Diving in La Paz

Went Diving here in Feb/March 2003. Was a great place, they have good package deals that gets you several days of diving. We kept wanting to see Hammerhead Sharks... but never was able to see them.

Club Cantamar

Techno and Trance on the Internet

I've found several internet radio stations.

DJDoboy.com - 128k

ETS-global.org 160k

Digitally Imported - 96k (free) - 128k+ (pay)

DJ Virgen - SD DJ

OK, I found something new. LAUNCHcast by Yahoo. Lets you listen to radio stations, create your own, and its all GOOD music.

Palm vs. Windows CE/Pocket PC, Round Four

I found this interesting review by someone that shares my experience with Palm vs Pocket PC.

For the fourth time, I've purchased a Windows CE device to see whether the Windows CE world is poised to crush my trusty old Palm devices. For the fourth time, I say, "No, it's not," but it's getting closer.

A few years ago, I purchased a couple of Windows CE-based Casio palmtops and an early Compaq iPAQ. Unfortunately, the devices ran an OS that looked like squashed Windows on a small screen, chewed up battery power like it was going out of style, lacked a useful software set, and required about 10 times the CPU power of a Palm device to do the same job. So you'd think that I'd have sworn off Windows CE (or Pocket PC 2002, as it's now called to try to escape the bad reputation of past OSs) platforms for good. Which brings us to device number four.

I recently found a nifty new productivity tool that combines a phone with a PDA--the Samsung i700. Although the device itself isn't all that exciting, I was intrigued by its ability to retrieve my email through its cell phone link and do so at 1xRTT speeds--144Kbps in theory, about 64Kbps in practice. Even better, Verizon, my cell phone service provider, serves the out-of-the-way rural area that I live in. It sounded like the perfect tool for a traveling guy like me.

The only catch is that the device runs Pocket PC 2002. But what the heck, I thought, who cares what OS it runs? After all, I'm just using it as a wireless email appliance that happens to have a cell phone attached. So, I thought I'd just ignore the PDA abilities. (Of course, in the end I couldn't ignore the PDA part: a big bright colorful transflective screen, a 200MHz processor, and 64MB of RAM--take that out and see what it can do.)

The first thing I noticed is that the Pocket PC platform has a desktop that's too sprawling to fit on a PDA, even a big PDA. At least I can no longer say that I'm being forced to view Windows through a very small pane. As with other Windows CE/Pocket PC devices I've tested, the memory management is still irritating; believe it or not, when you start a program, the application remains in memory until you either perform a soft reset or you navigate to Windows CE's Control Panel Memory applet, which lets you shut down the program. I'm not joking when I say that most experienced Pocket PC users agree that your best bet with Pocket PC PDA/phone combinations is to do a soft reset every night to keep the system reliable. Sounds a mite too Windows 9x-ish for my taste. I can't remember the last time that I ran out of memory or had to reset my Palm device.

Regarding power, the i700 has both a main battery and a piggyback battery that should give you 2 to 3 days worth of work, assuming you don't spend much time on the phone and are careful about shutting off the backlight. In contrast, I've run my Palm Tungsten-T system for about a week without a recharge, and my old Palm 505 once ran 3 weeks without a recharge. Then again, the Palm 505 ran on a 33MHz processor and had only 8MB of RAM, so the comparison might seem unfair--but consider that Pocket PC could never run on that slow of a processor and small amount of RAM.

Despite the fast processor, the OS is sluggish compared with the Palm. I found an interesting benchmark of sorts in the Handmark Software implementation of the Monopoly board game, which the company sells for both Palm and Pocket PC platforms. The Palm version runs quite quickly on its 33MHz processor; the Pocket PC version is quite laconic by comparison. Finally, the included synchronization software, ActiveSync, is just as irritating as always, disconnecting your host computer from your PDA at random times even though the PDA is in the cradle and the USB connector is snugly in place.

But some Pocket PC features have changed for the better. My first pleasant surprise happened when I wanted to transfer some of my Palm address book to the i700, so I attempted to beam a contact, and it worked. And my earlier complaint about the lack of Pocket PC software no longer applies. Yes, the Palm world still has more software, but Pocket PC is gaining on Palm. And that, in the end analysis, is why I think that Palm might finally find itself losing ground to Pocket PC. Now that third-party innovators have joined the Pocket PC party, anything can happen--inexpensive useful add-ons, lots of freeware, who knows. But until that happens, I'm going to keep carrying my Palm--and the Pocket PC device.

