|
Authored by: bugstomper on Tuesday, May 15 2012 @ 10:25 AM EDT |
Please post corrections here, summarizing in the Title box error->correction
or s/error/correction/ to make it easy to scan to see what errors have already
been reported.
[ Reply to This | # ]
|
|
Authored by: bugstomper on Tuesday, May 15 2012 @ 10:30 AM EDT |
Please stay off topic in these threads. Whatever topic that may be.
[ Reply to This | # ]
|
- Patent troll - Authored by: Anonymous on Tuesday, May 15 2012 @ 10:47 AM EDT
|
Authored by: bugstomper on Tuesday, May 15 2012 @ 10:33 AM EDT |
Please type the title of the News Picks article in the Title box of your comment
and include the link to the article in the body of your comment, preferably make
it a clicky link and posting in HTML Formatted Mode to make it easy for the
readers once the article has scrolled off the News Picks sidebar.
[ Reply to This | # ]
|
|
Authored by: bugstomper on Tuesday, May 15 2012 @ 10:39 AM EDT |
Please post your transcriptions of the Comes exhibits here with full HTML markup
but posted in Plain Old Text mode to make it easy for PJ to copy and paste.
[ Reply to This | # ]
|
|
Authored by: complex_number on Tuesday, May 15 2012 @ 10:44 AM EDT |
would it take to prepare that lot? I know there was a weekend but really?????
With Oracle's over long motion from last week, and not this lot, the costs in
legal bums on seats for both sides must be getting silly.
Will someone come to their senses and call a halt?
Yes and there goes that squadron of 'Flying Pigs (Gloucester Old Spot variety)'
taking off from Farnborough (UK) once again.
---
Ubuntu & 'apt-get' are not the answer to Life, The Universe & Everything which
is of course, "42" or is it 1.618?
[ Reply to This | # ]
|
|
Authored by: bugstomper on Tuesday, May 15 2012 @ 10:54 AM EDT |
This is really hard to write, since I am about to defend Dr. Mitchell's
testimony even though I think that the correct verdict would exonerate
Google.
I have been trying to answer one question that I did not see clearly
answered by either side in the notes that were reported, and which I could not
find delineated in the documentation I found on Dalvik. That documentation shows
that the Field@CCCC operand of the iget instruction represents a numerical
offset CCCC into a Field Table. It shows that the Field Table is an array of
elements, one per field, each element containing three numbers: An index into a
Class pool, an index into the Constant Pool pointing to a string which is the
name of the field, and an index into the Constant pool pointing to a string
which is the type descriptor of the field.
What it doesn't show is exactly
how you get from an iget Field@CCCC instruction, which refers to element or
offset CCCC in the Field Table, to the iget-quickened instruction which refers
to a numerical offset of the data in the field in the instance object.
That
conversion is done by dexopt. As described in the documentation, one of the
optimizations it performs is that conversion:
For [the] instance
field [of] get/put, replace the field index with a byte
offset.
Oracle quotes that to try to show that dexopt does symbol
resolution. Is the act of replacing the field index with a byte offset actually
symbol resolution?
I saw that, and thought about Dr. Mitchell's
testimony
Oracle: Do Dalvik instructions contain symbolic
references?
Dr. Mitchell: Yes.
Dr. Mitchell: What are they?
Dr.
Mitchell: The field indexes.
Oracle: But those are numbers, does it
matter?
Dr. Mitchell: Not at all, a number can be a symbolic
reference.
And compared it to Andy McFadden's testimony, I think
talking about an iget instruction with an operand of
Field@01
Google: What does "01" tell you?
Mr. McFadden: Index
into the field IDs table.
Google: Does it tell you a location?
Mr.
McFadden: Yes.
Google: What happens when you get to location 01 in the field
table?
Mr. McFadden: You read the data there, and chase that to the next
location.
Google: What happens when you reach the string data table?
Mr.
McFadden: At that point, you're no longer working with numeric values; you've
got string data, and you have to use those to find a matching
field.
Hmm, was Andy McFadden talking about a symbol after
all?
I finally chased down how the field index is replaced with a byte
offset by looking in the source code. What I found was a bit of a shock. Dr.
Mitchell is not talking nonsense, even if there are some semantics for him and
Dr. August to argue over.
The code to find the byte offset from the field
index ends up in a function in file
http://source-android.frandroid.com/dalvik/vm/oo/Object.cpp with the following
signature
InstField* dvmFindInstanceField(const ClassObject*
clazz,
const char* fieldName, const char*
signature)
and which does the following loop to find the
field with the requested name and type descriptor
(signature):
pField = clazz->ifields;
for (i = 0; i < clazz->ifieldCount; i++, pField++) {
if (strcmp(fieldName, pField->name) == 0 &&
strcmp(signature, pField->signature) == 0)
{
return pField;
}
}
return NULL;
For you non-programmers who skipped over the block of
code instead of just stopping reading this comment, here is what the above code
means: The function takes a string which is the name of the field, plus another
string that specifies some other information about the field called a
"signature", and it goes through every field of the class, one by one, comparing
the name and signature of that field with the name and signature being looked
for. When it finds a match, it returns a pointer directly to that
field.
