|
First, take a deep breath ...
The basic ideas Charles expresses are, in my view, on
track. But some of
the details are biased toward his interests. For example, VisualDOE
and eQUEST
have both benefited from substantial support from California utilities use
of "Public
Goods Charge" (PGC) funds. Both have had very substantial investment
by
their developing companies (Eley for VisualDOE and JJHirsch for
eQUEST.)
eQUEST is freeware for users and can be licensed by anyone for
further
development without per copy royalty. VisualDOE, currently, is
marketed
as a commercial product. So, I do not understand; under the
policy
proposed is it suggested that eQUEST should be sold for a fee and not
licensed
for further development, or that VisualDOE should be treated the same
a
eQUEST - freeware and source licensed to anyone, or that neither
should
have been funded in the first place?
Policy can treat private funded projects differently from
Government funded
projects but for the policy to work it must treat different Government
funded
projects similarly and different private funded projects similarly.
Obviously the key here is that both VisualDOE and eQUEST would not
exist without
the combination of public and private investment; the partnership that
Charles' paper
describes. In both cases, I believe, the public funding should
"extract" some
public benefit - such as the basic product being freeware (or very cheap)
for
end users and the source being available (cheap) for developers - but
the
private funding should provide some advantage to the developer as a
"reward"
for their private investment. I think the "reward" is that the
private developers
get a head start, they retain copyright (so have an advantage over open
source
in being able to do private products based on the starting point), and they
gain
the internal know-how/experience from the project. Future such
public-private
partnerships should allow the developers to continue to retain
ownership
of the copyright, but require that the "starting point" which is developed
using
the public funds be placed out as open source under a Mozilla-like license;
the
developer could, however, continue on with their original source copy to
develop
private version licensed in any way they see fit.
The California investor owned utilities (and EPRI), when deciding to
support
QUEST development "extracted" the public benefit but other California
utilities,
when deciding to support VisualDOE, Cool Tools, and other related
projects,
did not "extract" much that is yet seen. This later case should
change. A similar
problem exists with California Public Interest Energy Research (PIER)
funding; they are spending millions on software tools in this field, but
have
not published the terms to be used in the licenses for the resultant
software.
When the support for the software comes ENTIRELY from public funds - be
it
DOE, CEC/PIER, etc. - the ENTIRE resulting work should be made open
source under the BSD-like license so that anyone can run with it in any
way.
Such software should not use the Mozilla-like open source
license, as that format
was intended for use when the posted source contains substantial
private
investment; the Mozilla format "rewards" the copyright holder by leveling
the
playing field for all but him/her who decided to make their private work
open.
For EnergyPlus, Energy10, DBA, Radiance, Superlite, Window-4, BLAST,
CalERDA-1 through DOE-2.1C, and all the many such fully publicly funded
software,
the route to take is clear: BSD-like license so that anyone can run with it
in
any way on a level playing field. However, I also believe that we
don't need full
packaged commercial products from 100% government funds using 100%
directed procurements (sole source contracting); full products, if needed,
should
be bid out as open procurements with a requirement for real
cost-sharing.
If no bidder is willing to put up private money (not just another pot of
public funds)
it is doubtful the product is needed. If a full product is needed, it
is best developed
within the private sector by companies who will, in the end, need to market
it and
support it and who probably also understand the customer needs better than
the
US Government, National Labs, or other public funding sources.
That said, however, I am 100% in favor of Government funding
of basic algorithm,
model and technology development; as I have said before, many
advancements
in this field would not exist without this funding. EnergyPlus the
LBNL/UI
commercial product - NO; EnergyPlus the BSD-like open source
simulation
toolkit maintained by LBNL and UI (and hopefully others in
the future) - YES.
Energy10 the NREL commercial product - NO; Energy10 the BSD-like open
source
toolkit copyrighted by winning bidder XYZ company (and the private
by-product
being XYZE10+ enhanced $1000 version) - YES. VisualDOE the
commercial
product - NO; VisualDOE the Mozilla-like open source toolkit
copyrighted
by winning bidder Eley Associates (and the private by-product being
Eley
VisualDOE+ enhanced $1000 version) - YES. eQUEST the commercial
product - NO; eQUEST the Mozilla-like open
source simulation and interface toolkit
copyrighted by winning bidder JJHirsch Associates (and the
private by-product
being JJH eQUESTpro enhanced $1000 version
- well, we pass on that one for
now, but reserve the right to do it) - YES.
DOE-2 is a good example - we used several million dollars of private money
in addition to several hundred thousand of public money (DOE and CA
PGC
funds) to do our part of DOE-2.1D, DOE-2.1E and DOE-2.2. We
could
not have done this if the DOE-2 portions supported by DOE had a
royalty
attached or had licensing that restricted redistribution (like Mozilla or
like
the proposed, even more restrictive licensing for EnergyPlus) since
we
could not have demonstrated to our private funders how THEY would
benefit. DOE "extracted" from us royalty free DOE-2.2 distribution to
the
public; even though US Government contracting of software development
to
small businesses requires no such license grant for the privately
funded
portions, DOE felt this DOE-2.2 license grant was required to fulfill
their
public interest responsibility (though they have chosen, so far, not
to
use that license grant to distribute DOE-2.2 to the public.) This is
also
why I raised the question: does the user/developer community believe
we should move to open source licensing for DOE-2? The policy should not be to say that US public money
should be spend on engines but
not interfaces; the policy should be - if the need
exists, the priority is high enough,
and the private sector cannot do it well enough or fast enough alone,
plus
US public funds are available, then do it all ... but do it in a way that
partners with,
enables, and enhances the competitiveness of US companies in the world
market. --- Jeff Hirsch James J. Hirsch & Associates Building Performance Analysis Software & Consulting 12185 Presilla Road Camarillo, CA 93012-9243 USA phone: (805) 553-9000 fax: (805) 532-2401 email: Jeff.Hirsch@xxxxxxxx web: http://DOE2.com =====================================================You received this e-mail because you are subscribed to the BLDG-SIM@xxxxxxxx mailing list. To unsubscribe from this mailing list send a blank message to BLDG-SIM-UNSUBSCRIBE@xxxxxxxx |