SpamNet for filtering your OUTLOOK mailbox

People are always asking me what I do to fight the deluge of SPAM that attacks all of our inboxes. One thing I do is not use real email addresses when signing up online. But in the 12 months that I have done this, not one company has sold my addresses. The spam that fills my inbox is totally unsolicited, or is gathered from an email source older than 12 months ago. To fight this, I have found the SPAMNET program works AWSOME!. It will remove 95% of the SPAM out of your inbox. At which point for items it misses, you can manually 'block' it, and train it to block those messages as well.

In the off chance SPAMNET identifies something from someone I really want to receive email from, I periodically review the SPAM folder it creates, and when I find that it has FALSELY identified a email as spam, I am able to train it NOT to filter that email. Quite nice. Sign up using my info and get a special price.

For every friend of yours who subscribes using your
referral code noted above, you receive* one month FREE of the SpamNet service
and they lock in the special $1.99 USD per month price up to a year from time of
purchase. Just give them your referral code and send them to href="http://www.cloudmark.com/spamnet"> size=2>www.cloudmark.com/spamnet . During the
subscription process they will be asked to enter that referral code at which
point they will be awarded the special price you gave them and you will be
recognized for the referral and get *one month FREE.

Moving to OpenSourceHost.com

I have now completed my move to open source host (OSH). The move took several hours because I backed up the databases and moved them over. Here are my notes.

Gallery

Backing this up was easy, you need to zip up the albums directory, and the two default style sheets in the gallery/css directory. Make sure the permissions are correct on the new albums directory, especially if you are switching versions.

Had to go in and run the configure.sh / secure.sh scripts to reset some things, specifically the border color (which is not in the CSS) and the admin password.

Drupal

This was a little trickier.
Exporting Backup

mysqldump --opt -u user_name -p database_name > tfiske_drupal.sql

mysqldump --opt -udrupaltfiske -pdrpwtfiske drupal_tfiske > drupal_tfiske-opt.sql

There is more but I need to get onto steve's box to confirm. The --opt flag is needed to make this an update script, rather than something that creates the database from scratch. Remove --opt if you are going to start with an empty database.

Import Restore

mysql -u user_name -p database_name < hacked.sql

Note that there is no password, the mysql utility asks for it. The hacked.sql script was modified from the export to drop 3 tables that the upgrade portion created.

Update

Modify the update.php file to allow it to be in update mode:

// Disable access checking?
$access_check = 0;

Change back to a 1 after you are done.

Then goto your site http://www.corvettepowr.com/update.php . Be sure to select the version your database _WAS_ on. Its very important to double check the date that it tells you since it most likely guessed wrong (read I spent alot of time wasted because the wrong date was chosen). This will go through and produce a report telling you how it altered your tables... this is where i figured out that I needed to drop some tables. If it doesn't work right, you just go back to the import/restore part and re-import.

Change update.php back to the 1, and your set. 🙂

Some Links

Forum on backups

MySQL site docs on backups

mysqldump docs

mysqladmin

FIRESTORM 2003

Leave it to the news to give it a name like FIRESTORM 2003... drum roll please...

Everyone I know is OK, and it looks like the fires have spared my immediate circle of friends. There are a couple people I know in Scripps Ranch that I have not been able to contact to see how they are doing. The fire went right down their street, so I don't know if they got hit by the fire.

Lorinda and Jocie both had homes where the fire was in their yard! The houses survived, and both families are back in their houses.

The smoke is pretty bad still, and I think all of us are going to take some time before we are breathing right. I'm off to help Lorinda move her family back into their house, and try and get some semblence of normalcy started. Please email me if you have useful links like FEMA or other support information. Most of her neighbors lost their houses, and will need aid. 🙁

Oh... I know several people have asked about my old home in Scripps Ranch, the fire went right up Semillion and Avenita Magnifica... You turn onto Semillion to get to my house off of Pomerado.. Instead of turning right to goto my house, you turn left to get to Avenita Magnifica. The fire stayed on the left hand (south) side of Semillion and did not cross and head towards the houses in my old neighborhood. I know Jaime's parents are back in their house, and I have yet to chat with Jim and Carla to see how they are doing. An update later.

Some pictures to distract you

Drawing of the effected areas - As of 10/28/2003


NASA picture from space

Tell No Tales Pirate Party

Due to the fire, and my servers still being down (they are hosted in a still active area of town). I will have pictures up wednesday or thursday.

Pictures are UP!

Windows iTunes – When Hell Freezes Over

iTunes for windows is now available from the Apple.com site. Besides being a great MP3 player, and ripper, it does sync well with the Apple iPod. This player also allows you to legally purchase MP3's of your favorite artists for $.99 at a single click.

Surprisingly I also found that other people in my building showed up in my list of available music. I could 'listen' to their music as well as my own. Interesting concept.

New Digital Camera

I have finally decided to replace the ole' S110 with a new Digital Camera. I take enough pictures that its time to get one with better resolution.

