Depuis le début, je cherchais des faits, car les affirmations péremptoires des uns et des autres ca me fait penser aux joutes oratoires stériles de nos politiciens.
L'ennui, c'est que le 1er doc concret trouvé: - la note de Deutsche Bank Securities Inc. datée du 7 aout 2003 - me fait douter. En plus son style écrit avec beaucoup d'humilité: que des faits, aucune opinion des redacteurs qui prennent la peine de préciser qu'ils n'ont aucun interessement financier dans cette étude. http://equities.research.db.com.(...)
En voici quelques extraits qui m'ont paru bien troublants:
.../...
Clearly we are not passing legal judgment on these contracts, but below are
several observations.
The September 1995 agreement between SCO and Novell is titled
"Asset Purchase Agreement". It clearly appears as an asset purchase
not a licensing transaction.
The 1995 Asset Purchase Agreement was amended in 1996 to address
what looked like a potential conflict in Schedule 1.1b of the original
asset purchase. The original document potentially did not include the
copyright ownership by SCO, but Amendment X seems to further clarify
this. According to management, this amendment coincided with but
preceded a significant IBM license arrangement.
According to SCO management, Sun is the only company with a
license to distribute source code and derivative work. IBM (and its
Sequent contract) have rights, according to management for binary
distribution of the product.
According to SCO management, Sequent's license with AT&T is more
restrictive than IBMs. Sequent does not have the right to distribute
source code, methods and concepts while IBM is not restrained on the
concepts front. At issue is contributions of derivative work from IBM
and Sequent into the Linux kernal.
.../...
.../...
We reviewed a portion of the Unix V kernal source code and the 2.4 Linux
kernal source code. SCO management highlighted examples of code that
could potentially represent literal duplication of source code. Compared
side-by-side, the code was identical - same variables, declaratives,
nomenclature, style and comments. The primary difference was the
removal of the copyright notices that appeared in the UNIX code which
were missing in the Linux code.
The assertion by SCO management is that they own the copyright and
rights to the UNIX kernal code and that it was placed into the Linux 2.4
kernal (the kernal code we reviewed was NOT, according to them,
contributed by IBM) without their permission. Clearly this will be an issue
to be determined by the courts, but this sort of code review by
technologists and attorneys in large corporations is likely to raise some
potential concerns around legal exposure when running Linux.
On occasion in the source code, variable names may have been changed,
the style of declaring the variables differed, but nonetheless it appears near
identical. According to SCO management, in a 2002 judgement by the 3rd
Circuit court (Dun & Bradstreet versus Grace Consulting), IP violations were
upheld with only 27 lines of literally copied code out of hundreds of
thousands.
.../....
et plus loin:
.../...
SCO management acknowledges that portions of the duplicate code can be
rewritten, but this effort would have to be undertaken by companies in a
"clean" environment where the existing code could not be used as a model
for replacement code. But their primary assertion here is that the enterprise
scalable code contributed by IBM/Sequent (NUMA - Non Uniform Memory
Access, RCU - Read, Copy, Update, JFS - Journal File System, SMP -
Symmetric Multiprocessor Support, scheduler improvements, logical
partition support) all requires enormous resources to recreate and that the
derivative work contributed by commercial vendors was done without their
permission and in violation of their contracts. For SMP alone, they contend
that 150 header files and 600 source code files in Linux 2.4 are derivative
contributions (700 source code files for 2.5 Linux). This could represent, in
their opinion, 3-5 years of new effort and 25-50 million dollars of
commercially-inspired efforts.
The views expressed in this report accurately reflect the personal views of
the undersigned lead analyst about the subject issuers and the securities of
those issuers. In addition, the undersigned lead analyst has not and will not
receive any compensation for providing a specific recommendation or view
in this report. -Brian Skiba, Matthew Kelly
Je finis par y croire et me demander si l'une des 2 hypothèses suivantes ne serait pas la bonne:
1/ Un des équipes de devpt a effectivement trompé la communauté en pompant les lignes incriminées .
2/ C'est un coup fourré préparé de longue date, et le noyautage du noyau (Sic !!!)a été bien préparé.
Bref, une sale histoire ..... qui laissera des traces quelle que soit la suite des actions IBM / SCO sur la licence GPL.
