well i just spent the last two hours scouring the net for public domain or
otherwise copyright unentangled materials property data. the data is to be
used in the "skdb" hardware repository for calculating various things like
"how much torque is needed to tighten this bolt?" or "how much does a
brick of 6061 aluminum weigh?"
a CC/GPL list of specific common materials (i.e. not "plastic" or "steel")
and their properties in an easy to parse format would be awesome.
looking for things like:
coefficient of friction wrt other materials
density
tensile strength
modulus
fatigue and creep info
electrical properties
chemical resistance
changes due to temperature (including tempering processes)
machinability
elemental makeup
similarity to other materials
price and availability
there are numerous copyrighted databases out there already but i don't
feel comfortable putting a lot of work into parsing and beautifying a
dataset that i may get in trouble for using later.
i already know we dont have any lawyers on this list, but does anyone know
the legality of programmatically extracting the data from, say, a webpage
or an ebook, and then redistributing it in a different form?
what about by hand? at what point is it no longer "fair use"?
-fenn
By contrast, for example, Appropedia
http://www.appropedia.org/Welcome_to_Appropedia
has a "Creative Commons Attribution-Share Alike 3.0" license, that does not
restrict commercial use.
So, I would expect someone could move Appropedia content into OpenMaterials,
but not OpenMaterials content into Appropedia? Also, Wikipedia is GFDL
(incompatible both ways), but likely moving to CC-BY-SA, and again, you
could then only move content from Wikipedia to OpenMaterials, but not the
other way.
Licensing issues are a pain. Personally, I wish everyone in the world had
stuck to using either the GPL or BDS-revised. :-) Lawrence Lessig has
contributed to vast amounts of license proliferation, and he should have
know better IMHO. Even CC licenses are under multiple versions right now,
and it is not clear to me how compatible content from various versions are,
since unlike the standard GPL, there is no built in upgrade path (unless you
opt out of "or any later version).
I'm not even sure what the implications are of the same CC license but
"ported" to another language and how compatible they are under "Share Alike
— If you alter, transform, or build upon this work, you may distribute the
resulting work only under the same, similar or a compatible license." It
obviously sounds like the intent is that they are all compatible as
"similar" or "compatible" licenses, but what does that really mean in
practice? At the very least, it seems to me that someone still has to track
the details of what exact version of the CC license was used originally if
you aggregate content, and that is another small but real hurdle.
To me, "non-commercial use" is also a very problematical as a concept. Isn't
a teacher displaying information in the course of their work commercial use?
It seems that way to me.
Also, the open source definition suggests not discriminating against fields
of endeavor, so it is questionable to me if a "no commercial use"
restriction is really "open source". Example:
http://www.opensource.org/docs/definition.php
"""
6. No Discrimination Against Fields of Endeavor: The license must not
restrict anyone from making use of the program in a specific field of
endeavor. For example, it may not restrict the program from being used in a
business, or from being used for genetic research. Rationale: The major
intention of this clause is to prohibit license traps that prevent open
source from being used commercially. We want commercial users to join our
community, not feel excluded from it.
""
Now, as a practical matter, no commercial use may make sense as a choice for
other reasons. So it might work well for you whatever your plans are.
Even "open" is also a problematical term. So, one can have "open something"
without it being "open source something" or "free something".
Personally, I like the idea of ethical restrictions on how things are used
in practice (like "this content should not be used to do bad thing X"). But
how that translates into the world of open source licenses is a different
issue, as in, maybe it is better to restrict distribution in informal ways
that to have it in the license? Or it is better to surround the content with
an ethical framework of ideas?
With that said, OpenMaterials looks like a wonderful site and a great
project. And a very good idea implicit in OpenMaterials is that by focusing
more on materials than devices, OpenMaterials sidesteps some of the concerns
I had for OSCOMAK in terms of people putting in weapons or other problematic
things. I wish with OSCOMAK I had been clever enough to say, "what part can
I do with the least risk or controversy?".
By the way, and while I am all for whatever people are doing, and applaud
anything going in that direction, I still am enamored of the promise of the
"social semantic desktop"
http://www.semanticdesktop.org/
(basically, having content locally on desktops instead of mainly on servers,
somehow tagged collaboratively with metadata, with content in standard
machine-processable formats). This is for various reasons (security,
redundancy, accountability, privacy, liability, processability, cheapness,
etc.). But social semantic desktop research like Nepomuk is really bleeding
edge stuff with a tiny audience and still under development, so certainly a
Wiki and website is a safer way to go right now.
