www-legal-discuss mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ralph Goers <ralph.go...@dslextreme.com>
Subject Re: Legal questions about moving JXTA/JXSE project to ASF
Date Thu, 28 Oct 2010 04:24:16 GMT
Sam Ruby replied to you last week.  Normally, email is sent to the legal discuss mailing list
and Sam replied to the list, not to you individually. The assumption is that if you are going
to ask a question you will subscribe to the list to participate in the conversation. You can
view his reply at http://mail-archives.apache.org/mod_mbox/www-legal-discuss/201010.mbox/%3cAANLkTi=Y7BNu1eEEdSu_ZUuibVkLA=R5=rRBR=1BNF5C@mail.gmail.com%3e


On Oct 27, 2010, at 8:24 PM, Jérôme Verstrynge wrote:

> Dear Apache, 
> I have sent you the email below last week, but I have not received an answer since. Is
this email list/address still active?
> Thanks, 
> Jérôme Verstrynge
> On 21/10/2010 15:34, Jérôme Verstrynge wrote:
>> Dear Apache, 
>> My name is Jérôme Verstrynge and I am the lead for the latest release of JXTA/JXSE
(2.6). Oracle has officially announced this week that it was withdrawing from the JXTA/JXSE
project. This is something we expected to happen sooner or later. The community's voting is
an overwhelming victory for Apache. I am preparing an application to join the incubator.
>> We have a couple of legal issues we would like to discuss with you.
>> - All JXTA code has been release under Apache License 1.1 (see https://jxta.dev.java.net/jxta_license.htm).
We are currently trying to have it released under Apache 2.0.
>> - Many parts of the code are covered by patents. Some patent are obviously covered
by prior art, but not all.
>> - The string of email below summarizes the discussion in the community.
>> - One of our community member contacted its lawyer and says that since code has been
released under Apache 1.1, we are safe. Do you agree?
>> - Someone on the Apache incubator maillist said that Apache 1.1 license allows sub-licensing?
Does it mean we could move the code under Apache license 2.0 if we can't obtain this from
>> - Someone suggested we re-engineerd existing code, but the lawyer seems to say we
would be in grey area (c.f.,"The grey area comes where a modification of the existing code,
in an area covered by the patents, is so important, that it could be considered not as a modification
of the existing code (covered by the Apache License) but as a new implementation of what is
covered in the patents."). Do it make sense to re-engineer that code?
>> - Oracle/Sun is not eager to transfert the JXTA trademark. We now believe it has
to do with the fact that this name appears in many patent applications. Does it mean we should
pick another name to move to ASF?
>> Thanks, 
>> Jérôme Verstrynge
>> -----------------------------------------------------------------------------------------------------------------------------------------------------------------------
>> On 21/10/2010 9:32, Bruno Grieder wrote:
>>> On patents, here is some initial lawyer feedback (or what I understood), with
the warning that it is coming from an European lawyer and we are talking about software patents,
an American problem.
>>> -the answer is not trivial
>>> -Oracle's withdrawal does not affect the patent situation, in the sense that
a new (emphasis on new) implementation of what is described in the patents without their consent
has always been and is still a breach of their patents (that is true of any patent)
>>> -code made by Oracle released under Apache 2.0 can be freely used under the terms
of this license (in short the current JXTA code is safe since it is "patented" but was released
by Oracle itself under the terms of the Apache License).
>>> The grey area comes where a modification of the existing code, in an area covered
by the patents, is so important, that it could be considered not as a modification of the
existing code (covered by the Apache License) but as a new implementation of what is covered
in the patents. 
>>> IMHO, no immediate problem even fixing the current code but in the long run we
should strive to remove pieces of code clearly covered by Oracle's patents.
>>> Another reason for doing this, is that Simon did a quick patent search yesterday
and found quite a large number of Oracle patents related to JXTA.
>>> Most surprising is US patent 7,788,522 B1 by Ms. Abdelaziz and Traversat, Oracle
being the assignee, and dated August 2010, e.g. a couple of months ago, although no code has
been contributed by Sun/Oracle for a much, much longer period (two years?).  That patent,
like all other patents has the word "jxta" all over the place; that I guess, also gives another
angle to the "trademark issue".
>>> b.
>>> On 20 oct. 2010, at 18:43, Jérôme Verstrynge wrote:
>>>> ASF: It looks like an overwhelming victory for ASF. I'll wait until Friday
before taking action.
