The debate is on. The upcoming blogospheric analysis of this US Presidential debate will prove interesting. Kerry seems pretty smooth, but he keeps pointing with his thumb. a live blog of the event.
Update 8:35 PM - Kerry won in my opinion. Not a big win, but this debate certainly didn’t knock him out of the race. Others think differently.
It’s almost totally unbelievable, but in this post on LGF, despite what Occam’s razor tells us about stuff like this and all of the other material that makes up a "preponderance of evidence", we learn that some among us are still holding onto an irrational belief that the Bush/Killian memos might be authentic. David E. Hailey, Jr., Ph.D. has a new analysis proving this.
Dr. Hailey’s argument is full of a number of hilarious holes (like his use of a word processor to prove that a typewriter could have created the memos), but these errors in logic aren’t the only reason he looks like a fool…. His credentials and his analysis might be a lot more impressive, if he had spent a little more time proofing the section of his analysis that attempts to extol his and his Lab’s qualifications:
|
| Screenshot of Dr. Hailey’s analysis from his website |
These errors are also present in the PDF version of the analysis (local copy).
$ ls -l Bush_Memos.pdf -rw-rw-r-- 1 bohnsack bohnsack 1278587 Sep 30 07:17 Bush_Memos.pdf $ md5sum Bush_Memos.pdf c3a9b2af0544e725cc80ccda0d25065f Bush_Memos.pdf
There are few of us who do not make the occasional grammar or diction mistake, but an error of this type in a paragraph that’s supposed to lay out an English professor’s credibility is beyond amusing. Details matter, especially when you’re trying to prove something. If this English professor has so much difficulty with English, one has to wonder if he could have similar problems with critical thinking skills.
Update 2:39 PM - Not only does Dr. Hailey need a fifth grader to help with his grammar, it turns out he’s also a total fraud. Ha. Ha.
There seems to be a trend developing here…
Update 6:36 PM - Maybe not a fraud, but still lame.
Update 10/1/2004 7:20 AM - The professor’s website is down this morning and there are other developments
Update 10/7/2004 3:55 PM - Not a fraud, but unquestionably incompetent and blinded by bias.
Luiz DeRose from Cray
o Tools and Techniques for Performance Analysis
- profiling
- hardware counters
- PAPI
- perfsuite
- MPI logging
o Hands-on lab
o I/O Performance tools
- parallel I/O overview
- parallel I/O tools
Federico Bassetti
o Realistic Performance Expectations on Linux clusters
Made up for being new at NCSA here. Lots of good information.
o Hands-on lab
Today was mostly excellent. They earned their conference fees today.
I’m attending the Linux Cluster Institute’s Applications Module Program at UNM today and the rest of the week.
Today’s topics:
Doug Pase from IBM
o Performance Programming on Intel & AMD Architectures
Stuff about memory/register/cache arch, vectorization, etc. Some interesting things were discussed, but I was mostly familar with the material. It was good to cover it as intro before the rest of the day’s information.
o Performance Programming on Intel & AMD Architectures
Apply stuff from the previous discussion to basic performance techniques: loop unrolling, memory access patterns, data alignment, FP multiplication by recripocal instead of division, subroutine inlining, etc. Interesting stuff. It clarified at lot of stuff that I was only vaguely aware of.
Ron Brightwell from Sandia
o Linux Cluster Programming
Overview of MPI. Again, stuff I knew a little bit about is now much clearer. Excellent info. Good speaker. Poor slides: Sorry, not my slides, that point is incorrect, etc.
Luiz DeRose from Cray
o Debugging parallel applications with Totalview. Far too much basic information about how a debugger works in serial mode. I know that stuff. Skip to the parallel stuff please. Why show powerpoint slides of a program? Show the actual program in operation. Oh.. and we don’t actually have a way to run Totalview in parallel, because LSF on the machine we’re using doesn’t support X11 tunneling in interactive sessions. Nice.
Federico Bassetti from NCSA
o Local Conputing Environment. A discussion on how to use some of NCSA’s resources. Nice guy, but he didn’t seem an expert on what’s in NCSA’s machine rooms. Many incorrect slides. Again, why talk and point, when we could be typing.
I took a trip around southern New Mexico yesterday… Valley of Fires Recreational Area, Lincoln National Forest, White Sands National Monument, etc. There were a lot of interesting things to to see, and many images were captured, but processing and upload of the majority will be delayed until later in the week.
|
|
|
|
|
Sky Captain and the World of Tomorrow: The effect was cool for five minutes. Aside from that, the movie was pathetic and a waste of my money. Bad, bad, bad.
|
Canon’s new flagship 35 mm SLR: EOS-1Ds MarkII
Street price: $7,999 |
Wow. What kind of Canon-nut photography enthusiast wouldn’t want of these babies? Too bad its price puts it out of striking distance for the amateur. Personally, I’m going to wait to see what the 20D’s successor is going to look like, before I replace my 10D. It might be a wait of more than a year and a half, but I can probably handle it.
Head over to slashdot for Michael Badnarik’s answers to 15 questions asked by slashdot readers.
Nader’s back on the New Mexico ballot. Evidently, campaign finance records show judge Wendy York, who had previously issued a ruling blocking Nader from New Mexico’s November 2nd ballot, contributed $1,000 to John Kerry’s presidential campaign in April. Naturally, a GOP state senator wants judge York to step down.
Wheee….
Update: Nader’s back off the ballot via a different judge.
You are currently browsing the Matthew Bohnsack weblog archives for September, 02004.