Anyway, the OpenMaterials site looks fabulous. The look reminds me most of
Inhabitat for some reason (even though the emphasis is different):
http://www.inhabitat.com/
Both must have great graphics designer onboard. :-) And both have great
technical content, too. :-) You might look at Inhabitat to study their
business model, depending what your plans are for the OpenMaterials site.
Anyway, for me, I feel the deeper question is how free and open source
projects can interact and share content in such a way as to allow computer
based analysis of things like cradle-to-cradle manufacturing.
http://en.wikipedia.org/wiki/Cradle_to_cradle
And it is not an easy question. That's another big reason I like the Social
Semantic Desktop idea at this point, because it is still expensive for
individuals to run servers, but almost everyone has a desktop they could do
a lot of local number crunching on essentially for free. But with that said,
as with Appropedia, people can download the contents of a website and
process it locally, so using web sites to collect or view content does not
preclude doing local processing of it too.
Your email also raises, once again, a deeper issue of what the main mission
of this list is, which has ranged from discussing the nuts-and-bolts of
specific projects, to collecting notes about various ongoing open efforts,
to strategizing how different open groups can collaborate, to talking about
social/economic theory of an open world.
I feel the biggest value of a list like this might eventually be in talking
about how the different groups could work together somehow. But that is
maybe also in some ways the hardest, especially often for engineering types
who would rather do-it-themselves. :-) And to begin with, we are not all on
the same page about a lot of assumptions (especially about economics), so it
is even hard to come up with some well-defined common ground. Even this
issue of licensing (commercial use vs. non-commercial use) reflects a whole
set of assumptions about how to run a web site, expectations about the
future, and so on. So, that's a big social issue we have yet to resolve.
--Paul Fernhout
I was talking with fenn the other night and brought up a question that
I might as well post here. Kirsty and Katarina and I have been talking
off-list (and we'll now start talking off-list) but I think it would
be better if someone else (like Paul) responds to them first. In the
mean time, I have some legal questions about engineering, if we're
going to be talking about legal stuff again- I might as well bring
this up.
How does licensing of databases work in the engineering fields? This
isn't something that my university has gone over with me, even though
I am enrolled in mechanical engineering studies. If there is a number
in the CRC handbook, or the Machinist's Handbook, or some other
resource, and I use it in my engineering calculations for the design
and/or construction of some device, how does that play into the scheme
of things? Can I be sued for "illegally using a number" that I did not
manually curate by painstakingly doing whatever it is that NIST does
re: calibration and measurements? What if I use a certain number from
a book/manual in the calibration of an instrument in my lab? What if I
want to import the data from a giant database into SKDB for material
information? How does this work?
What I suspect is happening is that all of these issues are "grey
areas" that most engineers don't care about. I certainly wish that I
could care less about this. Because engineers aren't usually releasing
their calculations and mathematical models of their devices, they
don't have to make they're not upsetting people- all that everyone
else has to go off of is maybe a patent. But this is more a game of
making sure you hide your sources rather than sharing, and since the
legal system isn't used to sharing in a more productive manner, these
questions don't seem to have answers yet. And if these answers do not
exist, what then? Where can I get my data without being shot or yelled
at by Paul?
Disallowing commercial use is often the primary differentiator
Copyright holders use when "dual licensing" a project.
For example: SUN's (now Oracle's) http://VirtualBox.org is released
under the GNU GPL for non-commercial use, and is available under the
"Personal Use and Evaluation License (PUEL)" otherwise.
Commerce is an important part of freedom. Being able to share and
being able to charge a fee of compensation for that sharing has been
part of the GNU GPL since it's precursor = the "Emacs General Public
License" was used by Richard Stallman to sell copies of the GNU Emacs
editor on tape.
In fact a 1987 version of the document
http://Matt.Lee.name/gnu/MIT/gnu/emacs18.41/etc/DISTRIB includes
ordering information!:
"'
________ $150 GNU Emacs source code, on a 1600bpi industry standard
mag tape in tar format. The tape also contains GDB
(the GNU source-level C debugger), MIT Scheme (a dialect
of Lisp), hack (a rogue-like game) and bison (a
compatible replacement for yacc).
'"
Patrick
I agree that putting your work in the public eye creates issues that can
often be swept under the rug by others when source materials are hidden
away. Who knows how much proprietary software infringes on other software
illegally? It's one (illegal) advantage proprietary developers have,
although it breaks down the more employees they have who might report it to
someone.
On the general issue, see my other note on Database copyrights and how you
can't usually copyright facts.
We faced some of the same issues with our garden simulator and information
about botanical plants and their growth parameters. In the end, my wife went
through dozens of gardening books plus many other sources, and found at
least three references for each piece of data. But we were making
essentially a database of information about plants, drawing on other large
collections of data.
