|
Authored by: bugstomper on Friday, May 11 2012 @ 06:21 PM EDT |
Please summarize in the Title box error->correction or s/error/correction/ to
make it easy for readers to scan the list to see what errors have already been
reported, and for Mark and PJ to see what needs to be corrected.
[ Reply to This | # ]
|
- at 242 -> as 242 - Authored by: Anonymous on Friday, May 11 2012 @ 06:43 PM EDT
- rght -> right in "you have to rerun dexopt, rght? " -nt- - Authored by: Anonymous on Friday, May 11 2012 @ 07:33 PM EDT
- blidness -> blindness (n/t) - Authored by: Anonymous on Friday, May 11 2012 @ 08:13 PM EDT
- insturctions -> instructions (n/t) - Authored by: Anonymous on Friday, May 11 2012 @ 08:20 PM EDT
- sepcific -> specific (n/t) - Authored by: Anonymous on Friday, May 11 2012 @ 08:21 PM EDT
- Mr. Parr: TX-2672, what is this? -> Google: TX-2672... - Authored by: Anonymous on Saturday, May 12 2012 @ 12:49 AM EDT
- Terrence Parr -> Terence Parr - Authored by: Anonymous on Saturday, May 12 2012 @ 02:54 AM EDT
- Corrections Thread - Authored by: Anonymous on Saturday, May 12 2012 @ 06:36 AM EDT
- s/requies/requires/ - Authored by: hardmath on Saturday, May 12 2012 @ 09:40 AM EDT
|
Authored by: bugstomper on Friday, May 11 2012 @ 06:24 PM EDT |
Please stay off topic in these threads. Feel free to spice up your posts with
fancy formatting in HTML Formatted mode, and especially to make your links
delightfully clicky.[ Reply to This | # ]
|
- Bad patents in medicine/chemistry too. - Authored by: Anonymous on Friday, May 11 2012 @ 08:14 PM EDT
- IP Czar: Voluntary Industry Agreements Could Be Key to Combatting IP Infringement - Authored by: Anonymous on Saturday, May 12 2012 @ 01:03 AM EDT
- Under the radar - Authored by: Anonymous on Saturday, May 12 2012 @ 02:59 AM EDT
- Hands On With the World's Simplest Android Phone - Authored by: Anonymous on Saturday, May 12 2012 @ 03:45 AM EDT
- Patenting mathematics - Authored by: Anonymous on Saturday, May 12 2012 @ 12:45 PM EDT
- Holy moly, Mueller was actually right about something. - Authored by: Anonymous on Saturday, May 12 2012 @ 02:58 PM EDT
- Microsoft back to major OS anti-trust violation. - Authored by: Anonymous on Saturday, May 12 2012 @ 04:50 PM EDT
- It costs Google over $300/employee to upgrade Ubuntu - Authored by: Anonymous on Saturday, May 12 2012 @ 09:01 PM EDT
- Alison Courses - Introduction to Copyright Law in America - Authored by: Davo.Sydney on Saturday, May 12 2012 @ 09:11 PM EDT
- Why doesn't Joshua Bloch license RangeCheck to Google? - Authored by: symbolset on Sunday, May 13 2012 @ 12:18 AM EDT
- We can build the first generation of the USS Enterprise – so let’s do it ! - Authored by: Anonymous on Sunday, May 13 2012 @ 01:59 AM EDT
|
Authored by: bugstomper on Friday, May 11 2012 @ 06:27 PM EDT |
Please type the title of the News Picks in the Title box of your comment to make
it easy for readers to scan the headings. Also please include the link to the
article in the body of your comment, preferably as a clickable link posted in
HTML Formatted mode, for everyone's convenience once the article has scrolled
off the News Picks sidebar.[ Reply to This | # ]
|
|
Authored by: Anonymous on Friday, May 11 2012 @ 06:41 PM EDT |
lets say they have a 90% chance of getting $150,000 (though I would expect more
like a 90% chance of getting $50,000). Compare that with a 1% chance of $100
million.
Doing the math,
150,000 * 0.90 = 135,000
100,000,000 * 0.01 = 1,000,000
even a 1% chance of 15 million is still better.
Don't forget, having a 3rd phase increases the billable hours for the lawyers,
so they win no matter what.[ Reply to This | # ]
|
|
Authored by: Anonymous on Friday, May 11 2012 @ 06:43 PM EDT |
Maybe it would have made more sense having the evidence to see at the same time,
but reading that really put me to sleep.