>>>> Patent: Again, I am not a lawyer, but I have been trough the process of obtaining
a patent from the USPTO. I worked with Lawyers in Boston. The USPTO challenged our patent
twice and we rebutted them twice. We finally got the patent. 
>>>> I'll repeat myself here (and check with your Lawyer if you don't believe
me), but in a patent, the only thing that matters in court is the claims and the filling date.
If their is prior art covering the claims, then the patent is invalid. In our case, their
is prior art (Gnutella), hence any serious Lawyer would recommend Oracle (or whoever owns
the patent) to not challenge anyone. There is no need to be scared and size does not matter
in this case.
>>>> On the other side, it is not because Oracle announced its withdrawal from
JXTA that the situation was different in the first place...
>>>> JXTA trademark: I don't think Oracle's decision is fair. If they don't change
their mind, I plan to take further action. We don't have to tolerate and accept it this. Oracle
is not standing behind its own commitment. There is a case to put pressure here.
>>>> Jérôme
>>>> On 20/10/2010 12:00, cees_pieters@wxs.nl wrote:
>>>>> Adrian, although I think you are right, I believe that Larry Ellison
only likes to sue big fish, so I suspect we'll be of too little interest for Oracle to make
JXTA an issue. Besides this JXTA has been developed with open source licenses from the start,
so it would seem difficult (at least for me) to see how this could be turned back or restrained.
>>>>> -----Original Message-----
>>>>> From: Adrian Ivan [mailto:adrian@coena.com]
>>>>> Sent: Wed 20/10/2010 10:05
>>>>> To: dev@jxta.dev.java.net
>>>>> Subject: Re: [jxta-dev] Re: [jxta-announce] JXTA Future
>>>>>   don't want to sound paranoic, but maybe one/more company here gets
>>>>> bigger and successful.
>>>>> these giant companies seem to be very sue happy, so we need to be
>>>>> covered from the start.
>>>>> adrian
>>>>> On 10/20/2010 10:59 AM, Bruno Grieder wrote:
>>>>> > Adrian,
>>>>> >
>>>>> > I am not a lawyer (I have one at home + a corporate lawyer; I will
>>>>> > and share the result);
>>>>> >
>>>>> > This is my basic understanding right now. Until yesterday, official
>>>>> > date of withdrawal of Oracle, the code use is "safe". The code was
>>>>> > released, used, modified, etc.. with their full consent and knowledge
>>>>> > under terms clearly agreed and stated by them.
>>>>> >
>>>>> > The problem lies with future modifications to areas covered to their
>>>>> > patents and the use of these.
>>>>> >
>>>>> > I would not overestimate the "danger" though. First, I still hope
>>>>> > can get that letter and then, any "move" by Oracle will likely be
>>>>> > motivated by a business decision, not a legal decision. Right now,
>>>>> > only business decision I can see from them, is cutting their costs
>>>>> >
>>>>> > b.
>>>>> > Amalto Technologies - http://www.amalto.com
>>>>> >
>>>>> > Direct/France: +33 1 83 62 39 73
>>>>> > Mobile: +33 6 33 27 46 85
>>>>> > Skype: bgrieder
>>>>> >
>>>>> > On 20 oct. 2010, at 09:27, Adrian Ivan wrote:
>>>>> >
>>>>> >> To have a clear picture,
>>>>> >>
>>>>> >> as for "business users" of jxta, ie companies that build products
>>>>> >> using jxta technology, how safe are we from the juridical point
>>>>> >> view in regards to Oracle?
>>>>> >>
>>>>> >> Adrian
>>>>> >>
>>>>> >>> On 20 Oct 2010, at 06:25, Bruno Grieder wrote:
>>>>> >>>
>>>>> >>>> Jerôme,
>>>>> >>>>
>>>>> >>>> As you know, as a "business user" I am of those requesting
>>>>> >>>> written authorization on patents from Oracle.
>>>>> >>>>
>>>>> >>>> I am less worried by the validity of the patent or by
>>>>> >>>> chances of winning a patent case, than by the future
>>>>> >>>> offered to Oracle of using the patent as the ground
to start a
>>>>> >>>> case. They are big, and we may be "sued out" whatever
the outcome
>>>>> >>>> of the case may be.
>>>>> >>>>
>>>>> >>>> During the call, I understood from the Oracle representative,
>>>>> >>>> -It has never been and is still not  Sun/Oracle intention
>>>>> >>>> prevent free development and free use of the JXTA technology
in any
>>>>> >>>> way, and as a gesture Oracle will immediately move the
license to
>>>>> >>>> Apache 2.0
>>>>> >>>> -After replacement of GlassFish 2 by GlassFish 3, Sun/Oracle
>>>>> >>>> not be using JXTA in any of its products.
