|
Authored by: Anonymous on Friday, May 18 2012 @ 09:27 PM EDT |
So I agree with the GP regarding clarity of Dr. Parr vs. Dr. Mitchell, but my
point is that Dr. Parr may still have missed an opportunity that could be
critical given the "authority gap" between him and Dr. Mitchell.
Showing one major difference (stack usage) between the two algorithms was easier
to handwave around than demonstrating two major differences (stack usage and
performance differences for pattern matching vs simulated array initialization).
Showing performance improvements on large arrays from using pattern matching
(using nice graphs) would have cemented in the jury's mind that the Dalvik
method wasn't just different, but was used because it was superior.
On the other hand, I can't blame him or her for thinking that the stack argument
would be sufficient - I thought it was a slam-dunk too. It's probably van Nest
who, after seeing a first draft of Dr. Parr's report, should have asked him
"Is there any other way that you can demonstrate a significant difference
between the simulation and pattern matching approaches?"[ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Saturday, May 19 2012 @ 09:35 AM EDT |
Dr. David August is an
Associate Professor of Computer Science at Princeton University.
Dr. Terence Parr is an Associate
Professor and graduate program director of Computer Science at University of San
Francisco. He's also the author of ANTLR, among other things.
It's not as
if Dr. Mitchell is the only expert in this case with impressive credentials. [ Reply to This | Parent | # ]
|
|
|
|
|