At least Terence Parr's testimony was somewhat interesting. Arrays are probably
much simpler to explain than symbolic referencing though.[ Reply to This | # ]
|
- Andrew McFadden testimony - zzzzzzzzz - Authored by: Anonymous on Friday, May 11 2012 @ 06:55 PM EDT
- Mr. McFadden: Hopefully, 242. [He smiles, but no one laughs] - Authored by: YurtGuppy on Friday, May 11 2012 @ 07:25 PM EDT
- Andrew McFadden testimony - zzzzzzzzz - Authored by: Anonymous on Friday, May 11 2012 @ 07:26 PM EDT
- It woke me up! He devastated Oracle's case. - Authored by: bugstomper on Friday, May 11 2012 @ 07:26 PM EDT
- I agree, but... - Authored by: Anonymous on Friday, May 11 2012 @ 08:30 PM EDT
- I agree, but... - Authored by: Anonymous on Saturday, May 12 2012 @ 01:17 AM EDT
- But David August - excellent - Authored by: Anonymous on Saturday, May 12 2012 @ 12:46 AM EDT
- The "definition" in another court document ruins this defence - Authored by: Anonymous on Saturday, May 12 2012 @ 10:37 PM EDT
|
Authored by: lnuss on Friday, May 11 2012 @ 06:53 PM EDT |
Nice reporting -- thanks, well done!
---
Larry N.[ Reply to This | # ]
|
|
Authored by: Anonymous on Friday, May 11 2012 @ 07:02 PM EDT |
Christ how is the jury expected to make any sense of all this? I'm an Android
developer with limited knowledge of the dex format and its optimizer and I can
barely make heads or tails out of the level of detail we've dived into here. Why
on earth did the parties not seek a bench trial for this phase? Can the jury
just refuse to return a verdict on the grounds of incomprehensibility? [ Reply to This | # ]
|
|
Authored by: SilverWave on Friday, May 11 2012 @ 07:05 PM EDT |
Too good.
:-)
---
RMS: The 4 Freedoms
0 run the program for any purpose
1 study the source code and change it
2 make copies and distribute them
3 publish modified versions
[ Reply to This | # ]
|
|
Authored by: Anonymous on Friday, May 11 2012 @ 07:05 PM EDT |
Thanks for this detailed report. The mainstream media's coverage of the patent
testimony has been all but useless, but thanks to these details we might get a
sense of what is actually going on in that courtroom!
Oracle:
So the role of the iget instruction is to obtain actual field data from an
object and store it in a Dalvik register, true, sir?
Mr. McFadden:
Yes.
Oracle: iget with 01 as the field index, it doesn't store the number 01
in a Dalvik register, right?
Mr. McFadden: Right.
Oracle: It doesn't
obtain 2 or 76 and store those in a Dalvik register, does it?
It doesn't
obtain the name "byte", does it?
[five or six more questions along these
lines, getting more and more excited]
Mr. McFadden: Yes.
Oracle: The
actual data is what it stores, right?
Mr. McFadden: Yes.
Oracle: True or
false: the Dalvik iget instruction never contains the actual memory location of
the actual field it's supposed to get, true?
Mr. McFadden: True.
Oracle:
True or false: the va operand is not the memory location of the actual
field?
Mr. McFadden: True.
Oracle: The vb and field@CCCC are not the
memory location of the actual field?
Mr. McFadden: True.
[By this point
the Oracle lawyer looks like he's just triumphantly unveiled the murder weapon.
I have no idea what his questions were supposed to mean, though, and the witness
sounded like he didn't either. I didn't see the jury's
response.]
From the cross exam questions, Oracle's lawyers seem
to believe that any indirect reference to the field value is a "symbolic
reference" (or maybe they just want to get a jury to believe
it.)
Unfortunately for Oracle, it isn't true. A "symbolic reference" is
when you have a symbol (a string) that refers to something, and
"resolving the reference" means looking up that symbol in some kind of
table. (in a compiler, it would be called "the symbol table", for obvious
reasons.) A numeric indirection (an index into a different table, or into a
data segment) is NOT a symbolic reference. Based on these notes, it sounds very
much like if the claim requires a "symbolic reference", Dalvik won't infringe.
Oracle's poor lawyers, having to wade through all this technobabble. My heart
bleeds for them. /sadface
I skimmed through the '520 patent yesterday,
and it just describes a clinit optimization that Java compilers do so the VM can
initialize classes faster. I remember this from when I worked on a Java VM ~10
years ago; its sort of a kludge to permit some static initialization of class
stuff (instead of having to do it all at runtime). In a sane world this
shouldn't be patentable, being (1) entirely software/math and (2) not
particularly innovative, or hard to come up with from scratch. Anyway.. why
does it use such a kludgy mechanism? Well, Wikipedia says that JDK 1.0 came out
in 1996, and this patent wasn't filed until 1998 -- meaning Sun already had
working and deployed VMs that were doing something even stupider than this in
order to initialize their classes. When they designed this "optimization", it
had to fit into their existing system, and not break backward compatibility with
existing compiled .class files.
But if you were making a new system from
scratch, you would likely try to find a non-kludgy solution to this problem. So
I'd say its not surprising that Dalvik doesn't use the same mechanism, because
the one from the '520 patent is not particularly great. If you weren't
constrained to follow Java's .class format, you could do something simpler and
better, which it sounds like Dalvik did. [ Reply to This | # ]
|
|
Authored by: Anonymous on Friday, May 11 2012 @ 07:08 PM EDT |
Question 3b and the jury:
"[Has Oracle proven that Google’s conceded use of the
following was infringing, the only issue being whether such
use was de minimis:]
a. yes, infringing. [The rangeCheck method in TimSort.java
and ComparableTimSort.Java]
b. no, not infringing. [Source code in seven “Impl.java”
files and the one “ACL” file]"
And in the order above: "Here, Google has admitted to
copying the entire files. No reasonable jury could find that
this copying was de minimis."
