requirements for a free settlement library network

> What is really needed is something which links all
> these diverse sites into something cohesive that weeds
> out junk and organizes what is left over.

I agree with your point. There is a lot out there and it would be nice
to reference it, to know what it good, and even to edit it somehow.
Although for copyright licensing reasons, other than for linking as you
mentioned, probably much of such information is completely unusable
directly by cut and paste into derived efforts such a collection of
manufacturing recipes. Reuse and improvement would require a big hassle
of figuring out who all the authors and rights holders are, getting
permission from the original authors, etc. which sometimes is more
uncertain and will take longer than starting from scratch.

Here are some elaborations on requirements for a space settlement
library network -- in part related to issues others have raised in other
related posts in this thread.

== short term, mailing lists and web sites are useful ===

In the short term, Wikis, mailing lists, static webs sites, custom web
hosting (like Innertransit or Zope or Plone http://www.plone.org/ ),
newsgroups, and existing peer-to-peer apps (like Freenet, Groove, or
Sametime) are what we have to work with off-the-shelf. (And thanks for
all the great links people posted on related efforts and sources of
technical information on the web.) There is a lot to be said for using
web pages or mailing lists as part of a bootstrapping process. Right
now, we are using a mailing list to discuss this, which has the benefit
of distributing this information to everyone on the list. While the list
is a single point of failure, at least the information is broadly
distributed to form the basis of a recovery with a new list or other
approach. Something like a web site running Plone and mailman, for
example, could be a useful home base for starting a related effort.

== a bigger vision ===

Still, here is a big long term vision: a network of space habitats each
being a node of a huge distributed library of all human knowledge about
how and why to affirm life in space, including the design, simulation,
and monitoring tools to interact with advanced manufacturing systems to
apply that knowledge to life needs.

The mini short term vision: a network of people on earth each having
their PC as a node in a huge distributed library of similar purpose. See
Theodore Sturgeon's 1950s sci-fi short story "The Skills of Xanadu" for
some inspiration here.

On top of these distributed knowledge networks there could be various
distributed moderation systems where some people (as Ryan suggests)
moderate up the stuff they find useful and down the stuff. There would
be some aspects of trust flowing through the network as one decides
whose moderations and contributions to pay more attention to. These are
all issues discussed quite a bit in various online and academic
communities (Slashdot wrestles with this).

Ultimately (if we survive) this is the sort of system we will have in
some form or another as space habitat ensconced multi-cellular life (and
machine intelligence) spreads from the Earth even if we do not
intentionally begin one now. Does such a system exist even now?
Certainly there are aspects of this in various existing systems, but
more than anything, I think the potential for improvement is needed
within the system chosen if this is to be a large scale, long term
effort, and that requires a freely licensed infrastructure.

== the Debian project is a useful model ==

If we were to come up with a more detailed set of requirements for such
an open ended improvable design system it would, in my opinion, have
many aspects of something like the Debian Project http://www.debian.org/
for GNU/Linux. That is, it would become a collection of freely licensed
packages all well labeled and indexed, perhaps in a variety of free
formats and free languages, managed mainly by volunteers (eventually as
or under a non-profit), hosted on a variety of servers, powered by free
tools themselves part of the distribution. Debian also included a
non-free aspect of things that may be redistributed but not changed, but
those are kept clearly separated as "non-free". This is Debian's "social
contract":
http://www.debian.org/social_contract.html
Eventually the space settlement system might be more fine grained that
that though -- involving things like a large database of manufacturing
recipes for example, where the recipes themselves were distributed (and
sadly perhaps not always 100% shared). This project could even
eventually be under Debian in some sense (the Debian SpaceSettlement
package?) although other arrangements might make sense as well
(including this project someday hosting Debian as the OS for Space
Settlements Everywhere :-). And of course, then there would be the
RedHats and SCOs trying in various ways to make a buck off of all this
free work (either by hook or by crook as it were.)

