Latest · Mon · Tues · Wed · Thurs · Fri
         
Monday 25 February, 2002
Monday Morning Roundup
Dr. Pournelle had a couple of well thought out short essays last week (see them starting here as soon as he updates his site. Until then, you can see them here). I don't always agree with him but he probably writes more good stuff before breakfast than many people do in their lifetimes. So if you haven't read them yet, take a look. [By the way, you don't need to be a graduate student to read the Federalist Papers, although parts of it were required reading for one of my classes. You, dear readers, can read them here)
Speed is Good
I've been using a Plextor 8X CD-RW for a couple of years. And while it was a speed demon when it first came out, Moore's law affects CD-RW drives just as it does CPUs. So this weekend I bought a Plextor Plexwriter 40X12X40A burner.

Not only is this drive fast, but even better, it includes Sanyo's BURN-Proof technology. This technology avoids the main cause of coaster creation, buffer underun. This also makes it possible to get back to multitasking, something that was essentially impossible without it. Best of all, it allows me to use my DVD drive as the source and the Plextor as the target, thus allowing me to make copies "on the fly" without first creating a temporary image file to the hard drive and then burn that to the CD. Setup like this, I made a CD to CD copy in just under three minutes on my Pentium III 933MHz PC with 512MB of RAM. YMMV. Don't try this at home (go to your friend's house). Professional driver in a closed course. Void where prohibited.

Speed Kills
Or is it the sudden stop that does the trick? Fellow Daynoter Jan Swijsen and I have been going back and forth regarding the private contractor enforcement of the speed limits here. The abridged version is that I don't think it is appropriate for a private business to be be providing this service (what's next, private police?). Rather, I believe it is the proper role of the public sector, not the private. Even further, I don't believe the enforcement of arbitrarily set speed limits, above which a reasonable person can routinely drive safely, engenders resect for the law nor reduces the number or severity of crashes. I would rather spend money on drunk driver checks, reducing the distractions inside the car (fiddling with radios/CD/cassettes/talking on cell phones/using the GPS/etc. and finding alternatives for young boys who want to prove their manhood through street racing.

From: Jan Swijsen
To: Dan Seto
Date: Fri, 22 Feb 2002 18:53:28 +0100
Subject: Re: Politics! bhwa.

<snip>The question has always been how much are you willing to risk?</snip>

That is indeed a basic question. However you are never alone in traffic. If you are willing to risk driving trough a residential area at 50 Km/h and someone else is willing to risk doing the same thing at 150 Km/h then what YOU are willing to risk doesn't matter. Most people will accept the 50 Km/h risk as reasonable while only a very few will think 150 Km/h reasonable.

<snip> the statistics clearly show that the safest roads to be on, per mile driven, are the highways?</snip>

Aren't statistics worse than politics? <g> Reasoning off that finding : The average speed on the highway is 120 Km/h. So 120 Km/h is the safest speed to be driving at. So in residential areas we must make sure people drive as close to 120 Km/h as possible.

The same type of reasoning can be applied in reverse. Most (light) accidents happen on parking lots with cars driving less than 5 Km/h. Eliminating parking lots will result in a significant reduction in the number of accidents.

People residing in small, elongated, underground residences (sometimes called coffins) never apply for health care. People not living in such do apply for health care. Public health care cost loads of money and has lots of problems. So put everybody in a coffin and all your health care problems are solved.

(need more examples of 'applied statistics'? <g> )

Statistics are good when used well and when based on correct data. Actually statistics are always neutral, it's all and only numbers. It's people basing conclusions on these numbers that are making the mistake of putting the numbers out above reasoning.

What we see happening is that we have a speed limit of 90 Km/h (while most people ignore it and fly by at around 100 Km/h because the layout of the road allows that) and (say) 100 accidents per year 1010% of which result in the death for someone (10%). The limit is then brought down to 70 Km/h. Most people don't care and keep driving along at 100 Km/h because there is no police watching so only a few reduce their speed. Those who don't reduce their speed however look out better to make sure they are not caught on the odd chance of a patrol. This heightened attention alone is likely to reduce the number of accidents. So now he have a limit of 70 Km/h and (say) 80 accidents a year with 8 death (10%) . So the limitation seems to have had an effect because there are less accidents and less killings..

What is not analyzed however is that the severity of the accidents has not been reduced. What should have been expected if the speed reduction had actually been enforced : a few less accidents (say 90) and all accidents less severe (say only 3 deaths).

Yeah, statistics are almost as bad as politics. But turn politicians loose on statistics and chaos theory suddenly seems easy.

--
Svenson.

Et Fini
I see that JHR has apparently decided to stop posting and devote his time and energy to other pursuits (see it here). JHR is another person that I respect, but don't necessarily always agree with. His posts and emails have stirred the pot of ideas more than once. Some people have reviled, nay even pillioried his opinions. And yet he came back for more and gave as good as he got.

I'm going to miss his voice in the marketplace of opinions. Happy trails to you.

Aloha!

