decoration decoration
Stories

GROKLAW
When you want to know more...
decoration
For layout only
Home
Archives
Site Map
Search
About Groklaw
Awards
Legal Research
Timelines
ApplevSamsung
ApplevSamsung p.2
ArchiveExplorer
Autozone
Bilski
Cases
Cast: Lawyers
Comes v. MS
Contracts/Documents
Courts
DRM
Gordon v MS
GPL
Grokdoc
HTML How To
IPI v RH
IV v. Google
Legal Docs
Lodsys
MS Litigations
MSvB&N
News Picks
Novell v. MS
Novell-MS Deal
ODF/OOXML
OOXML Appeals
OraclevGoogle
Patents
ProjectMonterey
Psystar
Quote Database
Red Hat v SCO
Salus Book
SCEA v Hotz
SCO Appeals
SCO Bankruptcy
SCO Financials
SCO Overview
SCO v IBM
SCO v Novell
SCO:Soup2Nuts
SCOsource
Sean Daly
Software Patents
Switch to Linux
Transcripts
Unix Books
Your contributions keep Groklaw going.
To donate to Groklaw 2.0:

Groklaw Gear

Click here to send an email to the editor of this weblog.


Contact PJ

Click here to email PJ. You won't find me on Facebook Donate Paypal


User Functions

Username:

Password:

Don't have an account yet? Sign up as a New User

No Legal Advice

The information on Groklaw is not intended to constitute legal advice. While Mark is a lawyer and he has asked other lawyers and law students to contribute articles, all of these articles are offered to help educate, not to provide specific legal advice. They are not your lawyers.

Here's Groklaw's comments policy.


What's New

STORIES
No new stories

COMMENTS last 48 hrs
No new comments


Sponsors

Hosting:
hosted by ibiblio

On servers donated to ibiblio by AMD.

Webmaster
Some Observations on Oracle v. Google, by Mark Webbink, Esq.
Wednesday, May 11 2011 @ 07:12 PM EDT

Some Observations on Oracle v. Google,
~ by Mark Webbink, Esq.

Before reviewing the status of the lawsuit and the claims interpretation, a quick word on how the parties got to the point of a lawsuit. This was not a shot that caught Google off guard. Oracle and Google had been in discussions for months prior to this lawsuit over a license that would have permitted Google to move forward. Word is that the companies had all but reached agreement, but when the Google negotiators sent the proposal up the ladder for final approval, it got nixed at the very highest levels within Google. The fact that the two companies would have gotten that far and then failed to close the deal should not be interpreted against either company. Oracle had its reasons for seeking a license arrangement; Google had its for rejecting the arrangement. But when Google rejected the arrangement, they had to have known the lawsuit was coming.

Back up for a second to Sun before Oracle. Sun's licensing arrangements around Java have always been complex. While licensees were well aware that Sun had patents related to Java, those patents were rarely, if ever, mentioned. There was no need to mention them. If you agreed to Sun's licensing structure that would permit you to create your own version of Java, that could pass the test suite, and you could call Java, then Sun threw the patents in for free. Various open versions for Java, including jBoss, have been subject to this arrangement. You can see some of this structure in paragraphs 10 and 11 of Oracle's Reply [PDF] to defendant Google Inc.'s Amended Counterclaims.

Google believed (and believes) it avoided this licensing structure by implementing a clean room version of the Java runtime. The problem with clean rooms is that, while they may help avoid copyright claims, they are not particularly helpful in avoiding patent claims -- the ol' two-edge sword of software. So if you are going to develop a new implementation of something like the Java run-time environment, you have to not only use a clean room in order to avoid copyright claims, you also have to work around any relevant patents (and this doesn't require a clean room). Suffice it to say that the approach Google has taken has some potential holes in it with respect to patents. Of course, Google believes the Oracle patents are either invalid or not infringed in this instance. [Editorial aside – none of this commentary is intended to imply that patents are a good thing for software; in the eyes of this writer they clearly are not.]

One other thing to mention with respect to the copyright claims Oracle has asserted, some technical interpretations imply that Google's Dalvik merely recompiles Java classes into the Dalvik format. If that is true, then such recompilations could be construed as a derivative work, but let's not go too far down that path at this point. The patent claims Oracle has asserted are far more important than the copyright claims.

So let's turn to the claims construction as it is moving through the court. First, the parties agreed on the interpretation of a number of key terms in the stipulation [PDF] they filed on February 22nd. Stipulated interpretations include:

function;

machine instruction;

native machine instruction/native instruction;

virtual machine instructions [these first five all relating solely to the '205 patent];

a processor, a memory a class preloader . . .;

and a processor, a memory means for executing . . . [these last two all relating solely to the '720 patent].