Here is the link to the manufactures, click any one and get a list of
available cameras, in each camera description is a link to a full review
and within each review is a link to sample photos with that
camera.

http://www.dpreview.com/reviews/specs.asp

Here are a couple of suggestions based on the requirements you
mentioned...

Canon S400 - Very small (3.4 x 2.2 x 1.1 in) 4mp, camera, very
compact
http://www.dpreview.com/reviews/canons400/

Olympus C5050 - small (4.5 x 2.7 x 3.2 in) 5mp, very good camera,
fast
http://www.dpreview.com/reviews/olympusc5050z/

Nikon Coolpix 5700 - medium (4.3 x 3 x 4 in) 5mp, very good camera, best
zoom
http://www.dpreview.com/reviews/nikoncp5700/

Anyway, there are some ideas.  If you like zoom the Nikon is best,
if you like fast as in taking pictures quickly one after the other
without waiting, the Olympus is best; and if you want the absolute
smallest feature rich camera the canon is best.

In any event, you can poke around on the dpreview.com web site and find
out what you need to about the various cameras available.  Hope it
helps.

There’s a Sucker Born in Every Medial Prefrontal Cortex

There's a Sucker Born in Every Medial Prefrontal
Cortex
size=-1>By CLIVE THOMPSON
alt="" src="http://graphics7.nytimes.com/images/misc/spacer.gif"
width=1>
Published: October
26, 2003

src="http://graphics7.nytimes.com/images/dropcap/w.gif" width=46 align=left
border=0>hen he isn't pondering the inner workings of the mind, Read
Montague, a 43-year-old neuroscientist at Baylor College of Medicine, has been
known to contemplate the other mysteries of life: for instance, the Pepsi
Challenge. In the series of TV commercials from the 70's and 80's that pitted
Coke against Pepsi in a blind taste test, Pepsi was usually the winner. So why,
Montague asked himself not long ago, did Coke appeal so strongly to so many
people if it didn't taste any better?

Over several months this past summer, Montague set to work looking for a
scientifically convincing answer. He assembled a group of test subjects and,
while monitoring their brain activity with an M.R.I. machine, recreated the
Pepsi Challenge. His results confirmed those of the TV campaign: Pepsi tended to
produce a stronger response than Coke in the brain's ventral putamen, a region
thought to process feelings of reward. (Monkeys, for instance, exhibit activity
in the ventral putamen when they receive food for completing a task.) Indeed, in
people who preferred Pepsi, the ventral putamen was five times as active when
drinking Pepsi than that of Coke fans when drinking Coke.

In the real world, of course, taste is not everything. So Montague tried to
gauge the appeal of Coke's image, its ''brand influence,'' by repeating the
experiment with a small variation: this time, he announced which of the sample
tastes were Coke. The outcome was remarkable: almost all the subjects said they
preferred Coke. What's more, the brain activity of the subjects was now
different. There was also activity in the medial prefrontal cortex, an area of
the brain that scientists say governs high-level cognitive powers. Apparently,
the subjects were meditating in a more sophisticated way on the taste of Coke,
allowing memories and other impressions of the drink -- in a word, its brand --
to shape their preference.

Pepsi, crucially, couldn't achieve the same effect. When Montague reversed
the situation, announcing which tastes were of Pepsi, far fewer of the subjects
said they preferred Pepsi. Montague was impressed: he had demonstrated, with a
fair degree of neuroscientific precision, the special power of Coke's brand to
override our taste buds.

Measuring brand influence might seem like an unusual activity for a
neuroscientist, but Montague is just one of a growing breed of researchers who
are applying the methods of the neurology lab to the questions of the
advertising world. Some of these researchers, like Montague, are purely academic
in focus, studying the consumer mind out of intellectual curiosity, with no
corporate support. Increasingly, though, there are others -- like several of the
researchers at the Mind of the Market Laboratory at Harvard Business School --
who work as full-fledged ''neuromarketers,'' conducting brain research with the
help of corporate financing and sharing their results with their sponsors. This
summer, when it opened its doors for business, the BrightHouse Institute for
Thought Sciences in Atlanta became the first neuromarketing firm to boast a
Fortune 500 consumer-products company as a client. (The client's identity is
currently a secret.) The institute will scan the brains of a representative
sample of its client's prospective customers, assess their reactions to the
company's products and advertising and tweak the corporate image accordingly.