And yet, the jury answered No to 3b. Would the judge argue
that the jury was unreasonable? Presumably not, or else the
jury is not competent on other questions. How then did the
jury reach this decision if not through de miniumus, given
the phrasing of the question? Which is to say that a
reasonable jury COULD find de-minimus, and that the judge
has erred.
Sounds like grounds for an appeal to me. The jury is
reasonalble, or all verdicts reached by the jury are
suspect.
[ Reply to This | # ]
|
|
Authored by: Anonymous on Friday, May 11 2012 @ 07:14 PM EDT |
> No reasonable jury could find that the copying of entire computer files was
de minimis.
I disagree. They were a few files out of thousands and not a recognizable
portion of the whole work. Also, I think they were put there by a third party,
though I do not recall the exact details.
In short, I'm not sure why he's considers his jury unreasonable.[ Reply to This | # ]
|
|
Authored by: Anonymous on Friday, May 11 2012 @ 07:34 PM EDT |
"We're saying once you have proved infringement, we think we have a claim
for infringer's profits, and the burden of proof is on them. The question is not
how much we're entitled to, it's who has the burden." - David Boies
That is SO like BSF in the SCO case. Always trying to escape their own burden
of proof. "They did something! They can't prove that they didn't!"
This act is getting old.
MSS2[ Reply to This | # ]
|
|
Authored by: Anonymous on Friday, May 11 2012 @ 07:35 PM EDT |
I know how hard it is to teach a basic understanding of the concepts involved in
programming. I know the kind of cray toe curling misconceptions students are
capable of coming up with. And these are people who have an interest and have
decided to study the subject.
I'm now trying to imagine taking a bunch of completely random people off the
street who probably have no interest in or aptitude for the subject and getting
them to understand the difference between static and dynamic linking.
I'm trying to imagine teaching them about this under the handicap of not being
allod the freedom to explain properly but being constrained to only answer
yes/no questions.
I'm trying to imagine doing it while not being able to ask questions to test
their understanding, or have them ask me questions about the things they don't
understand.
They get no assignments to make them think. They don't get assessed or tested to
see if they are learned.
I can't imagine an educationally worse process. I shudder to think what is most
likely going on in their brains at the end of it all. And millions of dollars
hang off this? [ Reply to This | # ]
|
|
Authored by: dcs on Friday, May 11 2012 @ 08:03 PM EDT |
Oracle tries to show that, because dexopt happens when the
handset is turned on ("running", in Oracle's parlay), what
it does is "dynamic".
However, by Oracle's definition of "runtime", nothing is
static, since programs only run when the computer is
running.
"Runtime" means "when the program is executing".
"Dynamic"
is what happens to the program at runtime -- when it is
executing. "Static" is what happens to the program before
runtime -- before it gets executed.
This is the definition of static and dynamic, and Oracle is
portraying them as something else, something that doesn't
even make sense.
---
Daniel C. Sobral
[ Reply to This | # ]
|
|
Authored by: SilverWave on Friday, May 11 2012 @ 08:33 PM EDT |
Yep Devastating.
---
RMS: The 4 Freedoms
0 run the program for any purpose
1 study the source code and change it
2 make copies and distribute them
3 publish modified versions
[ Reply to This | # ]
|
|
Authored by: rsteinmetz70112 on Friday, May 11 2012 @ 09:22 PM EDT |
I have recently gotten the impression that the Judge is much more than an
impartial referee in this case. He may have an agenda of his own.
It is difficult to tell how good Oracles patent case was, but Google's seems
very persuasive. At best it's a he said he said case, and the jury has already
mostly sided with Google on copyright.
The jury has demonstrated they are not convinced by Oracles copyright evidence.
They couldn't reach a verdict on fair use but at least some on the jury felt
Google's actions were fair use. That does not bode well for damages.
Is it possible he is setting Oracle up to win a small award in the damages phase
followed by his ruling against Oracle on SSO. Setting up a no win appeal
Oracle?
If Oracle were to win on the patent appeal and get an injunction exactly what
would they be enjoined against? Who would decide if Google had coded around
these patents?
Seems more and more like the Fat Lady is a long way off.
---
Rsteinmetz - IANAL therefore my opinions are illegal.
"I could be wrong now, but I don't think so."
Randy Newman - The Title Theme from Monk
[ Reply to This | # ]
|
|
Authored by: Anonymous on Friday, May 11 2012 @ 09:25 PM EDT |
Now I understand. McFadden is describing a stored program machine,
calculated Go To.
Wait, I thought they banned that because it is risky.
When was it invented? 1940 something?
[ Reply to This | # ]
|
|
Authored by: Anonymous on Friday, May 11 2012 @ 09:47 PM EDT |
Mappers vs. Packers,
redux:
Judge: Is there a reason the numbering of the
opcodes skips 1?