That lookup of a name as a string in a list of items one of which
contains the same string, that is exactly what is meant in common programming
parlance by "resolve a symbolic reference".
What Dr. Mitchell was saying is
that the iget Field@CCCC instruction has an operand that is a number that refers
to a symbol that is the name of the location containing the data to be fetched.
When dexopt replaces the instruction with an iget-quicken that has as operand
the byte offset into the instance, that is what is usually meant my "symbol
resolution".
My conclusion: If you are using ordinary programming or
computer science terminology, iget Field@CCCC is an instruction with a symbolic
reference.
What about Dr August's testimony that disagrees with Dr.
Mitchell? With this deeper understanding of the underlying technology I go back
to his statements, recorded faithfully enough by reporter Zach that I can be
pretty sure of what went on. Notice how carefully Dr. August is to talk in terms
of "the court’s claim construction". He does that every time the result of using
ordinary computer science definitions would lead to a different
result.
August: [Fumbling to get a pointer stick and arrange the
flip chart.] Here we have an instruction 52 and a single operand 01. What is
that operand 01? Let’s apply the court’s claim construction and see. “A
reference that identifies data.” But does it do so by name other than numeric
memory location? Looking at this, 01 directly refers to the location of the
data; the entry is at memory location 01. So the claim construction doesn’t
apply. This is not a symbolic reference.
Google: Where is that entry 01
located?
August: A reference located in the instruction stream.
Google: Where does it point?
August: An entry in the field ID table in
the data portion of the program.
Google: Let’s move to the field ID table.
Do the data there qualify as symbolic references?
August: A two-part
question as there are two pieces of data: yellow and magenta. Start with 02. It
is a reference that identifies data, but 02 is the location of that data, so
it’s not a symbolic reference.
Google: What about the 76 in the field id
table?
August: The 76 also refers to data, identifies data. 76 is the
location in the table in which is occurs, so it is also the numeric memory
location.
Google: Is the 01 in the instruction stream referred to as an
index?
August: Yes. It’s an index into the table.
Google: What’s the
difference between an index and an offset?
August: Both are locations.
Offset may be used more commonly when talking about how many bytes into a table
you want to traverse. Index more commonly used for tables with larger size
entries.
Google: At entry 76 in the string id table, is that a symbolic
reference?
August: No, a numeric reference. Doesn’t fit the construction
because it identifies the data by location.
Google: String data at entry 8
(string “fun” again). Is that symbolic reference?
August: It is a reference
that identifies data. It’s not shown on this board, but I know from looking at
the code that it is. To find “fun,” we need to resolve it by looking for the
location where it refers to. In resolve.c, this is resolved dynamically, so this
does qualify under the claim construction as a symbolic reference.
Google:
What about in the course of dexopt? Would it qualify as a symbolic reference?
August: Not if it is resolved statically. Dexopt is a static linking
process, so it operates statically and not dynamically.
Google: Were you
here for Dr. Mitchell’s testimony? Did you hear him explain that some of the
dexopt references were symbolic references?
August: Yes. Dr. Mitchell is
saying that all these are symbolic references.
Google: How do you explain
the difference in your opinions?
August: Mitchell is applying the court’s
claim construction differently.
This is a bit of weaseling by
Google and August. They start with the 01 and painfully step through every
indexed redirection pointing out all the uses of numerical reference before
finally getting to the string and finally admitting that there is a symbol which
is being used to name the data being referenced. It sure won't seem to the jury
like the 01 operand of the iget is a symbolic reference even though it is. But
he never lied by saying that the instruction does not use symbolic reference to
reference its target data, that is just an implication by listing all those
"numeric references" before finally getting to the symbol. And then he can
honestly say that it is not a symbolic reference under the court's construction
when used in dexopt because it is not dynamic.
Ok, having defended Dr.
Mitchell's testimony, I have to go into why I think Google has it right after
all in regards to the '104 patent.
First, dynamic vs static resolution:
Google is making the most of the claim construction that the Court imposed that
hangs "dynamic resolution" on the definition of "symbolic reference". From a
computer science perspective that is just weird, as they do not necessarily go
together. From the '104 claims it was a strange thing for Judge Alsup to do,
since the six claims of '104 that are in this case do not include any of the
claims that mention "dynamic". The six claims are the broadest claims of the
patent. On what basis is his reasoning for including "dynamic"?
The Abstract
of the '104 patent is all about dynamic symbol resolution and an interpreter of
intermediate code instructions, with the resolution happening during
interpretation and instructions being rewritten.
