Re:ToolBook Business Models

[Eric Hunting]

> Date: Tues, Feb 10 2009 4:19 pm
> From: Bryan Bishop
>
>
> On Tue, Feb 10, 2009 at 5:42 PM, Eric Hunting wrote:
>> Let's have a look at a possible business model for the ToolBook
>> concept. This is a rough outline to start the processing of hammering
>> out details or spotting obvious bugs.
>
> Eric, I appreciate you going into so much detail on the ideas that we
> all share here. As I was reading (esp. the ending lists), I began to
> get a feel for another theme, among your suggestions for different
> prospective book series and media series. One of the points that I
> have been trying to push for a while now is the need for an organized
> "backend" behind all of this. I don't mean business management, but
> rather technical infrastructure management. Howto, documentation,
> instructions and other forms of information like that should be within
> the domain of computer infrastructure management. Many new forms of
> media, more fit for our computer age, need to be reconsidered in light
> of distributed bug trackers, distributed revision control systems and
> so on, which makes for a decentralized yet effective management of
> engineering infrastructure. To make these tools for managing this
> system of tools more accessible, I've mentioned before on this mailing
> list about wizards and guide tools and package maintainers to help
> package hardware and information about what the hardware requires,
> this way mostly everyone can package what they want into a form that
> can be added into the commons, in some reusable way, in some
> not-so-"one-off"-way. Here in Austin, the work on SKDB (and to some
> extent, unptnt) is centered around these ideas- especially what you
> originally were calling "an evolving fabber media format" (or
> something with that sentiment).
>
> Interestingly enough, diybio.org has recently expressed interest in
> the usefulness of those types of wizards for amateur experimentation
> in the wetlab, for "protocol management"-- we can imagine protocols as
> a type of 'fabber media'-- so there's some exciting progress on that
> front:
> http://groups.google.com/group/diybio/msg/587f1061e5d30e8f
>
> And of course, the more historical examples of these wizards are all
> over the industries. In print media, we see them in the
> choose-your-own-adventure novels, or in the old Infocom interactive
> fiction games. We also see them to help individuals install programs
> on nearly every operating system. On Windows, this consists of a
> click-through form that, now that I think about it, I am unsure of the
> advantage of. And on linux platforms, there's all sorts of
> configure-make-make-install tools (collectivelly called "autotools")
> as well as systems like APT, YUM, etc., which automate the
> installation of software packages and their requisite dependencies,
> either for their making or their operation. On debian, I always like
> to bring up the example of "dselect", which is this old ncurses
> interface to scrolling through the package lists (I think it's been
> superceded by 'aptitude' IIRC).

I can see this as an important area of development for ToolBook,
particularly as the sophistication of digital machine tools increases
and they integrate into the network computing environment on a bit
higher level than a printer. It seems to me the power in this rests in
the potential modularization of technique such that if you could
describe -and thus encode- technique in a sufficiently standardized
way it could be possible for computer programs to generate a recipe
for an object largely through a dialogue describing the product -
essentially the same process as software wizards work. The end result
could be something akin to a Maker Expert System where knowledge is
constantly being cultivated through use. The catch right now, though,
is the variability of information 'style.' In other words, when people
communicate knowledge, using a video or a book as an example, they
employ stylistic variations in language, graphic representation,
layout, etc. Somewhat arbitrary or subjective choices are made in how
to compartmentalize the information presented. For example, the number
of steps to describe a technique or process can vary greatly from one
presenter to another. To make a standardized way to encode this
knowledge such that a computer program could assimilate and distribute
it through something akin to an expert system it would require a
protocol that eliminates the variability of style and the
arbitrariness of step-divisions of tasks in much the way text based
expert systems demand a formalization of language structure. This
could be possible for higher levels of information. I think things
like the Lego instruction manuals offer a hint here. There may be
hundreds of thousands of models in the collective library of Lego
kits, and yet the instructions for every single one of them has an
identical format and graphic style. I suspect this protocol for
designing these instructions is so formalized that a computer program
could potentially generate them automatically from a deconstruction of
any particular model. We've discussed this general notion before in
the idea of 'process Taylorization' based on a software protocol that
can, through simulation, combine the topological and step-task
characteristics of machine processing to produce a 'Taylor script'
that is machine and network transportable and then 'compilable' to the
level of machine control language based on specific brand-name sets of
tools in the automation fab shop. Here we're sort of looking at the
same thing in the context of a means to machine encode the human
knowledge of technique so as to allow it to produce Taylor scripts
for both human use and, eventually, machine use. Seems like a very
promising area if we can devise protocols that can pull it off.
Perhaps there's something to be learned toward this in the area of
advanced educational courseware where they are also looking to find
means of standardizing the representation of knowledge for software-
based automated organization.