That's another thing I should have suggested to Ben in my list of ways to be
safe. If the data is in three different database, you would think it is
safer to use it?
Anyway, I'd also say here, if you can find three unrelated references
stating an engineering value, you are probably on safe ground. But in
general, if it is a fact, it can't be copyrighted, so for individual values,
that is less of a worry.
One might patent a fact though in some indirect way. For example, genes are
facts, aren't they?, but the USPTO says they can be patented. It would not
surprise me if there were some equivalent things that can be patented, like
boat hull curves or something like that. At least patents only last about 20
years and are narrower in their application in restricting personal use.
--Paul Fernhout
> For example: SUN's (now Oracle's) http://VirtualBox.org is released
> under the GNU GPL for non-commercial use, and is available under the
> "Personal Use and Evaluation License (PUEL)" otherwise.
>
> Commerce is an important part of freedom. Being able to share and
> being able to charge a fee of compensation for that sharing has been
> part of the GNU GPL since it's precursor = the "Emacs General Public
> License" was used by Richard Stallman to sell copies of the GNU Emacs
> editor on tape.
Just to clarify, you can use GPL code commercially. What you can't do is
hide your changes or extensions to it if you ship a product.
Some companies (like Trolltech with QT) have used this to offer a GPL
version for free, but also offering a paid-for proprietary version with a
different license for embedding in non-free sofware. (QT is now under the
LGPL because it was bought by Nokia and they have other goals that to sell
software licenses.)
"QT Becomes LGPL"
http://tech.slashdot.org/article.pl?sid=09/01/14/1312210
In order to do this, a company either needs to not accept outside code
contributions under the GPL (because that would force them to only
distribute GPL for the combined work), or they need to make sure any
contributions to the code base need to be given under contributor agreements
that let the company do this. But, a lot of contributors don't like
contributing to such projects under such terms.
--Paul Fernhout
I understand this concern in a "high level" sort of way, but I wonder
if it is the wrong point of attack.
What I mean is: Are we sure non-commercial organizations cannot also
do 'unethical' things?
If the concern has to do with building upon 'open' works while making
the result proprietary,
then I suggest using the GNU GPL which not only requires "share and
share alike", but much
more profoundly requires all *distribution* (conveyance) of the
derivative work include "at cost" access to the
'Sources' of that work - so that it remains 'open' or 'free' in perpetuity.
The GNU AGPL (A == Affero) goes even further by requiring 'Sources' be
available to anyone
who *uses* the Object.
Yes, that is always a risk. But, as long as information is on a public
server, even if the license is restrictive, people planning on doing bad
things can use the information in violation of the license. It becomes
weighing that risk against the benefit to the community at large of having
that information to do good things.
Even as I was writing about focusing on materials, I was realizing that most
explosives could be considered materials, and so presumably fall under open
materials. But, as I said recently in another thread, where would mining and
rocketry and even roads and railroads be without explosives?
I still feel that politically and practically it may be best to avoid
controversial materials for a small site. A group as big as Wikipedia is
better able to defend freedom-of-speech and public right to know issues etc.
> On the other hand I do agree with Paul and the Open Source Initiative
> that we should encourage commercial uses, namely for educational
> purposes such as workshops, classes, etc - that's the purpose of
> research anyway, right?
Just to be clear, while I feel non-commercial use would preclude teachers
using stuff in the course of their paid work, others might argue that
teaching is OK or that non-profit or government use is OK under
non-commercial. The ambiguity is part of the problem.
http://www.google.com/search?hl=en&q=what+does+non-commercial+mean%3F
http://ccelliott.blogspot.com/2007/12/you-say-non-commercial-i-say-huh-what.html
The last link links here:
http://wiki.creativecommons.org/DiscussionDraftNonCommercial_Guidelines
"Thank you for your interest in the NC license term. If you are looking for
the document previously posted on this page as the "NC Draft Discussion
Guidelines", please be aware that having served its purpose to prompt
discussion of the NC term, it has been taken down. As a result of feedback
on the draft discussion guidelines and many other conversations that have
taken place within the community, CC has begun a major study focused on
understandings of "noncommercial use". The results of this study, funded by
the Andrew W. Mellon Foundation, will be made publicly available in 2009.
For more information on the study, you can read the announcement. Please
watch the CC blog for opportunities to participate."
