Wednesday, February 25, 2004

photoSIG ? Little tear

Photo: Little tear

what a great photograph!

On Locke vs. Davey


from NYT - Supreme Court Approves Denial of Divinity Scholarships

The Supreme Court ruled today, in a case watched by public officials and educators across the country, that the states can withhold public scholarship money from students pursuing religious studies.

The justices decided, 7 to 2, that Washington State had the right to deny scholarship aid to a college student who was studying to be a minister.


The two justices disagreeing were, predictably, Antonin Scalia and Clarence Thomas. In their opinion,

"The indignity of being singled out for special burdens on the basis of one's religious calling is so profound that the concrete harm produced can never be dismissed as insubstantial," Justice Scalia wrote.


I totally agree. However, the question is whether the person was indeed "singled out" on the basis of his religious calling. The person, especially the specific person in question, had a choice of pursuing his religious calling without a state scholarship that explicitly does not provide money for devotional theology degrees or pursuing a non-religious carreer in which case he could have scholarship money. Mr. Davey tried having it both ways by pursuing two degrees at the same time. However, he was in no way prohibited from participating in any kind of religious ceremonies or activities by the conditions of his scholarship. He was not prohibited from actually studying theology - only of training to become a minister. Should Washington State pay for the training of ministers (and rabbis, imams, etc.) ? I think most people would agree that would be difficult to reconcile with the consitution.


Moreover, I think it is an interesting twist that were he allowed to proceed in his studies using scholarship money ACLU would surely claim that this was a blatant violation of the "wall between church and state". IMO, that is precisely the reason Washington State originally excluded full-time pursuit of theological degrees from a list of programs its Scholarship Program could be used for. The State erred on the "state side" as it were of the Amendment. How could they not?

Could they have extended excluding funding for any theological courses from being covered by the scholarship? Perhaps. Would that be undue restriction on one's freedom to exercise the religion? Probably not. However, since it would be impractical (impossible?) to determine with any certainly whether an individual is planning to take a class in order to learn history, ethics, or as a multi-cultural requirement instead of a solemn theological obligation it makes sense to keep the study of religion as a funded activity and merely prohibit intentional training for ministry as a degree fundable by a secular tax polity such as Wash. State.

"The American Center for Law and Justice, a law firm founded by the religious broadcaster Pat Robertson, said the ruling 'clearly sanctions religious discrimination.'"

Somehow I do not think this what this law firm would say this if Mr. Davies sought to become an imam or a budhist lama. [ed - how do firms "speak"? why was no spokesperson's name mentioned?] More importantly, the ruling clearly sanctions only the ability of State not to fund religious education. Perhaps distatesfully to the abovementioned firm, the Supremes kept to leveling the playing field for all religions, without singling out any one of them.

Greenspan Urges Social Security Cuts

Going to be news tomorrow when newspapers get into it - Greenspan Urges Social Security Cuts

Tuesday, February 24, 2004

The Kerry Cascade - How a '50s psychology experiment can explain the Democratic primaries. By Duncan?Watts

The Kerry Cascade - How a '50s psychology experiment can explain the Democratic primaries.


Yeah, what that guy said - Duncan Watts' article in Slate
Everything pundits are saying about Dean now could just as easily be used (and would have been used) to "explain" a Dean victory. Had that happened instead, we would all be walking around saying, "Well, of course Kerry lost—he's got all the charisma of a dead horse—and that Dean is a real firebrand." In each of these "parallel worlds," Dean and Kerry are exactly the same (more or less), and voters are (more or less) exactly the same as well. In terms of the inputs, the difference between the two worlds could be a coin toss. And yet the results, along with our collective memory of what happened and why, are absolutely, completely different, and we can't even imagine now what that other world would have looked like, let alone how vigorously we'd be rationalizing it.


read the whole article.

Monday, February 23, 2004

Etymology of First Names

Etymology of First Names

good site - Etymology of First Names

Thursday, February 19, 2004

YAWS - Yet Another Web Services Spec


WS-Discovery is a yet another spec. proposed to enhance the way devices can discover and consume Web Services. It is intended as an add-on or supplement to UDDI, although how much it adds is still unclear.

WS-Discovery is focused on two key things: notifying systems through a multicast protocol when a device or service is available, and providing a location "bootstrap" so that UDDI systems and event-driven systems can continue to locate and communicate with the device.

The protocol works by providing a means by which a device can "announce" itself to a local network, and then "chat" with interested systems and devices that are looking to communicate with it.


I have never seen anyone specifically explain how this "chat" is going to work. The only examples we ever get are the same rehashed version of "everything will just work together", sometimes with updated or newly popular terms (like RFID) thrown in. For example:

Steven van Roekel, director of platform strategy at Microsoft, said he envisions a WS-Discovery scenario where one day a truck can drive into a warehouse and RFID (define) tags can be used to register what inventory would be in the truck.


