|
| Authored by: sd_ip_tech on Monday, May 21 2012 @ 04:48 PM EDT |
I have to believe that Van Nest et al knew these issues but had no time to
present. Agreed, 104 is a slam dunk without the arguments you state. Oracle made
sure that Google had to cover multiple bases and here we are. Still a possible
hung jury. Then 104 goes away for next time.
---
sd_ip_tech[ Reply to This | Parent | # ]
|
|
|
| Authored by: Anonymous on Monday, May 21 2012 @ 04:53 PM EDT |
It feels like they missed making the point (assuming its
true) that in this context dynamic means the resolution
happens every time the code runs. Static means its done
once and every execution after that it doesn't have to be
done again.
The confusion is that the oracle dynamic variant here makes
the optimisation whenever the code first references the
symbol(s) but doesn't have to do it again whilst the code is
running. If the app stops or is restarted you're back to
square one.
The dexopt model runs once at install time. After that -
even if you turn the box off it doesn't have to run again.
(Another correspondent made this point a while back but I
haven't seen it raised in the transcripts.)[ Reply to This | Parent | # ]
|
|
|
| Authored by: Anonymous on Monday, May 21 2012 @ 05:17 PM EDT |
On the contrary!
Google defense was great and Oracle sounded mumbled.
Oracle has to prove the claim and any wrench thrown in the wheels by google is
fair game in the court.
Ordinary people don't get much of the patents anyways, it's just the show put up
that counts for them.[ Reply to This | Parent | # ]
|
|
|
| Authored by: PJ on Monday, May 21 2012 @ 05:22 PM EDT |
Google did respond to that accusation. Review the
closings.[ Reply to This | Parent | # ]
|
|
|
| Authored by: Anonymous on Monday, May 21 2012 @ 05:47 PM EDT |
On the '520 patent, they did a better job, but they should have
pushed home
the
point that simulation requires that feeding in different but
equivalent inputs
requires getting the same output, but pattern matching does
not. Then asked
why dexopt failed in the tests with tweaked
input.
I'm pretty sure I saw that exact argument put forward by
Google. Perhaps they
didn't stress it enough, because as you note - that is a
clear and definitive
proof against simulation.[ Reply to This | Parent | # ]
|
|
|
| Authored by: Anonymous on Tuesday, May 22 2012 @ 01:15 AM EDT |
Sorry bud, but it would take weeks to explain that properly to a jury.
I'm beginning to feel we need to go to an experts panel type system for highly
technical cases.[ Reply to This | Parent | # ]
|
|
