Authored by: Anonymous on Monday, May 14 2012 @ 08:31 PM EDT |
n/t [ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Monday, May 14 2012 @ 08:39 PM EDT |
Yes. For 50 years or so, practically no instruction has ever contained a
memory location as they have just been offsets (sometimes to more offsets)
into a table. Then the OS/CPU would load up the tables into memory and use
the offsets to calculate the memory locations. Google should have used their
invalidation arguments of the patents with Dr Mitchell's definitions.
Also, Dr. Mitchell is confusing a program being loaded into the VM as data
versus being loaded into the VM as instructions. Dexopt loads the code/
instructions into memory as data to do the static analysis like every compiler/
optimizer ever created. From my understanding, the patents are for when the
VM is running the code as instructions to perform the dynamic analysis on the
fly like .NET.[ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Monday, May 14 2012 @ 08:42 PM EDT |
Well, Gödel's proof works by assigning a numeric value to strings of symbols in
a proof-language. Every UCS-2 string of characters can also be viewed as a base
(2^16) number, as well... so I think there are references that can be viewed
as both numeric and symbolic. However, I'm sure there's prior art for that
concept. From the reports I don't see prior art being used to limit the
patensts; I wonder if that could be brought in again on appeal?
[ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Monday, May 14 2012 @ 08:43 PM EDT |
Well, Gödel's proof works by assigning a numeric value to strings of symbols in
a proof-language. Every UCS-2 string of characters can also be viewed as a base
(2^16) number, as well... so I think there are references that can be viewed
as both numeric and symbolic. However, I'm sure there's prior art for that
concept. From the reports I don't see prior art being used to limit the
patensts; I wonder if that could be brought in again on appeal?
[ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Monday, May 14 2012 @ 08:46 PM EDT |
You are obviously not familiar enough with Stanford to be able to say that. [ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Monday, May 14 2012 @ 08:52 PM EDT |
My professor for data structures used array indicies to demonstrate memory
addresses all the time. Never once were they thought of as symbolic. There is
a direct relationship between array indicies that is closer to memory addresses
that most modern logical/physical translation engines built in to the
processors.
The witness committed perjury because Oracle paid him to.[ Reply to This | Parent | # ]
|
|
Authored by: jbb on Monday, May 14 2012 @ 09:35 PM EDT |
If you keep Oracle's witnesses from lying then how can Oracle possibly
win?
--- Our job is to remind ourselves that there are more contexts
than the one we’re in now — the one that we think is reality.
-- Alan Kay [ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Monday, May 14 2012 @ 10:26 PM EDT |
This isn't about the tens of thousands that Dr Mitchell receives, it's about the
tens of millions that Stanford receives from Ellison/Oracle. Dr. Mitchel is a
department head, he represents Stanford in this.
Follow the money, google this set for precedence: Ellison, Henley, Lucas,
Boskin, Molina, Grundfest[ Reply to This | Parent | # ]
|
|
Authored by: rcsteiner on Tuesday, May 15 2012 @ 01:56 AM EDT |
If Oracle wins anything because of his testimony, I wonder if his actions would
open him up to legal proceedings?
---
-Rich Steiner >>>---> Mableton, GA USA
The Theorem Theorem: If If, Then Then.[ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Tuesday, May 15 2012 @ 02:20 AM EDT |
IANAL but I believe the law has trouble with either/or arguments. You and I can
look at this and see a very simple argument which says
... EITHER a numeric index is not symbolic in which case dalvik doesn't violate
the patent
... OR a numeric index is symbolic in which case the patent is invalidated by
prior art going back 50 years
... HENCE Oracle gets no billions.
However I understand the law has trouble dealing with this kind of argument. [ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Tuesday, May 15 2012 @ 08:16 AM EDT |
All of you people using C++ and vtables are in a lot of
trouble. I hope you have $6B handy...[ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Tuesday, May 15 2012 @ 08:57 AM EDT |
An index into the string pool is the way quite a few languages represent
strings.
I don't see why that would make a reference non-symbolic.[ Reply to This | Parent | # ]
|
- Unbelievable! - Authored by: Anonymous on Tuesday, May 15 2012 @ 09:34 AM EDT
|
Authored by: Anonymous on Tuesday, May 15 2012 @ 09:50 AM EDT |
Yep.
Everyone with basic knowledge of how an array works in memory
can tell you that. And several people have.
Memory offsets ARE memory locations. If programs didn't use
offsets they could only function in one part of memory. By
having compilers use offsets, the OS can stick the program
into any section of RAM it wants and it'll work.
(Assembly/machine code programs are the exception here)[ Reply to This | Parent | # ]
|
|
Authored by: jbb on Tuesday, May 15 2012 @ 01:04 PM EDT |
Please see
this
analysis by bugstomper.
I think we owe Dr. Mitchell an apology. I,
for one, am sorry.
--- Our job is to remind ourselves that there are
more contexts
than the one we’re in now — the one that we think is reality.
-- Alan Kay [ Reply to This | Parent | # ]
|
- nah - Authored by: Anonymous on Tuesday, May 15 2012 @ 09:59 PM EDT
|