Mr. Parr: Good question. It's basically the address, and
some instructions take more space than the others. For example, the "dup"
instruction takes one byte, so it's at address 3 and the next is at address 4.
Imagine going down the street, some people have a big fancy house and it takes
up a lot of space, some people have small houses. The more information you have
to pack into the instruction, the more space it takes.
Judge: I still don't
understand why there's no address 1.
Mr. Parr: The iconst_4 instruction
occupies both address 0 and address 1.
Judge: Ah, okay.
Dr.
Parr is a programmer, and used to "Mapper" style thinking. He gives an
explanation that "some instructions take more space than others", but it
probably doesn't even occur to him to state explicitly that "this instruction
takes two bytes", because to him it is an unnecessary detail, that should be
obvious anyway from the context.
Judge Alsup on the other hand, is trained
to carefully analyze facts and draw inferences, but to not assume
anything. He's not a domain expert and if he just went ahead assuming things,
he would probably get them wrong. So he asks a clarifying question. To Dr.
Parr's credit, he immediately understands what detail he skipped over and states
it clearly. I guess that's because he's also a professor: he's good at noticing
which concepts the student already understands and linking new things to those.
And he also gives feedback to his student ("good question").
I
love reading exchanges like this, because it shows how intent Judge Alsup is on
understanding the tech as well as he can in such a short amount of time (or at
least well enough to be able to make legally-sound rulings about it). I could
follow along with McFadden and Parr's testimony easily enough, because I've been
programming for over 20 years and know a lot about compilers and VMs. To a
non-programmer this stuff must be nearly bewildering.
I was thinking about
this earlier today and I realized that from a certain point of view, Judge Alsup
has one of the most interesting jobs in the world: most of his cases are
probably relatively mundane, but occasionally he gets something crazy like this
one, and then he gets real subject-matter experts coming into his courtroom and
giving him a crash course in some complex field like programming or biology or
whatever. So he gets this little "peek behind the curtain", and then when the
case is done he can move on to something else. :P
P.S. Big thanks to
our courtroom reporters! You guys and gals are doing a great service for the
rest of us. [ Reply to This | # ]
|
|
Authored by: Anonymous on Friday, May 11 2012 @ 10:02 PM EDT |
I wonder if anyone there at the trial who wasn't a computer scientist could even
begin to know what was being talked about in regards to the patents. Without a
background in computer science or significant programming experience, I don't
see how anyone could see the testimonies as anything other than just experts
disagreeing with each other, with gobbledygook spicing up the discussion.
The main contention has been Oracle's definition of the words used in the
patents. I don't see how the jury will be able to come to an informed decision,
if the exact definition of the words are not agreed upon. This makes me wonder,
if a patent can be challenged at the patent office to clarify terminologies
being used. Forcing them to add specific definition of words to the patent to
increase specificity.[ Reply to This | # ]
|
|
Authored by: Rubberman on Friday, May 11 2012 @ 10:02 PM EDT |
I have 30 years experience as a software engineer, with a
major software patent to my name (as exclusive inventor)
that covers software that runs most of the semiconductor
fabs in the world today. The testimonies today were just
fabulous! As I have said in other postings during this
trial, this is the best entertainment I've had in a long
time! Thanks PJ for the efforts you and your
reporters/observers have been putting in at this trial! It
is an incredibly important one, and I think that if Oracle
took just a few seconds to really think about the possible
repercussions that a finding in their favor would mean,
they'd drop their case/claims as quickly as they could![ Reply to This | # ]
|
|
Authored by: feldegast on Friday, May 11 2012 @ 10:03 PM EDT |
https://twitter.com/#!/Feldegast
Raw feed: https://twitter.
com/#!/Feldegast/oracal-vs-
google-trial--- IANAL
My posts are ©2004-2012 and released under the Creative Commons License
Attribution-Noncommercial 2.0
P.J. has permission for commercial use. [ Reply to This | # ]
|
|
Authored by: Anonymous on Friday, May 11 2012 @ 11:08 PM EDT |
Regardless of which legal team wins the patent phase of this trial, everyone
else loses. That includes all the technical people at both Oracle and Google
too, they both lose bigtime regardless of the verdict, because this very high
profile trial is validating the concept of software patents in the US and
killing the industry of which they are a part.
You might as well slit the throats of all software developers if this continues,
at least those in the US. It's no different to all chefs becoming liable to
culinary patent infringement every time they mix ingredients in one combination
or another, or all writers becoming liable for grammatical patent infringement
each time they use a particular combination of words. It's just madness.
It has been said a million times already that the only beneficiaries of this
madness are the lawyers, and that is certainly true. But why aren't all the
computing megacorps fighting tooth and nail against software patents, with the
possible exception of Oracle on grounds of greed-driven insanity?
I really don't get this situation. The software industry here seems to have a
deathwish.
Why?[ Reply to This | # ]
|
|
Authored by: xtifr on Friday, May 11 2012 @ 11:58 PM EDT |
A lot of people reading this seem fairly confused by some of the technical talk.