# Re: Affaire SCO/Linux : La note du 7 aout de la Deutsche Bank
Posté par Wharf . En réponse à la dépêche Affaire SCO/Linux : IBM contre-attaque. Évalué à 3.
L'ennui, c'est que le 1er doc concret trouvé: - la note de Deutsche Bank Securities Inc. datée du 7 aout 2003 - me fait douter. En plus son style écrit avec beaucoup d'humilité: que des faits, aucune opinion des redacteurs qui prennent la peine de préciser qu'ils n'ont aucun interessement financier dans cette étude.
http://equities.research.db.com.(...)
En voici quelques extraits qui m'ont paru bien troublants:
.../...
Clearly we are not passing legal judgment on these contracts, but below are
several observations.
The September 1995 agreement between SCO and Novell is titled
"Asset Purchase Agreement". It clearly appears as an asset purchase
not a licensing transaction.
The 1995 Asset Purchase Agreement was amended in 1996 to address
what looked like a potential conflict in Schedule 1.1b of the original
asset purchase. The original document potentially did not include the
copyright ownership by SCO, but Amendment X seems to further clarify
this. According to management, this amendment coincided with but
preceded a significant IBM license arrangement.
According to SCO management, Sun is the only company with a
license to distribute source code and derivative work. IBM (and its
Sequent contract) have rights, according to management for binary
distribution of the product.
According to SCO management, Sequent's license with AT&T is more
restrictive than IBMs. Sequent does not have the right to distribute
source code, methods and concepts while IBM is not restrained on the
concepts front. At issue is contributions of derivative work from IBM
and Sequent into the Linux kernal.
.../...
.../...
We reviewed a portion of the Unix V kernal source code and the 2.4 Linux
kernal source code. SCO management highlighted examples of code that
could potentially represent literal duplication of source code. Compared
side-by-side, the code was identical - same variables, declaratives,
nomenclature, style and comments. The primary difference was the
removal of the copyright notices that appeared in the UNIX code which
were missing in the Linux code.
The assertion by SCO management is that they own the copyright and
rights to the UNIX kernal code and that it was placed into the Linux 2.4
kernal (the kernal code we reviewed was NOT, according to them,
contributed by IBM) without their permission. Clearly this will be an issue
to be determined by the courts, but this sort of code review by
technologists and attorneys in large corporations is likely to raise some
potential concerns around legal exposure when running Linux.
On occasion in the source code, variable names may have been changed,
the style of declaring the variables differed, but nonetheless it appears near
identical. According to SCO management, in a 2002 judgement by the 3rd
Circuit court (Dun & Bradstreet versus Grace Consulting), IP violations were
upheld with only 27 lines of literally copied code out of hundreds of
thousands.
.../....
et plus loin:
.../...
SCO management acknowledges that portions of the duplicate code can be
rewritten, but this effort would have to be undertaken by companies in a
"clean" environment where the existing code could not be used as a model
for replacement code. But their primary assertion here is that the enterprise
scalable code contributed by IBM/Sequent (NUMA - Non Uniform Memory
Access, RCU - Read, Copy, Update, JFS - Journal File System, SMP -
Symmetric Multiprocessor Support, scheduler improvements, logical
partition support) all requires enormous resources to recreate and that the
derivative work contributed by commercial vendors was done without their
permission and in violation of their contracts. For SMP alone, they contend
that 150 header files and 600 source code files in Linux 2.4 are derivative
contributions (700 source code files for 2.5 Linux). This could represent, in
their opinion, 3-5 years of new effort and 25-50 million dollars of
commercially-inspired efforts.
The views expressed in this report accurately reflect the personal views of
the undersigned lead analyst about the subject issuers and the securities of
those issuers. In addition, the undersigned lead analyst has not and will not
receive any compensation for providing a specific recommendation or view
in this report. -Brian Skiba, Matthew Kelly
Je finis par y croire et me demander si l'une des 2 hypothèses suivantes ne serait pas la bonne:
1/ Un des équipes de devpt a effectivement trompé la communauté en pompant les lignes incriminées .
2/ C'est un coup fourré préparé de longue date, et le noyautage du noyau (Sic !!!)a été bien préparé.
Bref, une sale histoire ..... qui laissera des traces quelle que soit la suite des actions IBM / SCO sur la licence GPL.
Bye