IANAL, and certainly not a
patent lawyer. I did find a law blog article about writing abstracts for
patents that says, in part
Although only a summary, it is
prudent to draft an Abstract with care. The Abstract is a part of a written
disclosure of the application and Federal Courts may properly rely on an
Abstract to construe claims. For this reason, it is beneficial to draft an
Abstract at least as broadly as the broadest independent claim.
If
that's true, then Judge Alsup could have been correct to hold all the claims to
the limitations regarding dynamic resolution.
Oracle is doing their own
weaseling about the dynamic aspect, trying to first conflate "dynamic" with the
"at runtime" concept that is only relevant to the other patent in the case, and
then making their obfuscation worse by conflating "at runtime" with "any time
the phone is powered on and so the VM may be running something". In the face of
that I don't feel so bad about Google gaining some advantage from the claim
construction language they got from Court and then insisting on that language
instead of conventional computer science definitions.
Then there are
implications of the rejections from the PTO. The broader the patent is construed
the more stronger teh rejections based on prior art. As far s I can see from the
reexamination report, if any claims survive it will be only on the basis that
they are construed as referring to interpreted instructions that specifically
contain the symbolic reference and are replaced by the interpreter itself with
new instructions that contain numeric references. Java doesn't practice that and
I doubt anything else does.
[ Reply to This | # ]
|
- How Dr. Mitchell is right even if his side is wrong - Authored by: Anonymous on Tuesday, May 15 2012 @ 11:06 AM EDT
- Was Andy McFadden talking about a symbol after all? - Authored by: hardmath on Tuesday, May 15 2012 @ 11:07 AM EDT
- How Dr. Mitchell is right even if his side is wrong - Authored by: Anonymous on Tuesday, May 15 2012 @ 11:10 AM EDT
- How Dr. Mitchell is right even if his side is wrong - Authored by: Anonymous on Tuesday, May 15 2012 @ 11:37 AM EDT
- When is the reference resolved? - Authored by: Anonymous on Tuesday, May 15 2012 @ 11:47 AM EDT
- How Dr. Mitchell is right even if his side is wrong - Authored by: Anonymous on Tuesday, May 15 2012 @ 12:23 PM EDT
- How Dr. Mitchell is right even if his side is wrong - Authored by: tknarr on Tuesday, May 15 2012 @ 12:35 PM EDT
- Thanks for the analysis - we see the language problem here - Authored by: PolR on Tuesday, May 15 2012 @ 12:37 PM EDT
- Can't argue with the code - Authored by: Anonymous on Tuesday, May 15 2012 @ 12:38 PM EDT
- lacking complete testimony - Authored by: Christian on Tuesday, May 15 2012 @ 12:43 PM EDT
- Thank you bugstomper! - Authored by: jbb on Tuesday, May 15 2012 @ 01:00 PM EDT
- Warning: contains '104-Claim 11 - Authored by: Ian Al on Tuesday, May 15 2012 @ 01:19 PM EDT
- Just because it can resolve a symbolic reference - Authored by: Anonymous on Tuesday, May 15 2012 @ 02:25 PM EDT
- I disagree with your conclusions - Authored by: Anonymous on Tuesday, May 15 2012 @ 02:29 PM EDT
- How Dr. Mitchell is wrong even if his side is wrong (Warning Patent 104) - Authored by: Anonymous on Tuesday, May 15 2012 @ 04:17 PM EDT
- Everything after the compile step is dynamic? - Authored by: Anonymous on Tuesday, May 15 2012 @ 04:55 PM EDT
- How Dr. Mitchell is right even if his side is wrong - Authored by: jwrl on Tuesday, May 15 2012 @ 11:58 PM EDT
- Claim free refutation - Authored by: Ian Al on Wednesday, May 16 2012 @ 02:33 AM EDT
|
Authored by: Anonymous on Tuesday, May 15 2012 @ 11:23 AM EDT |
With many thanks to Feldegast and the reporters
Feldegast
tweets
Raw feed
[ Reply to This | # ]
|
- Tweets from the courtoom - Authored by: Anonymous on Tuesday, May 15 2012 @ 11:25 AM EDT
- Juror not showing up - Authored by: Anonymous on Tuesday, May 15 2012 @ 11:26 AM EDT
- Attorneys say dismiss missing juror - Authored by: Anonymous on Tuesday, May 15 2012 @ 11:27 AM EDT
- Juror will be dismissed - Authored by: Anonymous on Tuesday, May 15 2012 @ 11:28 AM EDT
- Juror is number 2 - woman from the front row - Authored by: Anonymous on Tuesday, May 15 2012 @ 11:30 AM EDT