Let me remind everyone that there were diagrams and charts available to the
judge and jury that we don't get to see. As an experienced programmer, I was
able to figure out approximately what must have been on the charts in most
cases, just from hearing what was said. If the charts were like what I imagine,
then I don't think the judge and jury were as confused as many readers here
are.
I can't be sure, because I already understand the topics being discussed, so I
may be overestimating how clear the charts would make things. But the fact that
the judge was able to ask some very insightful questions makes me fairly
confident that the material was actually reasonably clear.
---
Do not meddle in the affairs of Wizards, for it makes them soggy and hard to
light.[ Reply to This | # ]
|
|
Authored by: Anonymous on Saturday, May 12 2012 @ 12:17 AM EDT |
I find it amusing that the lawyers aren't supposed to lead the witnesses, but
there's a sort of "wrapper" they can apply to make it all
alright.
Google: Tell me that we're pure and innocent as the
driven snow
Oracle: Objection!
Judge: You have to say "Simon
says"
Google: Simon says tell me that we're pure and innocent as
the driven snow
Witness: But of course!
[ Reply to This | # ]
|
|
Authored by: Anonymous on Saturday, May 12 2012 @ 12:47 AM EDT |
"No reasonable jury could find that this copying was de minimis."
Balderdash. The judge is making up law. Of course a "reasonable"
jury could find these few files to be de-minimus. If anything, no reasonable
jury could find to the contrary.
You are wrong PJ, there is no evidence at all that this judge is anything but a
word-game playing moron just like all other judges and lawyers.[ Reply to This | # ]
|
|
Authored by: Anonymous on Saturday, May 12 2012 @ 02:15 AM EDT |
Let me see how much time we used... [calculates briefly, then stops]
I think you both must be keeping track. If you need me to figure it out, it'll
take five minutes.
Google: Our legal assistants have been keeping track. I
think we can do it.
Judge: Great. See you on Monday.
I'm
not sure I understand this bit, so maybe someone can correct me if I'm way
off.
It sounds to me like the Judge was asking them "do you need the real
number, which it will take me 5 minutes or so to calculate? Or are you OK just
relying on your own estimates for whatever planning you need to do between now
and Monday?" and the lawyers answered "we're fine going by our own
estimates".
Maybe this means the paralegals exchange counts, and unless they
are wildly different or something, each side just assumes the real numbers are
in that ballpark. (Just like baseball scorekeepers at community baseball games
-- there's one from each team, usually the spouse of one of the players, and
they compare scores every inning or two, to make sure there's no significant
discrepancies).
Or maybe they don't share with each other, and each team
just relies on their own estimate, until the Judge "re-syncs" them by giving
them his own counts.
...I feel almost guilty that I find a detail like this
interesting, when there's so much important stuff going on in the case itself!
But I admit that from a comfortable outsider's viewpoint, watching this whole
process occur is fascinating, and not just for the usual "train-wreck in slow
motion" reasons (but that too). [ Reply to This | # ]
|
|
Authored by: Anonymous on Saturday, May 12 2012 @ 02:57 AM EDT |
I-Programmer Reads
the
Patents
Comment on RE38104: "This is a patent that bears
the
name of James Gosling, the father of Java, so I would
rather not criticize it
too much - but I have no choice."
[ Reply to This | # ]
|
|
Authored by: Anonymous on Saturday, May 12 2012 @ 04:58 AM EDT |
I like this bit:
Judge {to Oracle}: ...I'll explain to the jury
that you've got no evidence...
Nothing new for BS&F
then...
...If you can prove real damages on that, you're a great
lawyer. [Smiles at Boies]
Lawyer's not the word I would use (but
it does start and end with the same letters)...[ Reply to This | # ]
|
- why is it... - Authored by: BJ on Saturday, May 12 2012 @ 03:15 PM EDT
|
Authored by: Ian Al on Saturday, May 12 2012 @ 05:07 AM EDT |
PJ wrote (and I selectively read in case it does not support my
prejudice):...test files don't normally ship, and users certainly
don't use them... Oracle has opted to go for actual damages, not statutory,
despite the judge's warning that they are making a mistake... so there will be a
phase three.
Then, there was what the judge ordered (I'm not daft
enough to chop that up!):Dr. Cockburn has not adequately valued that
[allegedly copied] code in his report and cannot do so at trial. This order
holds that the jury will be instructed that if Google is found not liable for
infringing the selection, arrangement, and structure of the API packages, then
Dr. Cockburn’s copyright damages analysis is
inapplicable.
Analysing the term 'actual damages' using the
English language (I know, I know! What has that got to do with legal language!)
that could mean damages actually incurred by Oracle because of Google's
copyright infringement.
I look first at the directly copied test files
put into the Android current version repository by the European code company and
against the terms of their contract with Google. Google only found that out when
Oracle pointed it out a couple of year's ago. Google took it out of the current
repository a year ago. Google would not have asked developers to use test files
that they did not know had shipped. As PJ says, they never got to the production
phones.
So the files were on countless (as in, we have no way of
knowing how many) developers' machines for three years and were never used. How
much actual damage was done to Oracle for files that were not known about by
developers or Google and were never used?
