Re: [office-comment] Patent back door for Sun? (ODF all versions)

[marbux <marbux@gmail.com>]

On Sun, Mar 1, 2009 at 5:17 AM, Alex Brown <alexb@griffinbrown.co.uk> wrote:
> As I understand it, the IPR regime governing ODF is in part comprised of
> Sun's non-assertion statement (http://is.gd/liv1). According to this,
>
> "Sun irrevocably covenants that [...] it will not seek to enforce any of
> its enforceable U.S. or foreign patents against any implementation of
> the Open Document Format for Office Applications (OpenDocument) v1.0
> Specification, or of any subsequent version thereof ("OpenDocument
> Implementation") in which development Sun participates to the point of
> incurring an obligation, as defined by the rules of OASIS [...]"
>
> We learn about such an "obligation" from the OASIS rules
> (http://is.gd/livR), and that, for a party withdrawing from a TC:
>
> "A TC Party that has incurred a Participation Obligation prior to
> withdrawal continues to be subject to its Participation Obligation but
> only with respect to OASIS Draft Deliverable(s) approved more than seven
> (7) calendar days prior to its withdrawal."
>
> IANAL, but is a correct reading of this that if Sun withdraws from the
> ODF TC, then all outputs of the TC (and implementation of outputs)
> dating from 7 days prior to that date, and extending into the future,
> are fully subject to any patent claims Sun might then make? (I notice
> the verb form here is "participates", not "has participated"). Perhaps a
> more legally-minded reader could clarify ...
>
> If this is the case, is this acceptable to OASIS? I doubt it would be
> acceptable to JTC 1 (newly sensitized to IPR issues by events
> elsewhere), or to the "FOSS community", who enjoy a distinctly
> problematic relationship with Sun.
>
> What is needed here is a clearly-stated, perpetual waiver of patent
> rights from all parties who potentially hold them, that applies to all
> ODF drafts, specifications and standards published by OASIS embodying
> those technologies; not a hard-to-parse, qualified and conditional
> waiver. DIS 29500 has taught us that such clarity is necessary.
>
> Enlightenment / reassurance please!

Not licensed to practice law since my retirement, so this is only a
personal opinion, not legal advice ---

In the hypothesized situation regarding Sun, yours is one reasonable
interpretation of the plain language. But I think it highly likely
that a court would limit applicability to new features introduced into
the specification after the seven-day period, possibly even to new
features proposed after that period. I.e., those aspects of a new
standard version that were part of the specification prior to Sun's
hypothesized departure would remain immune even if incorporated into a
new version of the standard after Sun's departure.

Sun's covenant not to sue extends beyond what is required by OASIS
policy in several respects, one of which is a provision that grants
patent immunity to any version of the standard in which Sun has
participated to an extent that it incurs a participant obligation.
<http://www.oasis-open.org/committees/office/ipr.php>. So under my
reading, the TC would remain free to add features potentially subject
to Sun patent claims after Sun's departure, for the then currently
in-development version of the standard. But this would be under the
Sun covenant, not the OASIS IPR Policy.

However, the situation is far more complex than just the interaction
of OASIS IPR policy with Sun's covenant not to sue.  Sun is not the
only company or individual that has or will participate on the ODF TC
who at least in theory may have relevant patents.

For example, IBM has its own covenant not to sue in regard to ODF that
is in some respects more restrictive than Sun's but still in the
ballpark for ODF implementation by open source software developers.
<http://www-03.ibm.com/linux/opensource/isplist.shtml> (don't miss the
footnoted definitions at the bottom of the page).

I am not aware of any other companies or individuals who have
participated on the ODF TC and have announced what promises they will
make as to implementation rights potentially subject to any patents
they may hold. For these folks, the only limits are those imposed by
OASIS IPR Policy sections 10.2.1 and 10.2.3.
<http://www.oasis-open.org/who/intellectualproperty.php#licensing_req>.

And squarely to the point, section 10.2.1 allows participants to
impose a condition that prohibits sub-licensing, a situation that
would  create an irreconcilable conflict with many open source
software licensing scheme, such as the Gnu GPL and LGPL.

While I can agree that DIS-29500 taught some hard lessons on the
subject, I think it unfortunate that Microsoft's managers were not
among those who took that lesson to heart. I think it all too likely
that Microsoft will announce at some point that its Open Specification
Promise, which severely limits implementation of ISO/IEC:29500-2008,
has been extended to cover ODF. See
<http://www.microsoft.com/Interop/osp/default.mspx>.

If Microsoft does so, that could conceivably have the effect of
casting a patent cloud over ODF implementation. See Roger Parloff,
Microsoft Takes On the Free World, Fortune (14 May 2007),
<http://money.cnn.com/magazines/fortune/fortune_archive/2007/05/28/100033867/>
(Microsoft claims to hold 45 patents reading on OpenOffice.org, some
of which conceivably could read on other implementations of ODF); see
also Sun-Microsoft, Limited Patent Covenant and Stand-Still Agreement
Dated April, [sic] 2004,
<http://www.sec.gov/Archives/edgar/data/709519/000119312504155723/dex10109.htm>,
especially section IV (Sun agreed that Microsoft could sue licensees
of OpenOffice.org for patent infringement without Sun interference,
whilst obtaining Microsoft patent protection for StarOffice).

For a studied critique of the Microsoft Open Specification Promise by
legal scholars, see this paper published by the University of New
South Wales Faculty of Law,
<http://law.bepress.com/unswwps/flrps/art71/>

But I stress that OASIS is not unusual in this regard. For example,
ISO/IEC's joint patent policy  allows even more restrictions on
implementation than the relevant OASIS policy sections do, including
royalty requirements. ISO/IEC offers three options in regard to patent
right promises, and Microsoft chose RAND-Z for OOXML. RAND-Z is RAND
with zero royalties. For a very quick overview of the common patent
right promise regimes in standards organizations, I recommend this
article written by Microsoft standards attorney David Rudin.
<http://standardslaw.com/wordpress/>.

In my opinion, It's largely a situation of most IT standard
development organizations not having yet adapted to the rise of open
source software business models. I suspect they either will adapt and
provide patent promise regimes compatible with the major open source
software licenses or become largely irrelevant.

Another plausible outcome is that software patents go away. My view is
that it is mainly the U.S. that is prolonging the demise of software
patents. And there are signs aplenty that the U.S. Supreme Court is
looking for a suitable case in which to decide whether software
patents are lawful. That Court has never approved software as
patentable subject matter. Software patents in the U.S. are the
creation of the U.S. Patent Office and the Court of Appeals for the
Federal Circuit (and a predecessor court with a different name).

Best regards,

Paul E. Merrell, J.D. (Marbux)

-- 
Universal Interoperability Council
<http:www.universal-interop-council.org>