== a distributed library for practical and legal reasons ==

While the existing systems mentioned may all play some role (Wikipedia
and http://www.innertransit.org being open systems readily usable) I
think that long term a web based system should only be thought of as a
readily accessible front-end to part of a larger distributed system.
This is for a few legal and practical reasons.

First, the practical ones -- if the system is successful, bandwidth
costs will scale exponentially, and that means one person needs to pay
them if there is a central server. A distributed system can better
manage bandwidth by sharing the cost across the network -- keeping it
under upfront prepaid monthly limits. There are places like Ibiblio
http://www.ibiblio.org/ or SourceForge http://sourceforge.net/ or
Savannah http://savannah.gnu.org/ that might be able to supply such
bandwidth for free, but that still creates a single point of failure for
the project if that site goes down or is dissolved for technical, legal,
or financial reasons. Also, hosting everything on one site means the
system as a whole cannot as easily evolve by parts of it trying a new
things and the successful innovations spreading (sort of like evolving
populations of information tool users in a space settlement knowledge
ecology). Also, aspects of the system, such as doing large scale
simulations of space settlements may require distributed computing power
to make them affordable, so there may be a need for client software
willing to participate in background simulations. Also, it is quite
likely the network of design recipe information may need to be
manipulated in various fine grained ways, hinted at for example this
paper prototype at:
http://www.kurtz-fernhout.com/oscomak/prototype.htm
which realistically might requires client hosted software packages with
local 3D graphics and significant local processing power if a central
server is not to get bogged down in also supporting analysis of
manufacturing webs.

More importantly in some ways, there are specific legal reasons arguing
against a single site. Any discussion of space settlement in a detailed
way is going to include discussions and designs related to core
life-and-death things such as explosive rocket fuel, guidance systems,
medical drugs, and even social and political organizational methods.
Even a discussion of a design for something as mundane as a cutting
blade made of lunar ceramic might have political and legal implications
(it might bypass metal detectors for example). While I have no interest
in supporting misuse of such information in illegal or immoral
activities, the fact remains that such information can be misused
(because even a rock paper weight can be misused to kill somebody!), and
that for an international effort on the design of space settlements such
as the internet makes possible, different countries may impose various
restrictions on what sorts of information may be accessible by their
citizens or legally discussed. So, a single web site may rapidly run
into these legal issues and be shut down, whereas in a distributed
system, users can set filters to not receive or store information which
was not legal in their country (if such information was appropriately
tagged or machine-searchable). For example, in the USA, according to:
http://www4.law.cornell.edu/uscode/18/842.html
it is still legal to discuss and explain explosives (i.e. rocket fuel)
to people when you have no expectation they will misuse that
information, but that may well not be the case in, say, Germany (or
elsewhere). It may well be that various discussions or designs may not
be made available to certain people without formal certification (some
form of security clearance?). There may be other restrictions as well,
for example, even in a Star Trek universe, the Enterprise's computers
probably aren't going to let a five year old fabricate a live hand
phaser from a Replicator (excepting emergencies etc.).

To support handling multiple incompatible international legal
requirements, a distributed system in my opinion is an absolute
necessity long term. Until such a system exists or is adapted and used
(say, like the distributed Usenet newsgroups) public discussions on
space settlement may have to stay somewhat innocuous, since even the
simplest discussion of how to make things typically involves handling
cutting tools, using potentially poisonous or explosive chemicals, and
so on, and the very warnings on how to do these things safely
unfortunately double as instructions on how to do bad things (i.e. the
common household warning not to mix different types of cleaners.) At a
recent SSI conference, one engineer told me that kids these days can't
realistically become rocket scientists anymore because chemistry classes
are so dumbed down and access to dangerous materials so restricted to
prevent injury that kids never learn how to handle explosives safely or
how they operate practically. That is the social framework a space
settlemnt design effort unfortunately needs to operate in, and our
technical communication tools need to reflect that social reality (and I
think distributed peer-to-peer knowledge tools can do that to some extent).