Definitions in dispute were/are:
resolve / resolving (‘104 patent);

computer-readable medium (‘104, ‘447, ‘476, and ‘520 patents) / computer usable medium (‘702 patent) / computer-readable storage medium (‘720 patent);

intermediate form code / intermediate form object code (‘104 patent);

the play executing step (‘520 patent);

reduced class file (‘702 patent);

and symbolic (data/field) reference (‘104 patent).

In his tentative claim construction order [PDF] dated April 28, Judge Alsup proposed definitions for each of these terms. Oracle and Google have now responded (Groklaw documents 132 [PDF] and 135 [PDF]). The parties seem to largely accept the court's definitions, with the exception of the last term “symbolic (date/field) reference,” to which Oracle objects on grounds that the court's interpretation is limited to dynamic referencing, while Oracle believes the definition should include both static and dynamic referencing. At the same time, Google seeks to preserve its assertion of invalidity with respect to claims relying on some of these definitions.

The final action to date is a proposed order [PDF] issued May 3rd by Judge Alsup seeking to streamline the case by concentrating on those claims of greatest import. Judge Alsup proposes this be done in a series of step-downs, starting with all 132 claims asserted by Oracle and all of Google's prior art references on which Google intends to rely. At the end of the step-downs, Oracle will be limited to asserting only three claims at trial (presumably those three it believes most greatly substantiate Google's infringement and most capable of surviving an invalidity challenge). At the same time, Google would be limited in the end to just eight prior art references with which to defeat those three claims. Not surprisingly, Google is thrilled with this approach (Groklaw document 134 [PDF]) while Oracle argues that it will be deprived of substantial due process and appropriate remedy by the approach (Groklaw document 133 [PDF]).

In addition to suggesting this streamlined approach to the trial (presently slated to commence October 31, 2011), Judge Alsup asked each party its opinion of waiting until all of the patents have cleared reexamination (two of the patents, '720 and '205 are in inter partes reexamination while all of the remainder are in ex parte reexamination). As both Google and Oracle point out in their responses, it is likely to be three years or more before the reexaminations are all complete. Oracle concedes that, depending on what happens in reexamination, the reexaminations may moot the issues at trial. Google is more emphatic. It points to the high probability (rising to the level of almost a virtual certainty) that all of the claims in all of the patents will be narrowed to some extent in reexamination and that waiting until the results of those reexaminations are known will in all likelihood moot the trial issues before the court.

It will certainly be interesting to see how Judge Alsup resolves these issues over the next days.


  


Some Observations on Oracle v. Google, by Mark Webbink, Esq. | 306 comments | Create New Account
Comments belong to whoever posts them. Please notify us of inappropriate comments.
corrections
Authored by: julian on Wednesday, May 11 2011 @ 07:33 PM EDT
please put the korrection -> correction
in the title block and a hint in the comment where the problem is.

---
John Julian

[ Reply to This | # ]

[NP] News Picks threads
Authored by: Aladdin Sane on Wednesday, May 11 2011 @ 07:48 PM EDT
News Picks comments and observations here, please.

---
Well, it's either Captain Kirk or Doctor Strangelove. —me, ca. 1984-1985

[ Reply to This | # ]

[OT] Off topic threads
Authored by: Aladdin Sane on Wednesday, May 11 2011 @ 07:50 PM EDT
Behave in a manner which obviates comments that might suggest that your comment would go to the topic of the matter here. Ahem.

---
Well, it's either Captain Kirk or Doctor Strangelove. —me, ca. 1984-1985

[ Reply to This | # ]

[COMES] Comes v. MS transcripts
Authored by: Aladdin Sane on Wednesday, May 11 2011 @ 07:51 PM EDT
Gosh, golly. Put Comes docs here and rock and roll.

---
Well, it's either Captain Kirk or Doctor Strangelove. —me, ca. 1984-1985

[ Reply to This | # ]

Compiler output and derivative works
Authored by: jbb on Wednesday, May 11 2011 @ 07:56 PM EDT
Mark Webbink, Esq. opined:
If that [Dalvik recompiles Java classes] is true, then such recompilations could be construed as a derivative work, but let's not go too far down that path at this point.
Mikko Välimäki Ph.D, LL.M (20-page pdf) seems to have a different opinion:
If compiler output is regarded as a derivative work of both the source code and the ‘work’ of a compiler, could the author of the compiler have also rights to the resulting output? If the output does not include any code or expression from the compiler, the answer seems to be straightforward no.
IMO, any other interpretation would lead to madness, making software patents look like a walk in the park by comparison. While the output of a complier could very well be a derivative work, it is a derivative work of the input, not of the compiler itself. Are you really suggesting there is even a possibility that Microsoft has a copyright claim on any binary code created by one of its compilers? If I use sed on a file is that file now a derivative work of sed? ISTM the entire software industry would grind to a sudden halt of what you are implying were true.