>>>>> >>>> It would therefore be very hard to understand why Oracle
would not
>>>>> >>>> accept granting in writing a free and irrevocable right
to use the
>>>>> >>>> patent.
>>>>> >>>>
>>>>> >>>> Oracle's patent(s) mostly cover the DHT. There are other
>>>>> >>>> algorithms and ready Java implementations out there,
Kademlia in
>>>>> >>>> particular (but also Tapestry, Pastry, etc...) or projects
>>>>> >>>> Apache Cassandra that implement advanced DHT and that
do not suffer
>>>>> >>>> the patent limitations above. I realize, though, that
changing the
>>>>> >>>> DHT algorithm may be a big work touching very low level
>>>>> >>>> of JXTA.
>>>>> >>>>
>>>>> >>>> I would like to hear your opinion and that of other
>>>>> >>>> members on this but I feel we should finish stabilizing
the 2.6
>>>>> >>>> branch "as is" and accelerate the design work on 3.0
>>>>> >>>> significant changes may be required, not to mention
those required
>>>>> >>>> to match Apache requirements, should we eventually go
>>>>> >>>>
>>>>> >>>>
>>>>> >>>> Bruno
>>>>> >>>>
>>>>> >>>> Amalto Technologies - http://www.amalto.com <http://www.amalto.com/>
>>>>> >>>>
>>>>> >>>> On 19 oct. 2010, at 20:19, Jérôme Verstrynge wrote:
>>>>> >>>>
>>>>> >>>>> Dear all,
>>>>> >>>>>
>>>>> >>>>> We just had the conference call with Bernard prior
to this
>>>>> >>>>> announcement. I have been anticipating this decision
and initiated
>>>>> >>>>> some contact with the Apache Software Foundation.
I registered to
>>>>> >>>>> their incubator mailing list and asked whether they
would consider
>>>>> >>>>> an application from the JXTA community, the answer
>>>>> >>>>> In general, I would say that JXTA is more than welcome,
if your
>>>>> >>>>> community thinks that is the best course of action.
>>>>> >>>>> The next step for us is to start an open decision
about where to
>>>>> >>>>> migrate next and to vote. Hence, if anyone has other
>>>>> >>>>> suggestions/proposals to make than ASF, please communicate
it now.
>>>>> >>>>> Then we'll proceed with voting.
>>>>> >>>>>
>>>>> >>>>> Three technical issues need to be solved:
>>>>> >>>>>
>>>>> >>>>> *The licensing* - JXTA is currently operating under
a modified
>>>>> >>>>> Apache 1.1 license. Bernard announced during the
conference that
>>>>> >>>>> he would change this into a Apache 2.0 license on
the current
>>>>> >>>>> JXTA-JXSE website. The SCA that all contributors
signed allows this.
>>>>> >>>>>
>>>>> >>>>> *The 'JXTA' trademark* - It does not seem like Oracle/Sun
>>>>> >>>>> willing to transfer or to authorize its usage to
third a party,
>>>>> >>>>> although it would only require a simple written
authorization from
>>>>> >>>>> them (c.f. our current 1.1 Apache license terms).
>>>>> >>>>>
>>>>> >>>>> On
>>>>> >>>>> http://www.oracle.com/technetwork/topics/oracle-open-source-faq-090399.html,
>>>>> >>>>> Oracle says:
>>>>> >>>>>
>>>>> >>>>> "Oracle is committed to developing, supporting,
and promoting Open
>>>>> >>>>> Source. Oracle has been, and continues to be, committed
>>>>> >>>>> offering choice, flexibility, and a lower cost of
computing for
>>>>> >>>>> end users."
>>>>> >>>>>
>>>>> >>>>> I don't see how disengaging from an Open Source
project and
>>>>> >>>>> forcing the remaining community to fork it under
a new name fits
>>>>> >>>>> with the above commitment of promoting and supporting
Open Source.
>>>>> >>>>> It is plain contradiction and shows little respect
>>>>> >>>>> contributions made by other members of the community
in the past.
>>>>> >>>>> The JXTA project has always been an open source
project and all
>>>>> >>>>> participants have a legitimate claim on the name.
I hope Oracle
>>>>> >>>>> will change its mind and walk its talk.