To summarize, whatever the approach in the short term, in the long term,
a distributed system may be the best way to manage the legal liability
of conflicting individual country legal codes, and also has the
practical benefits mentioned above -- distributing bandwidth cost,
making possible distributed simulations, enabling client manipulation of
fine grained information in graphical ways, and avoiding a single point
of failure or shutdown.

== moving forward with specific recommendations ==

For the short term it may make sense to have a mailing list, Wiki, web
site, Innertransit space, Plone site, etc., but I would suggest to only
view this as an expedient gateway to a distributed system for people who
don't want to download and install related software. It might make sense
to host related web things on a site like Ibiblio, Savannah, or
elsewhere (all depending on the licensing), or perhaps I could host it
for as long as the needs were small from one of my virtual servers
(although I'm nervous about this for the legal reasons mentioned).

For the long term, it makes sense to find, improve, or develop free
applications (specifically in Python (and occasionally C/C++) using
wxWindows and on Debian) which interact with such a knowledge base. (I
have the beginnings of some prototypes along those lines, mainly related
to the Pointrel Data Repository System I have been working on.) I would
suggest most new stuff plus the core be in Python because it is
maintainable and well supported (and used already in the space community
like at NASA) plus wxWindows because it is also potentially
cross-platform on Macintosh and Windows, and Debian for a start because
it is probably the best free platform out there in terms of social
organization. These are similar choices to shoe made by the Open Source
Applications Foundation (funded initially by Mitch Kapor)
http://www.osafoundation.org/
for an information management tool "Chandler" they are working on. And a
space settlement design tool system might also have some linkages with
their work long term. Conveniently, these are all systems I know and use
now :-) but that is specifically because of all the systems I have
looked at (and I have used lots of languages and libraries etc.) these
seem the most practical for this sort of application right now given
many factors, and have a reasonable growth trajectory, and are
maintainable. For example, part of the effort might include improving a
system like PythonCAD:
http://www.pythoncad.org/
to integrate it into a system for designing space settlement infrastructure.

All this is not to say other solutions might not be part of this effort
as long as they mainly used free standards and free software (for
example documents written in Open Office, or made using a free CAD
program written in Lisp, etc.) just that for the greatest success, this
project needs to support, to paraphrase Doug Engelbart
http://www.bootstrap.org/
"a coevolution of the user community, knowledge, and tools". That
coevolution probably isn't going to happen unless the overall system is
open ended and based on a common free core and free standards and an
understanding and commitment to free licensing.

== beyond books and papers ==

As has been suggested, I think it would be great to publish books,
peer-reviewed papers, etc. related to such a project (e.g. Mark Prado's
Permanent book), given that formal processes exist for disseminating
these sorts of works to large numbers of people. Still, practically
speaking, those are old ways of publishing -- the new way is more the
collaborative discussion (perhaps with moderation like at Slashdot
http://slashdot.org/ ) resulting in both designs of artifacts (even if
just a Wiki web page) as well as the history of the discussion and the
prioritization of various comments. It is that new way that I would like
to focus on myself. The major reason for this is that while I think
summaries of free designs for artifacts are important, I think the
creation and distribution and discussion of the actual designs related
to those artifacts is just as important, and supporting that process
goes beyond making books or papers (even if there can be books and
papers about artifacts and discussions and such are useful, if sometimes
rapidly outdated).

== the biggest problem is cultural, not technical ==

I think the deepest issue with this project is a cultural one: that it
runs directly counter to the (I think correct) point that Jay Huebner
made on 9/13/2003 on this list in "Where's the Industrial Research?",
which is:
> We do need free forums for the development of ideas, but once a
> good idea surfaces, do not expect it to be discussed further here.
As long as space settlement is seen as something that is going to happen
by mainly a proprietary effort, we are in the same position Unix vendors
were twenty years ago when they were fighting it out to lock everyone
into their own proprietary version of Unix, and the big winner in the
end was not any Unix company but instead Microsoft (and we have all
probably suffered endless Blue Screens of Death etc. because of it).
While I think Jay's point is correct as regards commercial issues, I am
glad to see there is some interest here in moving beyond that approach
and out of a proprietary space settlement world (if proprietary space
settlements could even come about given short term investment horizons
of most organizations focused more on greed and near-term risk avoidance
than other issues).