Tuesday - 26 February, 2002
Down and Out
This is one of those it will get posted when it gets posted kind of days. Our Internet access, which has been a little flakey recently, completely went south this morning. First the DNS went, then the firewall went into lockdown mode and is denying all inbound traffic. Sigh.
Down But Not Out
The latest from JHR is that he has reconsidered completely shutting down. He is, however, substantially cutting back his frequency of posting. I can understand that and he is not the only one to come to that decision (see Matt, Al, Doc, Jonathan, Greg, and Bob W. to name a few). Well, post'em when you got'em JHR.
Corporate Values
Yes, I know, some would consider the title to be an oxymoron. And for the most part you would be right. But every once in awhile, someone does a study to determine what, if any, are the economic effects of good corporate values. Thank you to InfoWorld's Bob Lewis (see his column here) for the link to a study which says good HR practice adds value to the bottom line (see the study here). Below is a short excerpt:
Do better people management strategies actually create higher market value? Or do financially successful companies simply have more resources to allocate to human capital initiatives?

We now have powerful insight into the answer. Our second Human Capital Index study allowed us to compare one set of companies at two points in time to analyze the correlation. The results are in and they are dramatic. Superior human capital practices are not only correlated with financial returns they are, in fact, a leading indicator of increased shareholder value. Further, we found that superior HR management leads financial performance to a much greater extent than financial outcomes lead good HR. We were also able to identify certain HR practices as value drivers and throw a cautionary flag in front of some conventional practices actually associated with a decrease in financial performance.

Opera 6.01
Just a quick note to let you know Opera 6.01 is out.

Aloha!

Wednesday - 27 February, 2002
Search Me
I'm in the process of updating my search engine software and things are not going well. The first pass through the everything ran without error but when all was done, I had multiple directories such as foo and foo?, bar and bar?, with the only difference being the question mark. I assume the install has some bugs in it so I tried redoing the config file. But right now, I get 'make' errors so I guess it's back to the drawing board. If I can't get the newer version going, I'll reinstall the old one. So, please don't use the search function until further notice because all you will get is an error message.
They Hate Us, They Really Hate Us
For what it's worth, USA Today is reporting a Gallup poll which says, on average, 61 percent of Muslims believe Arabs were not involved in the September 11 attacks. In fact, 89 percent of the nice folks in Kuwait don't think any Arabs were involved. Kuwait, isn't that the country that got invaded by its Arab friend and where US and NATO troops died getting it back? Perhaps they thought the Israelis had taken over their neighbor and were invading? And who do they think were in the planes that hit the WTC and Pentagon? Scientologists? Fijians? Laplanders? Those blood thirsty Swiss Guards? I really don't understand these people. Sheesh.

Aloha!

Thursday - 28 February, 2002
Searchin'
Okay, it took awhile to get the search engine back up and running. The problem yesterday was in the paths found in the configuration file. Specifically, I used wordpad to edit the file not realizing how wordwrap would break longer lines like paths. This is not a good thing. The second problem with the paths was that one of them pointed to the wrong place, a definite boo boo on my part. So once I got the paths corrected, 'make' and 'make install' did their things.

Of course, since its been almost two years since I originally installed the earlier version and since I changed the directory where it is installed, I had to look back here to see how to create/edit the files needed to get a crontab file going. Hopefully, all is back to normal but if you see any problems, please let me know. Mahalo.

I have to get ready for our staff meeting this morning so I gotta' go - Aloha!

Aloha Friday - 29 February, 2002

It's Friday!

Tali-vans
The State Transportation Department is trumpeting, as I figured they would, that traffic crashes for January were down 14 percent as compared to January of last year. And as I said before, the numbers mean nothing because they have not been normalized. By that I mean you need to take into account the number of people, and the distance they drive (and probably a bunch of other conditions) because these factors may skew any effects you are trying to find. Hence, if you aren't at least reporting the rate of crashes per mile driven, you aren't reporting anything relevant.

Oh, by the way, the department doesn't seem to be letting too many people know that the number of traffic related deaths is running at an all time high, as compared to last year. But then, they say that isn't relevant.

But you know, truth be told, even if the rate of crashes per miles driven went down and could be directly related to the traffic cams, I would probably still oppose it. Why? Because I think it is bad public policy to give police powers to private businesses. Businesses who are there solely to make money. To make money off of you and I.

So?, you ask, what's wrong with that? Think about this. One of the other "services" this company will be performing is to catch people running red lights. Now, I don't think anyone is for running red lights. But how many tickets a company can give is directly related to the number of seconds a yellow light is displayed. That is, the shorter the time a driver has to exit an intersection, that is the shorter the yellow light before the red comes on, the more tickets can be issued.

And yet, at least one study has shown that if you want to substantially reduce the number of crashes caused by running a red light, the answer is to increase, by on average, one second, the time the yellow light is displayed. Hmmm. We seem to have competing values here. On the one hand, the company wants to maximize income so it will push to decrease, or at the very least, not increase, the length of time for a yellow light. On the other hand is the public interest of increasing safety. Which do you think wins? The answer to that question is up to you. YMMV.

Have a Great Weekend Everyone - Aloha!


© 2002 Daniel K. Seto. All rights reserved. Disclaimer

Home

Diary Index

Last Week

Next Week

The Daynotes Gang

Contact Dan