How about rangeCheck?
Supposing Google had to get a high school coder to write one from scratch? How
much would that have cost Google? Say, $150? What about getting whiz coder
Josh Bloch to do it? $1,500? How much would Google have been prepared to pay
Oracle for those 9 lines? What about the ninth line that just had '}' in
it?
Perhaps I am looking in the wrong direction. Perhaps it was all
about the value to Google of using the code. What does it do? If a programmer
makes a total Cockburn of his program [his, because girlie programmers have more
sense] getting the range of an array wrong and it is picked up by rangeCheck,
would the incompetent programmer have thought to program a suitable response to
the returned error message? Probably not. So, what would the problem be of
leaving it out, altogether. Nothing in those circumstances. Supposing he got the
range wrong, but programmed in a recovery in response to the error message? Who
knows?
But wait, what did Mitchell say was important? He said that the
rangeCheck function was used 2600 times (whatever, I can't find the quote) in
the Dalvik startup phase. If rangeCheck had found an error in the startup phase
it would have been a major failure in the Dalvik platform code and not the apps.
Even if rangeCheck had not been used during Dalvik coding, the error had to be
sorted before Android ever shipped to developers, let alone
'phones.
So, the use highlighted by Mitchell was examples of code
wasting time in the Dalvik startup phase because any such errors had to be
resolved during Dalvik development. How much actual damage was done to Oracle,
because their code was substantially slowing down Dalvik startup? Perhaps it was
putting the Java ecosystem at risk of ridicule? No, that cannot be. Oracle
explain to us that Android is fragmenting the ecosystem. In fact, they demand
that the ecosystem is fragmented even more by making Java APIs unavailable to
Google.
For completeness, lets consider what Mitchell studiously
avoids; the apps. The whole point of the APIs and the patents is the apps
written in Java language. If an app has a range check error, it is likely to
crash. What is the actual damage to Oracle if an Android app
crashes?
The judge will tell the jury that Dr. Cockburn’s copyright
damages analysis is inapplicable. I'm assuming that he will have ruled on the
SSO by the time of the damages trial phase and that what he has signalled will
come to pass. So, Cockburn's damages report is out and Dr. Kearl's report is
based on the whole of Java, or at least, the whole of the value of the
APIs.
So, the lucky jury get to go to the damages phase three. What
actual damage to Oracle will they find for Google using nine lines of code that
does nothing in Android or development and test files that, just do nothing but
sit there, unseen until Larry wants money? You know, I think I'm beginning to
see Judge Alsup's point! I am not unhappy that Jacobs does not. Brain the size
of the planet, and he has to spout so much Oracle drivel. I hope he gets better,
soon. This is not good for him.
--- Regards
Ian Al
Software Patents: It's the disclosed functions in the patent, stupid! [ Reply to This | # ]
|
|
Authored by: Anonymous on Saturday, May 12 2012 @ 06:19 AM EDT |
Thank you for this wonderful and crystal clear explanation - it should have been
placed as a new topic instead of being buried in a thread :-)
Hopefully the Google folks will find and read it anyway!
Bernd[ Reply to This | # ]
|
- 100% ON topic IMHO - Authored by: Anonymous on Saturday, May 12 2012 @ 06:37 AM EDT
- Thanks - Authored by: Anonymous on Saturday, May 12 2012 @ 10:51 AM EDT
|
Authored by: SLi on Saturday, May 12 2012 @ 06:22 AM EDT |
Have I understood correctly that Google agreed to not contest
the validity
of the patents? So, they have even waived the
argument that they, as software
patents, are invalid under Bilski, because they are
directed to an
abstract idea?
Why would Google do this? [ Reply to This | # ]
|
|
Authored by: Anonymous on Saturday, May 12 2012 @ 06:53 AM EDT |
that the Judge susses out that all these patents are basically algorithms and
thus the patents themselves are not valid as the subject matter is not
patentable being merely mathematics...[ Reply to This | # ]
|
|
Authored by: Anonymous on Saturday, May 12 2012 @ 10:11 AM EDT |
This reminds me of my 6502 days.
Swapping zero page memory
/Arthur [ Reply to This | # ]
|
|
Authored by: Anonymous on Saturday, May 12 2012 @ 11:42 AM EDT |
*cough*taking a leak*cough* a break?
As a more tech person i'm not calling for some fancy foo/software, but please,
that problem is really solved since ages.
No matter if a real chess clock can be stopped or not, two stop watches or what
ever will do.
Or maybe yes, a special clock, but which should be a requisite in every
courtroom, like the hammer.[ Reply to This | # ]
|
|
Authored by: webster on Saturday, May 12 2012 @ 12:44 PM EDT |
1. It must be depressing to the Oracle team to realize that the
jury is going the extra mile for Google. Even on the admitted copying the jury
found for Google by stretching with the "de minimis"
concept. Oracle's burden of proof is high for this jury. There is at least one
juror that will hang for Google. Oracle has also annoyed the judge more than
Google has. [Does anyone have a score of tantrums with a vehemence scale and
presence of jury breakdown?]