>>>>> >>>>>
>>>>> >>>>> *The 'JXTA' patent* - Oracle now owns a Sun patent
related to JXTA
>>>>> >>>>> (Patent N°: US 7,464,168 B1). It describes a 'system
and method
>>>>> >>>>> for decentralized entity presence'. The claims of
this patent
>>>>> >>>>> covers for many parts of the technology implemented
in JXTA/JXSE code.
>>>>> >>>>>
>>>>> >>>>> Several participants to the conference call requested
that Oracle
>>>>> >>>>> provides a written authorization to use the technology
>>>>> >>>>> in the patent. Bernard is to investigate and follow-up
on this issue.
>>>>> >>>>>
>>>>> >>>>> On the other side, I am not a Lawyer, but:
>>>>> >>>>>
>>>>> >>>>> i) What is relevant in patents is the content of
claims. The rest
>>>>> >>>>> of the document has no value, but informative value.
>>>>> >>>>> i) From Wikipedia (and my own experience with patents):
"Prior art
>>>>> >>>>> (...) constitutes all information that has been
made available to
>>>>> >>>>> the public in any form before a given date that
might be relevant
>>>>> >>>>> to a patent's claims of originality. If an invention
has been
>>>>> >>>>> described in prior art, a patent on that invention
is not valid."
>>>>> >>>>> ii) In
>>>>> >>>>> http://www.uspto.gov/patents/resources/methods/aiplafall02paper.jsp,
>>>>> >>>>> at section IV.E., the USPTO says: "Applicant's own
work, available
>>>>> >>>>> to the public more than one year prior to the filing
date, is
>>>>> >>>>> prior art."
>>>>> >>>>> iii) The JXTA patent was filed on October 19, 2004.
>>>>> >>>>> iv) In July 2003, the status of the Gnutella 0.4
standard was
>>>>> >>>>> already available on:
>>>>> >>>>> http://rfc-gnutella.sourceforge.net/developer/index.html
>>>>> >>>>> v) The Gnutella 0.4 protocol defines a ping functionality
'used to
>>>>> >>>>> actively discover hosts on the net'.
>>>>> >>>>> vi) The above (v) constitutes prior art on several
claims of the
>>>>> >>>>> JXTA patent. Since remaining claims rely on those
claims, they
>>>>> >>>>> fall too in a domino pattern. What is left is insufficient
>>>>> >>>>> itself to constitute an invention.
>>>>> >>>>>
>>>>> >>>>> If Oracle were to refuse to provide a right to use
the technology
>>>>> >>>>> described in the patent, and therefore, walk their
talk, it would
>>>>> >>>>> be very risky for them to attempt any challenge
against any user
>>>>> >>>>> of the technology. I believe most court would consider
>>>>> >>>>> evidence for prior art and declare the patent invalid.
>>>>> >>>>>
>>>>> >>>>> Waiting for your feedback,
>>>>> >>>>>
>>>>> >>>>> Jérôme
>>>>> >>>>>
>>>>> >>>>>
>>>>> >>>>> On 19/10/2010 19:10, Bernard Traversat wrote:
>>>>> >>>>>>
>>>>> >>>>>> It has been a long journey since Sun started
the JXTA open-source
>>>>> >>>>>> project more than 10 years ago. For the past
>>>>> >>>>>> of years, investments in the JXTA technology
from Sun diminished
>>>>> >>>>>> to the point where the project and contributions
>>>>> >>>>>> completely managed by external community members.
As Sun, Oracle
>>>>> >>>>>> has no plan to further invest in JXTA.  One
of the reasons
>>>>> >>>>>> that we developed JXTA has an open-source project
was to allow
>>>>> >>>>>> the community to evolve the technology if the
>>>>> >>>>>> of the core team were changed.
>>>>> >>>>>>
>>>>> >>>>>> We already started to engage community leaders
to ensure a smooth
>>>>> >>>>>> transition. We will be maintaining the source
>>>>> >>>>>> and jxta.org <http://jxta.org/> site while
a new home for both
>>>>> >>>>>> the code and supported content is investigated
by the community.
>>>>> >>>>>>
>>>>> >>>>>> I would like to thank all those who have supported
the activities
>>>>> >>>>>> of JXTA over the years, and to extend my best
wishes to
>>>>> >>>>>> the community for future success in their endeavor
to evolve the
>>>>> >>>>>> technology.
>>>>> >>>>>>
>>>>> >>>>>> B.
>>>>> >>>>>>
>>>>> >>>>>
>>>>> >>>>
>>>>> >>>
>>>>> >>
>>>>> >

View raw message