>> What is it?: ToolBook is a combination media production/publishing
>> and
>> Maker cooperative. It combines the facilities of Fab Labs and
>> industrial design workshops with facilities for the creation of
>> print,
>> video, and digital media along with small-run economy media
>> production
>> and/or digital distribution. These facilities would be concentrated
>> in
>> some key locations as well as distributed across the globe relying on
>> Internet integration and would be used by a community of Maker and
>> media talent that are co-owners of the venture. Creation of key
>> concentrator facilities could form the basis of physical community
>> development. (ie. Vajra project)
>
> Overall that sounds fine. I wonder though if the focus on media is a
> bit too extreme? The alternative that I can see is something akin to
> working to increase the amount of industrial capacity and
> manufacturing tools and machines that one can compress into the fablab
> itself. This is a similar task and similar mission, but overall
> directed back into internal self-improvement rather than feeding
> media-hungry minds (which, of course, is important, and provides some
> economic motivation, yes).

Yes, it does look like the focus on media is a bit heavy but that
because I was describing a business model that assumes media will be
the primary revenue stream. I agree completely that the key objective
would be to cultivate open industrial capability, self-improving the
fab lab and its applications. This would be particularly important to
the Vajra community whose objective is to demonstrate as much
industrial independence in a community context as the available
technology and ingenuity will allow. But research itself doesn't make
money and the tools at hand can't yet provide you with a means of
subsistence without big compromises in standard of living. So I had to
consider what was going to be the simplest near-term mechanism for
sustaining a cultural movement and the most obvious answer to come to
my mind was media and the free-lance writing and co-operative
publishing model. Ultimately, ToolBook is intended as a knowledge and
technology engine. But you can only monetize knowledge in a few ways.

In the open source software world, the primary way has been through
specialized service industry. The consulting model. Basically, the
majority of participants in open source software -if they're not
already salary employed- earn their living from it by helping novices
and companies figure out how to use it -and there is a sort of
economic disincentive in that to making the technology accessible to a
mainstream society if true ubiquitous literacy means fewer people to
pay you for your help. This, I suspect, may be partly why Linux has
cultivated a reputation as a computing platform for a technorati and
not something for the everyman to use. Now, I imagine this model will
most certainly emerge in open manufacture. You will have developers of
open manufacturing technology who basically make a living as
industrial consultants designing and setting up systems for companies
who see in open fab tech a cheaper and more flexible means to meeting
their industrial capability needs with a lot fewer strings attached.
But right now open fab tech isn't at a level where it can actually
supplant, wholesale, conventional industrial automation. So that's a
little off in the future.

The other way to monetize knowledge is to turn it into a product,
either in the form of a manufactured product that encodes the
knowledge or the form of media communicating the knowledge. Both these
are already being done in the Maker community right now. You have some
people using the knowledge they get out of the Maker community to
start small businesses making simple products to sell -exploiting the
novelty factor associated with the alternative approaches to design
the new tools compel you to employ. (flat pak furniture is a prime
example of that) And you also have people making money by
communicating fab knowledge in the form of media, which amplifies the
potential of the first model. Since this is an already proven way to
earn an income from this knowledge, I figured I would combine the two.
The former is best suited to cultivation in a community setting where
the capital investment in tools can be shared. I call this the
'atelier' model because it's akin to fine art and craft studios,
communes, or ateliers. The latter is more suited to widely dispersed
communities of freelancers making more intermittent use of collective
facilities and cooperating mostly over the Internet. This I call the
'publishing' model because it's akin to how book publishing works.
Since there may be no 'angel' financing at hand for something like
ToolBook, I chose to lead with the publishing model and follow with
the atelier model, the former building your community from a globally
dispersed population, the latter giving it reason to come together in
specific geographical locations and really get down to business. This
is why there's a focus on media publishing up-front -easy for anyone
anywhere to participate in- but then I start to note things like
building communities and shared job-shops that are about bringing in
that atelier approach. That atelier approach has potential to spin-off
pretty big business. I fully anticipate it could produce things like
the next PayGen/Freeplay Energy company or the next Mondragon.

>> Mission: To match Maker talent to media development talent and
>> share a
>> collective and evolving pool of media and industrial resources for
>> the
>> specific purpose of systematically developing Maker-oriented media
>> that focuses on a functional knowledge-base of tools, technology, and
>> technique rather than just discrete personal artifacts and novelties.
>
> It is important to emphasize the meaning of what a "functional
> knowledge-base" would be. SKDB, or the societal engineering knowledge
> database project that originally set me on this path, was meant to
> functionally encode- in the sense of computer programs and executional
> software- to help with this. Rather than storing information in a
> giant wiki, or some other semantically ambiguous apparatus, there is a
> way to make it functional in the sense that one's computer tools
> function, and through these tools modify the 'fabber media' or open
> source hardware packages, both computationally and physically (if you
> have the tools physically available), and consequently keep
> improvements or bug reports in the stream / in the loop of things
> overall.