2. Nevertheless, the judge has taken the
sacrosanct jury verdict away from Google on the admitted
copying point. This is risky business for the judge. Although that peculiar
judge is gone from the Federal Circuit Court of Appeals, trifling with jury
verdicts pricks all their ears. Remember Kimball and SCO I.
3. Why is the
jury trial so sacred? It is a safeguard against a totalitarian state. It takes
the decision away from the branches of the government and puts it directly on
citizens. As counsel is wont to say to a jury:
___________"It is
a decision so important that it is not left to a president, governor or mayor,
or a majority of a house, senate or city council. It is a decision so important
that it is not determined by a plurality, majority, or super-majority. It has
to be unanimous!"________
So with this love for the jury, the
commons has to put up with some irrationality. Or as is repeated here every few
months, with sympathy over reason. The jury will try and do
what they want to be fair despite what the facts, law and judge might indicate.
Cf. State v Simpson, 1995.
4. Of course the jury does not know the law so
they may produce some unintended consequences. If they decide to throw Oracle
a patent bone, they may cause far more damage than they
understand. However, at least one juror hung them for "fair use" despite
finding API infringement. A reasonable patent defense should produce at least a
hung jury. The jury may have hung for infringement, 1a, if they didn't have the
1b alternative.
5. This day continues to illustrate the judicial absurdity
of trying patents before laymen. It is experts at dawn with an
array of seasoned programmers opining on software code. They may as well be
opining on complex math formulas for all the jury or any lay person can
understand. The Judge hired Dr. Kearl on damages. Why doesn't the judge insist
on a neutral patent expert or panel to present to the jury instead of the de
riguer contradictory dueling party banjos? The jury should say, "If biased,
knowledgeable experts won't agree, why and how should we? We need an unbiased
analysis and opinion."
6. Just being in court arguing about software
patents in concepts and terms the jury can't possible understand gives the
patents a dignity they don't deserve. The party is not allowed
to criticize the Patent Office. The patents are presumed valid. The jury has
been properly screened so they should be unbiased, and definitely ignorant,
about software patents. They can be incredibly destructive here even in trying
to help Google.
Caution to all observers and
Disclaimer. Most that read, report and comment on Groklaw have a bias
against software patents. They are not limited to just what the jury hears, and
the jury's deliberations. The jury has an entirely unique perspective. The
more experience one has with trials, the less one can predict because so much
can and has happened. In so far as the above notes a sympathy for one party, it
must not be considered as a prediction. So much for the small print.
~webster~
[ Reply to This | # ]
|
|
Authored by: Anonymous on Saturday, May 12 2012 @ 01:14 PM EDT |
That's the tragedy of the USPTO, that they seem to have no
engineers who are able to spot a goofy patent application when they see
one.
I think the tragedy may lie deeper then quite potentially the
patent examiners.
Not being part of the USPTO, I can only speak to what
occurs outside the USPTO. And what occurs outside the USPTO is - quite
simply:
The Legal Profession keeps telling the engineers that they are not
qualified to interpret a patent's claims!
We keep getting that drilled into
us over and over again. That makes absolutely no sense to us because:
A:
We're the ones who are supposed to be responsible for the invention in the first
place that the patent was filed against.
B: We're the ones who are
supposed to be able to build the invention out of reading the patent
filing.
C: We're the ones who are held responsible for trebble damages
if we are aware of the patent.
Taking C into context.... I'm seriously not
qualified to understand the text that I read but I'm qualified enough to face
trebble damages?
Let's extrapolate that to - maybe - what might be
happening at the USPTO.
Patent examiner reviews the patent filing with hands
tied. Remember, they have limitations placed on them such as not being able to
reference the internet.
Patent examiner denies patent filing on various
grounds including obviousness.
Path 1: Lawyer obfuscates wording of
patent till examiner no longer recognizes it and stamps valid - by default
(remember, I'm hypothesizing) if the examiner can't understand the invention,
the invention must be really, really advanced.
Path 2: The Lawyer
appeals to the USPTO appeals board - quite possibly populated with members of
the Legal profession with no Engineers. The appeals board overrules the patent
examiner.
So... here's the magic question:
If the patent examiner was
an advanced engineer would s/he still be constrained on what s/he could
effectively do the same way the rest of us are? By the Legal Profession telling
the individual they are not qualified to interpret/understand the patent
claims?
Perhaps the Tragedy lies much, much deeper then the USPTO lacking
engineers.
Perhaps the real tragedy lies in the Thomas Jefferson's no
longer being the gate keepers of what is allowed to be patented.
RAS[ Reply to This | # ]
|
|
Authored by: Anonymous on Saturday, May 12 2012 @ 07:02 PM EDT |
Clearly, te difference between symbolic and non-symbolic references Is
when te symbolic references, like employee_ssn translated into an
address.