---
[ ] Obey DRM Restrictions
[X] Ignore DRM Restrictions

[ Reply to This | # ]

Some Observations on Oracle v. Google, by Mark Webbink, Esq.
Authored by: Anonymous on Wednesday, May 11 2011 @ 08:09 PM EDT
There are two cases here; 1) patent and 2) copyright. At the time, I do not
think Google really understood the copyright law or the case, but in Aug 2008
the copyright law change to incorporate a concept of breach of license/contract
CAFC ruling, creating the $0 copyright law. The way the claims are written
against Google and what Oracle has covered, Google will be found guilty in
summary Judgment on Copyright Infringement (IMHO). The trial will be on patent
claims, but Google is a copyright infringer under copyright law and CAFC 2008
and 9th circuit case law.
<br>
What is interesting about the 9th circuit copyright case law is the standard –
it is so low. If I read the case law that generated this new CAFC action, the
case law is based on a collection of items made by a conversion tool. The case
that generated the claim was based on a mistake a contractor made. Given that
the basis of copyright in 9th circuit is so low, Google does not have a license
defense, so it will be interesting to see.
<br>
What is bad news for Google on this is that fact that Google is looking at
Sanctions + fees, so if Oracle wanted to… it is completely within Oracle’s power
to have all of the android devices destroyed under an injunction, and to
potentially enjoin Google from shipping any droid device. <br>
I do not think we will truly know the damages until we see the quarterly SEC
filing from Google. It is so close to the November date that they will need to
take a reserve to address the lawsuit. Since the copyright damages alone are in
the $15 – $30B range, this will be something that is not going to be hidden in
the SEC filing.

<br>
When Google tried to show that the copy right code was not copyrightable as a
defense that was truly telling, because the standard on copyright is so low.
<br>
Should be interesting to see how this plays out.

[ Reply to This | # ]

Questions for Mark here
Authored by: webster on Wednesday, May 11 2011 @ 08:10 PM EDT
.

Is there a way to suggest or argue directly, that given the legal questions
around software patents in general as reflected in recent higher court
decisions, and the record of review before the USPTO, and the fairness and
competence of the USPTO in issuing patents, that the court should put it all off
pending patent review?

[This is essentially saying that software patents don't have presumptive
validity. Many of these companies and lawyers are at times on both sides of
this issue.]

.

.

[ Reply to This | # ]

Claim term defintions in the tenative claim construction order
Authored by: dio gratia on Wednesday, May 11 2011 @ 08:33 PM EDT

The parties seem to largely accept the court's definitions, with the exception of the last term “symbolic (date/field) reference,” to which Oracle objects on grounds that the court's interpretation is limited to dynamic referencing, while Oracle believes the definition should include both static and dynamic referencing.
It's no surprise Oracle objected, this is the only definition of claim terms found contrary to their asserted definition. Judge Alsup's reasoning can be found in his tenative claim construction order ( page 20, line 22) -
The ’104 patent teaches two different types of data references: numeric references and symbolic references. The claimed invention includes an interpreter with two different sub- routines: “a static field reference routine for handling numeric references and a dynamic field reference routine for handling symbolic references”
is derived from the claims and description and appears to be on firm ground. Oracle's response ( Document page 2, line 20) -
Oracle remains concerned that the meaning of “dynamic” in the context of the '104 patent has not been fleshed out and may lead to a “construction of the construction” problem.
...
Oracle further suggests that deciding upon any particular gloss on “symbolic reference” is better done in the context of the infringement or validity issues, rather than in the abstract.
may well reflect shortcomings in patent prosecution wherein Oracle is effectively asking for the '104 patent to be expanded rather than potentially indefinite (Moreover,...). The expansion falling in Oracle's favor, of course. From an extrinsic perspective numeric also encloses enumerated. Recourse to expert testimony might demonstrate indefiniteness.

It seems likely both sides have already gamed the consequences of contrasting definitions, otherwise there would not have been points of contention. Oracle appears as the butcher who just can't resist putting a thumb on the scales out of habit.