== most space design work is not useful directly for legal reasons ==

And it will be a very long climb out of the pit of proprietary space
settlement information we have all helped dig -- since even NASA makes
most of its stuff proprietary -- despite being a government agency, by
preferring to license things as revenue generating spinoffs. For
example, a NASA center has on one web site: "Licensing agreements
demonstrate a monetary value of a center's research in that outside
entities are willing to pay for the center's space-based research."
http://spd.nasa.gov/sourcebook/patents.html
And their sofwtare typically requires signing license agreemtns and
paying licensing fees:
http://www.openchannelfoundation.org/cosmic/
(I've read of NASA engineers deciding to quit over this keeping
everything proprietary issue...) NASA does not even have the rights to
much of the details of space technology (like CAD files) it pays
contractors to develop. For example, two SSI conferences ago a NASA
engineer told me NASA was having trouble simulating the docking of the
Space Shuttle with the International Space Shuttle because the
contractors were only obligated to supply blue prints according to the
original contracts and they refused to turn over the CAD files (claiming
they owned them). So realistically, despite decades of space investment
by the US government, it is sadly mostly unavailable and unuseable (for
licensing reasons) at the detailed design level. Probably most of the
printed stuff SSI has (including conference proceedings) can't be put
online for similar reasons.

So, I think much of this work may have to be redone from scratch,
although obviously informed by previously printed materials and seeing
what is possible. And it will be a difficult road, as much of the
knowledge of how to make things (in detail) is kept proprietary as trade
secrets in the USA. Patents are often useless as references on how to
make things because they are written by patent lawyers to be overly
vague and also often intentionally omit key details so the originating
company can retain a proprietary advantage (despite what is stated in
the law about the need to be complete). When you think about it, the
issue of cultural mindset may be the single biggest issue in creating
sustainable settlements for everyone (whether on Earth on in space)
because there are a lot of great tinkeres and engineers in the world who
could design space settlements if only encouraged enough by seeing it as
possible to do such a big task as part of a larger effort.

== the future of engineering as a profession as open source ==

Note, I am not saying here that engineers should never be paid for their
efforts as long as we have an economy based on money, rather I would
suggest that like lawyers or doctors or accountants, engineers should
ideally be able to draw from a large pool of freely available
non-proprietary information (perhaps requiring years of training to
begin to comprehend) and could then make their money on making design
choices from a large body of public knowledge and doing local
customizations to specific local needs. To a large extent much of
engineering today does involve such activities (e.g. an engineer can
study old bridges whose internals are not kept hidden from view unlike
many software applications), but there are probably enough proprietary
pieces of essential manufacturing information out there to make it hard
to build complete space settlement networks until we can move beyond
these problems by collaboration.

There is another issue here which is that many engineers may have signed
NDAs or related work agreements preventing them from participating in an
open design process, but hopefully there are enough knowledgeable people
(especially retired ones) to move things forward. The USA seems to have
outsourced much of its manufacturing, sadly, leaving fewer and fewer
people active in the US economy who know how to make things, making
retired engineers likely to be many of the key contributors to any such
approach.

== 1970s technology is all we need for a first version ==

Fortunately, as Gerry O'Neill showed in his initial MIT studies, 1970s
technology is probably enough to get most of the way to the first space
settlements (networked computer workstations for design excepted of
course, but they don't need to mass much as a percentage even if needed
as Earth supplied "vitamins" for the first iterations).

--Paul Fernhout
http://www.pointrel.org

[This document is hereby placed into the public domain (except as noted,
which is presumed fair use); but remember that moral and legal laws
other than copyright say you can not modify it in a defamatory way such
as to indicate I wrote the modified version or to claim that you wrote
it if substantially unaltered.]