I don't think I have seen a clear explanation of when this appens in the two
systems. A real missed opportunity when the judge asked te witness to
illustrate te differene between his take and Dr. Mitchel' (for Oracle).[ Reply to This | # ]
|
|
Authored by: Anonymous on Saturday, May 12 2012 @ 11:20 PM EDT |
So, I'm not sure if I can really do this topic justice, but I wanted to point
out that the availability of infringer's profits in a copyright action is quite
reasonable. To keep things simple, I'm not going to go anywhere near the facts
of this case, though.
The basic idea is difficult to disagree with: that bad actors should not be
allowed to profit from their bad actions, whether or not their profits are
directly at the expense of the victim of those bad actions. This is the
morality (equity) underlying the remedy of unjust enrichment, part of the law of
restitution.
A general example (not copyright related) would be if I owe you a duty (of
confidence, say) and I breach this confidence (sharing the confidential
information) because it is profitable for me to do so. Even if you would not
have been able to profit from sharing this information (if you intended to and
could have done so, then you would be seeking lost profits), since I breached a
duty and made a profit, I am made to pay over profits flowing from the breach in
restitution of my unjust enrichment (not necessarily at your financial
expense).
Here's a slightly more detailed example of why this kind of remedy is available
for copyright infringement:
Let's say that I have written and published a book in English, I have no
intention of translating this book into Spanish or allowing someone else to do
so. My profit margin per copy sold is $5. You come along and decide that the
book would sell well in Spanish, so you translate and publish my book without
getting my permission, and you sell a millions copies. Your profit margin is $7
per copy. I sue you for copyright infringement, seeking actual damages and
disgorgement of profits. You would be able to argue that my actual provable
damages are $0, because I was not offering a Spanish translation, and so I have
not lost any sales of my English book - for the sake of simplicity, we can
presume that no one who bought the book in Spanish would have bought the book in
English - so we are selling our books in entirely different markets.
You have profited millions of dollars from the sale of your translation of my
book (much more than what's available in statutory damages), you have infringed
my copyright, and yet you have caused me no actual damages. If I was unable to
also claim disgorgement of your profits, you would (happily) pay me the small
amount of statutory damages and keep your millions in profits. In order to
prevent copyright infringement in cases like this (and I'm sure they would be
common), the law allows me to also claim your profits.
A further example, dealing with the issue of double recovery, where we are
selling in the same market:
Let's say that now you don't translate my book into Spanish, but simply sell
your own unauthorized edition of my book. Your profit margin is still $7, mine
is still $5, and you still sell a million copies. I would be able to argue that
every copy you sold was one that I didn't, so at my $5 profit margin I could
prove $5 million in actual damages. If I was not allowed to also claim
disgorgement of your profits, you'd still be able to make a $2 million profit by
infringing. Here is where the bar on double recovery comes into play: you have
made a $7 million profit, and I have proven $5 million in damages. If I was to
get all of your profit plus my losses that would be $12 million, which would
also be an unjust and excessive recovery. So instead I only get that portion of
your profits that were not included in my losses, and so I get $7 million in
total.
Hopefully this now all makes (some) sense.
Obviously the case at hand is a bit more complicated in that Oracle is going to
have to show that some of Google's profits flow from the use of those nine lines
of code. I believe that this is the part the the judge is not going to allow
Oracle to introduce evidence on, since it was left out of their first two
attempts at a damages report.
I am surprised that the jury will be determining the amount, though. My
understanding was that even though it's in the statute, it's still an equitable
remedy (as all restitution is, IIRC) and so should be determined by the judge.
I am by no means an expert in US law, however, so maybe the fact that it's in
the statute somehow transforms it to a legal remedy, or maybe equitable
statutory remedies are their own kind of thing.[ Reply to This | # ]
|
|
Authored by: BitOBear on Monday, May 14 2012 @ 06:48 AM EDT |
The jury was ordered to consider the entire SSO of the entire API (e.g. the
"166 APIs") where "each API" had multiple files.
Seven files out of thousands is de minimus (spelling) by definition. The jury
was ordered to consider the "wholeness of the whole" and the judge
just said that these tiny parts couldn't de minimus because of their
"wholeness of the parts". [sarcasm quotes mine.]
Mandatory car analogy: "vandal" takes the four valve stem caps off the
tires of a car. Jury is asked "did vandal steal a significant fraction of
the car". Jury says no. Judge says jury is wrong, the car was largely
stolen and no reasonable jury could find that the "theft" was minimal.
Judge is wrong.
====
Next, the act of decompiling the purely functional object code and ending up
with "identical" source demonstrates that the source elements copied
were -exactly- and -completely- functional. They were, after all, the
disgorgement of the functional element back into the form of source code by the
decompiler.
So if the jury was wrong about the files being copied, then any ruling that the
files have any "non-functional" a.k.a. "expressive" value is
-impossible- by definition. Anything "expressive" would have been
purged when the code was vacuumed up by the compiler and turned into executable
object code. There is no such thing more functional in computer science than a
valid object file.
So Oracle, in proving the original source and the decompiled source
"identical" proved in turn that it was wholly devoid of all
expressive, non-functional elements.
So the parts of the files copied by this method were wholly demonstrated to be
outside the scope of copyright law as -proven- by Oracle.[ Reply to This | # ]
|
|
|
|
|