Truck? Warehouse? Is this a warehouse full of fridges that can order more food from the store for you? Register what with what? And why? It is exactly this type of vague and unusable examples that do not help readers to learn anything about the new technology and push adoption rates down and time frames outward.

It is entirely possible that we needed another protocol for connecting devices to web-services rather than just web-services to web-services, I just wish the need, use, and landscape into which this spec. needs to fit into was explained much more clearly.

Wednesday, February 18, 2004

The Peter Principle - from Marginal Revolution



Remember the Peter Principle? It suggested that we would all be promoted to a level where we are incompetent.



Economist Ed Lazear asks whether we might expect such a result from profit-maximizing businesses. His article, "The Peter Principle: A Theory of Decline," appears in the latest Journal of Political Economy.



Here is the abstract:



Some have observed that individuals perform worse after being promoted. The Peter principle, which states that people are promoted to their level of incompetence, suggests that something is fundamentally misaligned in the promotion process. This view is unnecessary and inconsistent with the data. Below, it is argued that ability appears lower after promotion purely as a statistical matter. Being promoted is evidence that a standard has been met. Regression to the mean implies that future ability will be lower, on average. Firms optimally account for the regression bias in making promotion decisions, but the effect is never eliminated. Rather than evidence of a mistake, the Peter principle is a necessary consequence of any promotion rule. Furthermore, firms that take it into account appropriately adopt an optimal strategy. Usually, firms inflate the promotion criterion to offset the Peter principle effect, and the more important the transitory component is relative to total variation in ability, the larger the amount that the standard is inflated. The same logic applies to other situations. For example, it explains why movie sequels are worse than the original film on which they are based and why second visits to restaurants are less rewarding than the first.


In other words, firms know that you sometimes get lucky, and they set the promotion bar high on purpose. After your promotion you experience a "regression toward the mean", and your observed performance declines in quality, relative to your promotion-winning triumphs. But on average the promotions are still deserved. In other words, the Peter Principle will appear to be true in a well-functioning organization, even when promotions are handed out rationally.



My take: Lazear offers a characteristically nice demonstration of a clever idea. Behavioral factors may skew promotions in less efficient directions, but they will not overturn the central argument. People often overweight recent observations, but of course worker skill levels change through time. It is not obviously inappropriate to weight some observations more than others. Furthermore if people overvalue first impressions as well, the two behavioral effects may cancel to some degree.



Here is an earlier version of Lazear's paper. Thanks to Eric Crampton for the pointer.


very interesting.

Tuesday, February 17, 2004

FOXNews.com - Politics - Ashcroft Sued by Terror Case Prosecutor

Interesting - Ashcroft Sued by Terror Case Prosecutor


An assistant U.S. attorney involved in prosecuting a high-profile terrorism case is suing Attorney General John Ashcroft (search), alleging he and his agents at the Justice Department are retaliating against him for his testimony to the Senate Finance Committee.

I wanted this thing for years, I just did not know its name.

from Techdirt: More MiniPCs - But You Still Can't Buy One



Just last week we were talking about all those companies building very cool looking fully powered handheld computers that you couldn't buy, and now there's one more to add to the list. Paul Allen is backing a company called FlipStart, which is working on yet another different miniPC design. This one basically just looks like a laptop that has been shrunk down to a quarter of the size and fits in your palm. It definitely looks cool, but does anyone else remember that IBM sold machines that looked exactly like this in Japan years ago? At the time, the theory was that not enough people really wanted such a tiny PC in the US. Have times changed? In the meantime, this is yet another company that has announced a cool looking miniPC, but hasn't actually sold one yet. Will any of these products ever hit the market?


From The Guardian article referred to by Techdirt

Antelope is taking a different approach. The company is a spin-off from IBM's research into wearable computers and head-up displays for the military, and rather than a miniature notebook PC, it is producing a modular system. At the heart of it is the Antelope core; a 3in by 5in black box with a bronze top that looks like a giant version of the Crusoe chip inside it. This is a complete Windows XP PC with a custom connector on it. Plug it into the desktop dock and you get a PC with the standard complement of ports and sockets. When you're ready to leave your desk, pop it in your pocket or plug it into the rugged shell which has a 6in touch screen and cursor buttons, with a grip so you can use it in one hand (so it is ideal for engineers and field workers).

There are plans for other shells that will let you plug the core into a laptop or the dashboard of your car, but Antelope is hoping that companies will come up with their own shells, too.


Of the few companies described in that article Antelope seemed like the closest to my ideal. I also did not realize I can buy wireless network attached servers

a Network Attached Storage file server with built-in Wi-Fi (Sony and Linksys both have devices, though not yet for the UK market


I found this device from Vendotto, and could not find anything by linksys or sony.