I agree. I'm just not sure we're quite there yet in terms of the
software and hardware capability at-hand, or that most people can
comprehend what this means. But, most certainly, this is a key thing
to pursue.

>> Things that empower people to make, not just give them something to
>> make. This is intended to advance the Maker movement beyond the level
>> of disorganized hobbyist activity to a vital open industrial movement
>> and provide a media bridge between the fractured dispersed hacker
>> knowledge of casual Maker activity and the non-applied abstract
>> academic-oriented knowledge of formal engineering. Typical media
>> pursued would consist of user guides for the common tools of the
>> personal industrial workshop and Fab Lab and available building
>> systems/methods such as various modular building systems like Grid
>> Beam, T-slot, modular space frames, and the like. Media design would
>> be based on the evolving visually-oriented aesthetic approaches and
>> media forms of contemporary Maker media as epitomized by the current
>> Maker blogs and magazines. Kit products are also likely. Development
>> efforts would also focus on a systematic duplication of all the
>> common
>> tools of the Fab Lab in open source versions along with key
>> infrastructure systems and standard-of-living artifacts supporting
>> community development and developing world outreach such as
>> independent power systems, farming equipment, relief and durable
>> housing, independent manufacturing systems, open source laboratory
>> equipment, major domestic appliances, educational technology and
>> prefab facilities, and open source vehicle designs. Room would still
>> be made for the more typical Maker novelties as fun is still
>> important
>> to life. But the emphasis of the organization would be on empowerment
>> over amusement.
>
> I agree with all of this. Although some of it feels like a "tack-on"
> in the tree of reasons and ideas, instead of flowing from a central
> organized point. Empowering people to make is important. Maybe a more
> comprehensive approach, however, would be empowering people to make
> and modify the overall process in which they are contributing and
> learning. For instance, one of the issues in traditional organizations
> is that you can engineer an amazing tool, but that tool simply isn't
> going to be able to modify the system in which it was created because
> of human motivational issues or something silly like that. Another
> example is the link that I've been spreading around regarding Kevin
> Kelly's civilization-in-a-box, or civilization as an organism, and
> this "fabber media" at its core in order to allow Freitas-style
> self-replication. So not only is it empowering people to make within
> this paradigm, but allowing them to go off and make their own process
> and tools and their own forms of empowerment- whether this is by kits
> to draw some income to fund the fablab, replicating the fablab or
> simply making all of the tools over again as an educational
> demonstration by a couple of monks over a decade, so be it.
>
> Re: [Open Manufacturing] Dave Gingery and some more bootstrapping
> stuff
> http://groups.google.com/group/openmanufacturing/msg/e4c375acce772250

I see you point here and I agree. However, I'm not sure we have at-
hand the means to it in as sophisticated a manner as you describe.
Most certainly, I think we're heading in that direction. But look at
where even the semantic web is right now. Still not yet even in what
you could call an infancy. Now, the approach I propose does have a
mechanism for a closed loop of feedback between the knowledge output
of the ToolBook authoring (for want of a better word) community and
input from the general Maker community through the existing Internet-
based Maker media. The ToolBook community wouldn't obsolesce that by
any means. Indeed, it would actively participate in it and everybody
would have potential for be an author for ToolBook if they can
organize themselves well enough and/or invest in and share its
facilities. I anticipate a project submission system open both to
ToolBook shareholders AND the general public. Maybe we can even
implement a CSOP. (consumer stock ownership program -the more you
consume the more your share of ownership in the company producing.
Imagine what this world would be like if oil companies had mandatory
CSOPs...) This isn't as direct a means of co-cultivation of knowledge
as you suggest, but it's something. I'm just not sure we have the
technology yet to go to that more advanced level. When I think about
what you suggest -and correct me if this is an erroneous vision- what
comes to mind is an Internet-wide expert system with a natural
language front-end and a sophistication to where complex multimedia
and machine control software can be auto-composed from a user
dialogue. Do we have any hope of being able to do that today, or is
that a couple decades down the road?