Which links to:
http://creativecommons.org/weblog/entry/9557
But non-commercial can get even vaguer. Is a person putting content on their
site so as to drive interest to that site so they can get consulting
engaging in commercial use? :-)
> And what if someone takes some content
> published on oM, works on it to improve it, and then wants to
> distribute it commercially? That person is bound by our share-alike
> license - do we really want to prevent him/her from making a living on
> it? Aren't we restricting research by not allowing people to do this?
Yes, that's a complex issue. And it is difficult to be caught between the
scarcity world and the post-scarcity world.
> Then I remember that this license allows for exceptions: if people ask
> us we can decide if we want to let them use oM's research for
> commercial purposes, or not.
That is true, as long as the "rightsholder" to the copyright agrees. More on
that below.
> And I start to lean more and more towards
> the notion of using a more limiting license, such as the non-
> commercial one we currently have, and open the door to commercial uses
> in a more informal way, specifically by stating on our site that while
> we do hold a more restrictive license we are open to hear and consider
> all ideas in terms of how to use this work, and that people looking to
> put it towards commercial projects are encouraged to contact us.
Except in practice it is too much trouble in most cases for people to ask.
It breaks the notion of stigmergy, just picking up things and building on them.
On the other hand, it can be a business model. :-)
> Naturally, we'd have to come up with some kind of framework for this
> as these decisions couldn't be based just on our likes and dislikes.
> This framework can in fact be very problematic. And, of course,
> there's still the issue Paul raises about oM being able to use
> Appropedia content but not vice versa, and I don't have an answer to
> that either.
Yes, it is problematic.
> Finally, there is one other issue that worries me: we're an ever
> growing research group, people publishing content on oM tacitly agree
> with the license we are under, but how would this play out when it
> comes to using that content for commercial purposes? Who makes that
> decision? Even with an ethical uses framework there's always some room
> for interpretation and exceptions are bound to arise.
The "rightsholder" decides, or people they delegated that authority to.
Without going back to the original contributor and asking, you have no
obvious right to take content others supply you and relicense it under other
terms. You might try to argue there was implied consent or something, of
course.
> So now, again,
> I'm back at thinking we should allow for commercial uses on the
> license and not make those decisions ourselves. And then I start
> considering possible unethical uses and wondering if people would
> still be so willing to share their research on oM if we switched to a
> commercial license... Finally, if we're faced with that choice, what's
> more important: to provide a 'safe' way to share knowledge and
> resources or to allow it to be used for commercial purposes according
> to the open source standards?
Well, you are at least in good company. Here is a big site, wikiHow with a
similar license to what you have now (although they use CC 2.5, you have CC
3.0, and I don't know if that matters for interchanging information?):
http://www.wikihow.com/Main-Page
http://www.wikihow.com/wikiHow:Creative-Commons
"wikiHow is a collaborative writing project to build the world's largest,
highest quality how-to manual. With your contributions, we can create a free
resource that helps millions of people by offering solutions to the problems
of everyday life. wikiHow currently contains 55,414 articles — written,
edited, and maintained primarily by volunteers. Please join us by writing on
a topic not yet covered, or editing an article that someone else has started."
So, there are a mixture of opinions.
Safety is relative. It may be safe to restrict information to use to prevent
bad things, but it also may be unsafe in preventing some progress.
Almost everything commercial has externalities (both good and bad). But, we
probably can't improve on that without engaging people who do that work.
It's not an easy issue.
GPL vs. LGP vs. BSD have some of the same discussions.
--Paul Fernhout
OpenMaterials also has a wiki, we hope to gather information about
different types of materials, as well as detailed information on
physical properties, uses, techniques and hacks... Whatever people
want to add, and think others will find interesting.
Regarding 1) some associates of ours based at the Futurelab in Linz,
Austria are working on an initiative to help people access materials.
Primarily at this point they are aiming this at artists, but in theory
anyone will be able to access this service. I think they want to
call it 'we will build it', and are hoping to launch in September. so
we would really like to support their efforts to get this up and
running at a community level.
2) There seems to be a current discussion on the open manufacturing
list regarding a map? It would be great if we could find a way of
listing all the fab labs, community workshops, hacker spaces etc that
have resources that people can access (to help manufacture/make
things).
> And what if someone takes some content
> published on oM, works on it to improve it, and then wants to
> distribute it commercially? That person is bound by our share-alike
> license - do we really want to prevent him/her from making a living on
> it? Aren't we restricting research by not allowing people to do this?
Yes, that's a complex issue. And it is difficult to be caught between the
scarcity world and the post-scarcity world.
1) a way for people to source materials
2) a way for people to access fabrication tools, workshops, hacker
spaces etc
We'll be posting and promoting about this google group, and hope to
help promote the open manufacturing list. What other comments,
recommendations do you have to help us support open manufacturing?