- Jury instructions read to empahsise willfulness higher burden - Authored by: Anonymous on Tuesday, May 15 2012 @ 11:44 AM EDT
- Jacobs wants jury to focus - Authored by: Anonymous on Tuesday, May 15 2012 @ 11:48 AM EDT
- Jacobs: "You don’t avoid infringement because Android is big” - Authored by: Anonymous on Tuesday, May 15 2012 @ 11:49 AM EDT
- Jacobs: "Google has no defense & concedes most elements of claims..." - Authored by: Anonymous on Tuesday, May 15 2012 @ 11:51 AM EDT
- Oracle showing slides of tipping scales... - Authored by: Anonymous on Tuesday, May 15 2012 @ 11:53 AM EDT
- Jacobs discussing symbolic references: these appear in android - Authored by: Anonymous on Tuesday, May 15 2012 @ 12:03 PM EDT
- Jacobs admits flaws in Mitchell's report, but they were honest mistakes, and the rest is golden - Authored by: Anonymous on Tuesday, May 15 2012 @ 12:05 PM EDT
- Jacobs finished around 45 mins. Van Nest up next - Authored by: Anonymous on Tuesday, May 15 2012 @ 12:21 PM EDT
- Jacobs stresses importance of patents - Authored by: Anonymous on Tuesday, May 15 2012 @ 12:40 PM EDT
- Jacobs shows ostrich photo - willfulness - Authored by: Anonymous on Tuesday, May 15 2012 @ 12:43 PM EDT
- Jacobs will have 6 minutes for rebuttal - Authored by: Anonymous on Tuesday, May 15 2012 @ 12:48 PM EDT
- Van Nest begins closing args: Nothing happens on time in a 5wk trial - Authored by: Anonymous on Tuesday, May 15 2012 @ 01:09 PM EDT
- Van Nest: No evidence Google saw the patents before lawsuit - Authored by: Anonymous on Tuesday, May 15 2012 @ 01:12 PM EDT
- Van Nest - Question 1 is the most important - Authored by: Anonymous on Tuesday, May 15 2012 @ 01:20 PM EDT
- Van Nest good at presenting in layman's terms - Authored by: Anonymous on Tuesday, May 15 2012 @ 01:21 PM EDT
- Van Nest: "Dalvik only contains numeric references NOT symbolic - Authored by: Anonymous on Tuesday, May 15 2012 @ 01:34 PM EDT
- Van Nest: Oracle trying to change claim construction - Authored by: Anonymous on Tuesday, May 15 2012 @ 01:42 PM EDT
- Continued in next article comments n/t - Authored by: Anonymous on Tuesday, May 15 2012 @ 01:50 PM EDT
|
Authored by: Anonymous on Tuesday, May 15 2012 @ 10:51 PM EDT |
1130 (Google's Memorandum in
Opposition to Oracle's Motion to Defer Phase Three) finishes a great
parting-shot. The part of the memo before the end, argues that the
determination of damages resulting from phase 1 (including for those test files
where Judge Alsup overruled the jury's verdict) should be done in this trial, by
this jury. Then this:
Finally, Oracle argues that it would be
prejudicial to have the jury that found no liability
on the test files now be
told to consider damages as to those files. Dkt. 1126 at 5-6. This
concern
can easily be addressed through the Court’s instructions to the jury on
statutory
damages. “Juries are presumed to follow the court's instructions.”
Aguilar v. Alexander, 125
F.3d 815, 820 (9th Cir. 1997) (citing
Richardson v. Marsh, 481 U.S. 200, 211, 107 S. Ct. 1702,
95 L. Ed. 2d
176 (1987)); see also Jules Jordan Video, Inc. v. 144942 Canada Inc., 617
F.3d
1146 (9th Cir. 2010) (“There is a strong presumption that juries follow
curative instructions. See
Doe ex. rel. Rudy-Glanzer v. Glanzer, 232
F.3d 1258, 1270 (9th Cir. 2000).”).
But if Oracle finds that still too
risky, it could join Google in waiving the right to have the
jury decide the
damages for the eight test files and rangeCheck and agree to have these issues
tried to the Court.
For all of the foregoing reasons, Oracle’s motion
to defer phase three should be denied.
(the bold emphasis
is mine, the italics are from the original).
Google basically says "if
Oracle doesn't trust the jury, We'd be happy to let the Court decide". But
somehow I don't think Oracle would want that...
I smirked when I read it,
because the Judge Alsup has expressed a couple of times now, in stronger terms
each time, his extreme scepticism with the infringers-profits tack Oracle is
pursuing. [ Reply to This | # ]
|
|
Authored by: Anonymous on Tuesday, May 15 2012 @ 10:57 PM EDT |
1127 is Oracle's list of the
next anticpated witnesses. Both Larry Page and Eric Schmidt are on
it.
Sigh.
Google, for their
part, doesn't have Larry Ellison on their list.
Too bad, I'd love
to hear him explain under oath how this case got started. [ Reply to This | # ]
|
|
|
|
|