[ Reply to This | # ]

The thing that is so frustrating with this
Authored by: Anonymous on Wednesday, May 11 2011 @ 09:26 PM EDT
... is that no one, not a single competent person, could possibly think that any
of the patents are valid. The patent office knew they were invalid, but they
don't care. They want the fees, and their position is that their granting of a
patent does not say in any way whether it is valid or not. Programmers and
lawyers at both Google and Oracle know that the patents are invalid. The judge
has to know this too.

I suppose a naive person might say "well, if the patent office grants it
then it MUST be valid even if the patent office says otherwise."

The lawyers want to go to trial because they make money that way, regardless of
the outcome. But why in the world would the officials at Oracle or Google want
this. Is it just a hussy fit, or are they really so stupid they think these
patents mean something?

[ Reply to This | # ]

Some Observations on Oracle v. Google, by Mark Webbink, Esq.
Authored by: Anonymous on Wednesday, May 11 2011 @ 11:43 PM EDT
The simple fact of the matter is this:

Java stole the ideas from UCSD -- which was also public domain. Go check. The
real figurehead behind it was Dr. Ken Bowles, professor emeritus, UCSD. He's
now 82+ years old.

UCSD development was financed by primarily thru the Navy. Sun was financed by
the CIA -- via Kleiner Perkins Caufield and Byers. KPCB financed Sun.
Netscape. (Golly, I wonder why Bill Gates got nailed.) Amazon. Google (with
Schmidt at the helm). Facebook. Many others.

FOOTNOTE: Eric Schmidt was a CIA operative back as early as 1985. KPCB
financed Sun (with Eric Schmidt as CTO.) Eric Schmidt was mentored by IBM's
Lesk. They wrote a book together and created the utility called
"lex." Lex is a lexical analysis tool. Why would the CIA want a
reknowned expert in Lexical Analysis to run Google? Easy: SPYWARE.

Why would the CIA want a "compile once, run everywhere" computer
programming language and "platform"? What is the Ultimate Trojan
Horse? ANSWER: One that everyone has been conned to install, voluntarily.
Whala. Instant Eschelon/Carnivore.

The ideas of the "Java Language" and "Virtual Machine" isn't
patentable. It was common practice by 1995. It was "stolen" by Sun.
Read the Java documentation. Even the Java documentation credits UCSD Pascal.

DOES ANYONE KNOW *** THE REAL REASON *** that Eric Schmidt stepped aside as the
CEO of Google? I do. His ass was exposed and Google Lawyers could not defend
Schmidt and Google at the same time. (Don't Forget-- Lawyer David Boies is
Oracle's main man and knows who the real parties of interest are. Larry Ellison
is hoping for a CIA/Govt handout to keep his mouth shut.)

Even the ideas in the Java Language (as patentable as the English Language) was
taken from PRIMARILY Nik Wirths Modula-2 (Pascal 2nd generation) with some
syntax borrowed from C.

[ Reply to This | # ]

Why is Google doing this?
Authored by: Anonymous on Thursday, May 12 2011 @ 01:48 AM EDT

Thanks for this valuable commentary.

Something I don't understand is why Google, having negotiated a deal that the people actually doing the negotiating found acceptable, then decided to kick it out at top-executive level and go for a legal fight.

OK, maybe under the deal, Oracle stockholders would have gotten a few cents more and Google stockholders a few cents less, than Google's CEO would like. But agreeing an out-of-court settlement is usually better than spending millions of dollars, and a huge amount of management attention, on a years-long mega-lawsuit.

Whoever wins this lawsuit: stockholders of both companies and the industry generally have already lost.

Lawyers and law firms will be the only big winners.

[ Reply to This | # ]

ROOTKITS...
Authored by: Anonymous on Thursday, May 12 2011 @ 02:23 PM EDT
More on the Trojan Aspect of Java. The "virtual machine" of the Java
"platform" is the key. To have spyware run "anywhere", you
have to have a common CPU. It doesn't matter if you are running Netscape,
Firefox, IE, Opera, or any other. It doesn't matter if you are running Windows,
Linux, Unix, AIX, or even DOS/VSE. It shouldn't matter if you are running
Intel, ARM, AMD, Motorola, or some Chinese chip that you cannot even pronounce
or even spell. (Programming language is irrelevant. It's the CPU that's
crucial.)

The most dangerous form of virus is the creation of a microcode rootkit.

A few years ago, the only way to inject malware into a CPU was to boot the
computer with a diskette and load it manually.

In April, 2008, Microsoft finally complied with USGovt demands to give the
ability to update microcode in a PC. Failure to apply the update negated your
ability to download patches.