Shake it like a Polaroid picture

interesting note via blogdex - Should you shake Polaroid pictures “like a Polaroid picture" ?
Question :

Ever since the song "Hey Ya" by Andre 3000 of Outkast came out, everyone is shaking their Polaroid pictures. I have always been told that you should not shake a Polaroid picture, but I'm having a hard time convincing those around me that this is true. What is the answer?

Answer:

The short answer is no, you don't have to (and shouldn't) "shake it like a Polaroid picture."


Shaking or waving a Polaroid picture to help the development process originated in the early days of peel-apart film. After peeling the negative, the image needed to dry before it could be handled, so waving the photo helped it to dry more quickly.


When using the integral films (600, Spectra, 500, SX-70/Time-Zero, i-Zone) that are used in our most popular current camera models (Polaroid One, OneStep, JoyCam, etc.), the image develops and dries behind a clear plastic window and never touches the air, so shaking or waving has no effect.


In fact, shaking or waving can actually damage the image. Rapid movement during development can cause portions of the film to separate prematurely, or can cause "blobs" in the picture.


The best way to ensure a perfectly developed image is to simply lay the picture on a flat surface immediately after it exits the camera. Shield it from the wind and avoid bending, twisting, or otherwise disturbing it during development.


good to know.

The new fax - Japanese Teen Girls Text-Message in Elaborate "Girl Talk" Code



From Japan.com: The abbreviated language of cell phone messaging has gone beyond smileys among Japanese teenage girls. They're now using something of a secret language called "gyaru-moji," with words replaced by symbols and special characters. ("Gyaru" means gal in English; "moji" means letter or text.) Japan.com writer Arjen van Blokland describes it as "a mixture of Japanese syllables, numbers, mathematical symbols and Greek characters" that "resembles hieroglyphics."


Recently a new trend to send personalized messages was introduced by these girls. They do not send emails to friends anymore. Instead, they jot down their message on a piece of paper, take a photo of it and send it as a picture message to their friends.
-- by way of Japan Media Review


except... is not this just a more hi-tech way of faxing? I friend of mine from college used to fax letters to her family in Taiwan in the early 90's before email became wide accessible. It was an economical blend of the speed of email and personal expression. I guess the new thing here is that they are using new codewords to communicate - but is not that just a limitation of the medium, i.e. current phones? Perhaps we should bring stenography back into schools as a subject?

Do not perish that last thought. For years I have been wishing I knewhow to stenograph in order to take better notes in meetings and phone conferences.

ps. why does almost every article about Japan deal with "japanese teen girls" ?

Who You Calling "Arab"? - Considering today's New York Times story about Arabs. I mean, Muslims. No, brownish people from the Middle East. Or possibly Sout

Excellent points - Who You Calling "Arab"? - Considering today's New York Times story about Arabs. I mean, Muslims. No, brownish people from the Middle East. Or possibly South Asia.


When reading the abovementioned article in NYT today, I had a similar level of confusion. Slate's Today's papers also had a brief reference to the confusion. Perhaps we can assume that for most Americans the Arab-Americans *are* a relatively monolithic block and so the paper is justified in bundling their political contributions into one article serving as an overview. However, should NYT really be playing to the lowest common denominator of ignorance that would put all of these people together as "those Arabs"?

The article honestly alludes to the fact that these "arab" donors are as diverse a group as "americans" are. Some are first-generation immigrants and many are not. Some are Arabic and Muslim, some are Arabic and non-Muslim, some are Muslim and not Arabic, and some are merely from countries that are Muslim. As an example, if an Egyptian Copt family donated money to the Bush campaign they would also be included in the article by virtue of being from Egypt. The proper name for the article should thus be, "non-Jewish people with origins in the Middle East donate to Bush's campaign" ... I guess that would be a lot less exciting as a headline.

Putin Watches As Missile Launch Fails

Putin Watches As Missile Launch Fails


Putin, who is expected to easily win the presidential election March 14, went to the Barents Sea on board the giant Arkhangelsk submarine to observe maneuvers set to involve numerous missile launches and flights of strategic bombers.


But the ambitious exercise hit a snag when a ballistic missile a missile that is launched on a high-arch trajectory to hit a designated target failed to blast off as scheduled from another submarine, the Novomoskovsk, a government official said on condition of anonymity. The official said the automatic safety system blocked the launch for unspecified reasons.


I am glad the failsafe system works. I am worried about what would happen if the two missiles aimed at Kamchatka missed, and landed in Alaska? Japan?

On a different note - does US have a law or provision for similar failures to be announced? How do we know if our strategic missiles work any better?

