (which are staged in the research gallery, so these are the more important projects)... [more]
... if you laser them, then you will always see on a separate screen, the whole email thread conversation and earlier riffs that have led to these riffed up projects.
So I'm gonna cut short, but I tell you that I would tell you otherwise that, by other practices like dropping stuff in space, by in multiple ways, by leaving stuff behind and handing it over, this whole space starts to feel more and more like a brain.
And very often I had, and other people who come into this lab have the same experience, that you are in a living brain when you're in this spatial dynamic medium, which is kind of like a very, very surprising experience.
And I'm not making this up. This is something that a lot of people say.
It's a giant brain! And you can enter it. And it's bootstrapping itself.