>> Aside from this, ToolBook would function very similarly to current
>> publishing cooperatives as common in the independent book and comic
>> book publishing industries. Members of the co-op work primarily as
>> freelancers specializing in areas of their talent and interests;
>> Makers, writers, graphic and photographic artists, programmers, etc.
>> Any of these individuals -and in certain circumstances non-members-
>> can propose projects which are intended to culminate in the
>> production
>> of one or more media products (book/eBook, video, web site, DIY kit,
>> and occasionally small-run production whole artifacts) and which the
>> whole shareholder community ultimately votes on the investment in
>> through the access to co-op facilities and the investment in
>> publishing. (later on, the company may open up project submission to
>
> I don't know what this voting stuff is about. I figure that's only a
> matter of whether or not people actually spend time on the projects.
> And if nobody does, it's not that hard to store project details on a
> hard drive somewhere, waiting for somebody else to pick it up again.
> "Voting" issues would probably be left up to however individual
> facilities end up operating, is my guess.

You're more-or-less correct here. Basically, a project represents a
capital investment in materials and wear on facilities as well as an
investment in production and marketing of some form of media. For most
things this is small stuff. Many projects may need no more then a
little time in workshops, on-demand printing, and listing on a web
site catalog. No one would think twice about it unless the subject
matter is controversial. (a DIY video on the craft of hand-made glass
sex toys may raise some eyebrows...) But some projects -let's say
something on the scale of a house or a vehicle or some pretty
elaborate form of media like video produced on a stage set- are going
to be much more expensive and would be impractical to allow everyone
to pursue at once. And some projects are a little dangerous -such as
making an open source rocket. More generalized research is also
another issue here because it's pay-off in media that can generate a
can be very long and indirect -dispersed among a variety of future
projects as opposed to yielding a specific product itself, ToolBook
would have specific budget constraints based on cash flow and volume
of shareholders. So, when you are using shared facilities and the
ToolBook company is paying for materials and such, shareholders will
want in on the decision-making process for what is an acceptable
project to invest in.

The bigger ToolBook gets the more casual it can be about project
investment. I think on a small project level, access to facilities
would be mediated by a club model as I suggested with the Austin Fab
Lab. And if we can automate on demand production with media in the
manner of LuLu.com, then that club model could work for that too. But
if everyone is trying to fab a yacht at once and print 10k hardcover
copies of books on spec, you would have a serious problem on your
hands. Now, in a community setting like Vajra facilities access can be
much more casual because, as a live-in community, it's vetting its
users. And so it can come up with a simple average budget overhead on
an yearly basis. Communities would tend to be treated as a single
project with a running budget. But then, if they suddenly want a dozen
custom atomic force microscopes for something or are building a ten
storey vertical farm system and want ToolBook to pay for it, there's
going to have to be discussion and a democratic decision. That's all
this voting is about. Everyone in ToolBook has an economic stake in
it. So they are naturally going to want some say in these things. This
would be a bit more participatory than a typical corporation where
people just buy stock and only care about the economic performance.

>> individual percentage shares according to anticipated work. The
>> remainder of the profits are rolled back into the ToolBook enterprise
>> to recoup project investment, produce stock dividends, and invest in
>> additional co-op resources/facilities according to the collective P2P
>> strategy of the whole co-op community. Occasionally, projects may
>
> I think a model of "self-improvement" would be useful here, but from
> my history of looking at some other academic areas and seeing no
> reasonable signs of models of self-improvement, I'm not going to
> recommend anyone immediately goes off looking for it. It would be most
> interesting if resources can be fed back into the belly of the beast
> for upcreation/bootstrapping purposes. A while ago I sent out a few
> emails about "philanthropical bootstrapping" as a way to add coal and
> roar up the fires in the belly of the beast, or even giving away free
> biofuel from production centers implemented via these fablabs. Could
> be dangerous.
>
>> culminate in not only media products but physical facilities and
>> resources generating persistent income. For instance, a project
>
> Persistent income would be nice, but at some point
> maintenance/operational costs are going to generally be met, and it
> would be best if there is somehow an understanding that "what's next"
> is the internal 'improvement' or enhancement of tool-based capacities,
> via the construction and nurturing of that 'fabber media' and hardware
> packaging as well as computer infrastructure etc.
> - Bryan
> http://heybryan.org/
> 1 512 203 0507

Exactly. I think a key aspect of this is going to be the ToolBook
community communications platforms and its ability to set a group
agenda for specific technology development goals. I know it looks like
what I'm describing here is just a co-op for freelance DIY book
authors, and to a large degree that's how it would initially function
until it can start getting its people gathered together in specific
physical communities and has a budget big enough for more speculative
research. But ultimately it needs to set specific goals reflected as
preferences for certain projects over others -let's say, the book on
building a reconfigurable open source design CNC machine as opposed to
one on puppets or novelty lamps. (much as I personally like puppets
and novelty lamps...) Again, this is where that voting process comes
in. Through this the whole community enforces its collective agenda as
to what's relevant to use its facilities for, though everyone has pet
projects and I think you do need to let people have their fun too or
they won't stick around. Like most such ventures, if it isn't fun,
people won't stay with it.

Eric Hunting
erich...@gmail.com