Linux has "microcodectl" running as a process, today.

CONCLUSION: CPU Microcode can be updated with a software library, today;
unbeknownst to the computer user.

A Microcode Rootkit is the ultimate virus. NO Antivirus Software can detect a
virus within the CPU, itself.

Java's VM (really a CPU) either IS or CAN be readily changed into a virus that
monitors every single action on your computer with a simple update or maybe even
a secret API service call.

Wake up people. It's the "name brand" you trust. Java is a household
word, now.

That's marketing to clueless computer users who can't even understand how a
mouse or keyboard really interacts with hardware and the CPU bus.

RESULT: The ultimate CIA spy and CyberWarfare weapon.

[ Reply to This | # ]

Cerberus... Kerberos = SECURITY
Authored by: Anonymous on Thursday, May 12 2011 @ 04:11 PM EDT
Final comment from me... If you doubt what I have said, then think seriously
about your doubt. The entire "Unix" platform is embroiled in the
culture of the occult.

For you systems programmers out there and laypersons alike... REMINDER: we have
daemons (pronounced demons) and "unix" (eunuchs ... castrated
guardians of whores) and... even Kerberos...

Cerberus ( /&#712;s&#604;rb&#601;r&#601;s/),[1] or Kerberos,
(Greek form:
&#922;&#941;&#961;&#946;&#949;&#961;&#959;&#962;
, [&#712;kerberos])[2] in Greek and Roman mythology, is a multi-headed hound
(usually three-headed)[1][3][4] which guards the gates of The Underworld, to
prevent those who have crossed the river Styx from ever escaping. Cerberus
featured in many works of ancient Greek and Roman literature and in works of
both ancient and modern art and architecture, although, the depiction and
background surrounding Cerberus often differed across various works by different
authors of the era. The most notable difference is the number of its heads: Most
sources describe or depict three heads; others show it with two or even just
one; a smaller number of sources show a variable number, sometimes as many as
50.

[ Reply to This | # ]

Some Observations on Oracle v. Google, by Mark Webbink, Esq.
Authored by: TomWiles on Thursday, May 12 2011 @ 05:15 PM EDT
Circa 1980 Microsoft claimed joint ownership of any program compiled by a
Microsoft compiler. Any programs written for commercial distribution had to be
submitted to Microsoft for approval and apportionment of royalties.

When the DOD moved to Micros circa 1982 (Z100 and the Z248) these machines came
with full licensed versions of Microsoft's Basic, Fortran, and Cobol compilers.


The DOD bought these licenses to bypass the control that DOD lawyers felt
Microsoft had over any software written by DOD personnel.

Tom

[ Reply to This | # ]

Lawyers; or trial by combat?
Authored by: Anonymous on Friday, May 13 2011 @ 02:25 AM EDT
Anyone ever realize how lawyers make a "living" (if you can call it
that)?

1) They argue.
2) They argue about how to argue.
3) They argue about how to argue about arguing.

Can a society survive by argument alone?

No.

Everyone's answer to population control: Send everyone to law school and let
them be lawyers.

Reinstitute "trial by combat."

No arguments. No appeals. Cost of lawsuit = the cost of bodyarmor.

Put Larry Ellison up against Schmidt. Who'd win?

[ Reply to This | # ]

Some Observations on Oracle v. Google, by Mark Webbink, Esq.
Authored by: Anonymous on Tuesday, May 17 2011 @ 02:23 PM EDT
>>One other thing to mention with respect to the copyright
claims Oracle has asserted, some technical interpretations
imply that Google's Dalvik merely recompiles Java classes
into the Dalvik format. If that is true, then such
recompilations could be construed as a derivative work, but
let's not go too far down that path at this point.<<

As you point out, this is an aside, but it's an IMPORTANT
aside.

Because according to the Church-Turing thesis, EVERY
computer program written in a Turing-complete language can
(in theory anyway) be mechanically transformed to an
equivalent computer program written using ANY OTHER
programming language.

The result is that at some level, every Turing-complete
language (or Turing-complete language runtime) is a
derivative work of every other, by the argument you made
above.

Is liquid water the same as steam? If they're being used to
accomplish the same effect, I think the answer is a clear
"yes".

The same might be said of any two Turing-complete
programming languages or language runtimes.

But does it make sense to say this?


Dave Orme

[ Reply to This | # ]

Groklaw © Copyright 2003-2013 Pamela Jones.
All trademarks and copyrights on this page are owned by their respective owners.
Comments are owned by the individual posters.

PJ's articles are licensed under a Creative Commons License. ( Details )