From ventureblog - Standardizing The Comparison Slide



On Friday I was meeting with an interesting company. Their presentation was going along well and then we hit the dreaded competition comparison slide. You know the slide -- it is the one where you try to make clear that your product/service is better than everything else in the market. I've written about the competition slide before. It comes in a bunch of different forms but it is usually a list of attributes that your product/service has and that your competition's product/service doesn't. The degree to which an attribute is held by your product or the competition's product is indicated by a number of different codes -- circles, checks, percentages, even smiley faces. And, of course, there are always degrees of attribute compliance. Circles become circles, half circles, shaded circles, empty circles, full circles. Checks become check plusses, check minuses, half checks, grey checks. Percentages are used with incredible granularity (how does a product have 36% of an attribute?).


Given the myriad of codes used to depict product superiority, I find myself spending the first minutes of any such slide just trying to crack the code. Is a grey check better than a black check? Is a black circle faster/slower/bigger/smaller/cheaper/deeper/whatever than a red circle? (hint: look at the presenter's product column -- whatever they've got is what you want). In this era of standardization, I think it is time to standardize the comparison slide. And, for the sake of debate (we all know it takes a long time to reach consensus on any standard), I propose a standard -- the Consumer Reports ranking system. We've all gone to Consumer Reports when shopping for a car or a vacuum cleaner or a stroller. We've all seen the Consumer Reports bubbles. And we all know that it is better to have a full red circle than a full black circle (let me review -- the good folks at the Consumers Union have determined that the full red circle is excellent, the half red circle is very good, the empty circle is good, the half black circle is fair, and the full black circle is poor -- got it?). OK, truth be told, I don't care if you adopt my standard. But I sure would like some standard, so that I can focus on the information in your presentation, and not on deciphering your code.


I wonder what organization one would submit this specification to.

$15-billion bribe aims to free Russian CEO

$15-billion bribe aims to free Russian CEO


That's a serious amount of money that top shareholders of Yukos have reportedly offerred the Russian government for the release of its top people currently in russian prison on various charges:

Worried about their boss's health and desperate to free him from jail, top shareholders in Russian oil giant Yukos made the Kremlin a $15-billion offer on Monday: Let him go and we will give you the company.

After months of proclaiming Mikhail Khodorkovsky's innocence of the charges against him, and their chief executive officer's willingness to tough it out in prison, several big Yukos shareholders blinked and proposed what may amount to one of the largest bribes ever.


In exchange for the release of Mr. Khodorkovsky and fellow top shareholder Platon Lebedev, who are in jail facing fraud and tax-evasion charges, the Kremlin would receive the 44-per-cent stake in Yukos that is controlled by holding company Group Menatep. The offer's value was estimated Monday at $14.6-billion (U.S.).


I doubt this is something the government will even acknowledge as being oferred, but it is an interesting PR play.

Monday, February 16, 2004

IBM delivers autonomic tools | CNET News.com

Interesting - IBM delivers autonomic tools


The Autonomic Computing Toolkit brings together the latest versions of IBM's tools, including updates to its software for resolving and monitoring problems. It also contains tools designed to streamline software installation in complex situations. In addition, it has created a system for building a viewing console that can be used across several applications.


Going to have to check it out.

Requiem for the V-Chip - A relic of the last battle over indecency on TV. By Thomas?Hazlett

many good points. - Requiem for the V-Chip - A relic of the last battle over indecency on TV

Sunday, February 15, 2004

Betrayed by the Game

Good article - Betrayed by the Game

Friday, February 13, 2004

Neat Trick, or how to make $145million for the government


from Slate Moneybox

Clinton Treasury Secretary Robert Rubin, who had enormous influence on policy and decades of experience on Wall Street trading desks, mastered this subtle art. In his memoir, In an Uncertain World, Rubin recounts an episode in June 1998. The yen was at 141 to the dollar, the weakest the yen had been in eight years, and a bit too weak for the administration's liking. Rubin told the Senate Finance Committee that intervention is "a temporary tool, not a fundamental solution." Traders took the phrase as a sign that the administration wouldn't intervene to support the yen. And so in the next few days the yen slipped further, to about 146 to the dollar. Then, without warning, the U.S. government bought $2 billion in yen. "Currency traders were caught by surprise and the exchange rate moved all the way back to 136 yen to the dollar," Rubin recounted.


It is worth noting that this little game netted US government over $145million, leaving a couple million for expenses related to the trade. How so?

$2,000,000,000 * 146 yen/$ = 292,000,000,000.00 yen.

292,000,000,000.00 yen / 136 yen/$ = $2,147,058,823.53

the difference: $147,058,823.53 - expenses.


Not bad for a government official.


Friday, February 06, 2004

Op-Ed Columnist: Get Me Rewrite!

Good Krugman column today - Get Me Rewrite!


Although it is pretty clear that NYT would be even better off just hiring Marshall in the first place. He is (IMO) partisan, but also honest, smart, and determined.