Not long ago, M.R.I. machines were used solely for medical purposes, like
diagnosing strokes or discovering tumors. But neuroscience has reached a sort of
cocky adolescence; it has become routine to read about researchers tackling
every subject under the sun, placing test subjects in M.R.I. machines and
analyzing their brain activity as they do everything from making moral choices
to praying to appreciating beauty. Paul C. Lauterbur, a chemist who shared this
year's Nobel Prize in medicine for his contribution in the early 70's to the
invention of the M.R.I. machine, notes how novel the uses of his invention have
become. ''Things are getting a lot more subtle than we'd ever thought,'' he
says. It seems only natural that the commercial world has finally caught on.
''You don't have to be a genius to say, 'My God, if you combine making the can
red with making it less sweet, you can measure this in a scanner and see the
result,''' Montague says. ''If I were Pepsi, I'd go in there and I'd start
scanning people.''

The neuroscience wing at Emory University Hospital in Atlanta is the
epicenter of the neuromarketing world. Like most medical wards, it is filled
with an air of quiet, antiseptic tension. On a recent visit, in the hallway
outside an M.R.I. room, a patient milled around in a light blue paper gown. A
doctor on a bench flipped through a clipboard and talked in soothing tones to a
man in glasses, a young woman anxiously clutching his arm.

It was not a place where you would expect to encounter slick marketing
research. And when Justine Meaux, a research scientist for the BrightHouse
Institute, came out to greet me, she did seem strangely out of place. Clicking
along in strappy sandals, with a tight sleeveless top and purple toenail polish,
she looked more like a chic TV producer than a neuroscientist, which she is. Her
specialty, as she explained, is ''the neural dynamics of the perception and
production of rhythmic sensorimotor patterns'' -- though these days she spends
her professional life thinking about shopping. ''I'm really getting into reading
all this business stuff now, learning about campaigns, branding,'' she said,
leading me down the hallway to the M.R.I. chamber that the Institute uses. Three
years ago, after earning her Ph.D., she decided she wanted to apply brain
scanning to everyday problems and was intrigued by marketing as a ''practical
application of psychology,'' as she put it. She told me that she admired the
'' value="Intel Corporation" />Intel Inside'' advertising campaign, with its
TV spots showing dancing men in body suits. ''Intel actually branded the inside
of a computer,'' she marveled. ''They took the most abstract thing you can
imagine and figured out a way to make people identify with it.''

When we reached the M.R.I. control room, Clint Kilts, the scientific director
of the BrightHouse Institute, was fiddling away at a computer keyboard. A
professor in the department of psychiatry and behavioral sciences at Emory,
Kilts began working with Meaux in 2001. Meaux had learned that Kilts and a group
of marketers were founding the BrightHouse Institute, and she joined their team,
becoming perhaps the world's first full-time neuromarketer. Kilts is confident
that there will soon be room for other full-time careers in neuromarketing.
''You will actually see this being part of the decision-making process, up and
down the company,'' he predicted. ''You are going to see more large companies
that will have neuroscience divisions.''

The BrightHouse Institute's techniques are based, in part, on an experiment
that Kilts conducted earlier this year. He gathered a group of test subjects and
asked them to look at a series of commercial products, rating how strongly they
liked or disliked them. Then, while scanning their brains in an M.R.I. machine,
he showed them pictures of the products again. When Kilts looked at the images
of their brains, he was struck by one particular result: whenever a subject saw
a product he had identified as one he truly loved -- something that might prompt
him to say, ''That's just so me!'' -- his brain would show increased activity in
the medial prefrontal cortex.

Kilts was excited, for he knew that this region of the brain is commonly
associated with our sense of self. Patients with damage in this area of the
brain, for instance, often undergo drastic changes in personality; in one famous
case, a mild-mannered 19th-century railworker named Phineas Gage abruptly became
belligerent after an accident that destroyed his medial prefrontal cortex. More
recently, M.R.I. studies have found increased activity in this region when
people are asked if adjectives like ''trustworthy'' or ''courageous'' apply to
them. When the medial prefrontal cortex fires, your brain seems to be engaging,
in some manner, with what sort of person you are. If it fires when you see a
particular product, Kilts argues, it's most likely to be because the product
clicks with your self-image.

This result provided the BrightHouse Institute with an elegant tool for
testing marketing campaigns and brands. An immediate, intuitive bond between
consumer and product is one that every company dreams of making. ''If you like
Chevy trucks, it's because that has become the larger gestalt of who you
self-attribute as,'' Kilts said, using psychology-speak. ''You're a Chevy guy.''
With the help of neuromarketers, he claims, companies can now know with
certainty whether their products are making that special connection.

To demonstrate their technique, Kilts and Meaux offered to stick my head in
the M.R.I. machine. They laid me down headfirst in the coffinlike cylinder and
scurried out to the observation room. ''Here's what I want you to do,'' Meaux
said, her voice crackling over an intercom. ''I'm going to show you a bunch of
images of products and activities -- and I want you to picture yourself using
them. Don't think about whether you like them or not. Just put yourself in the
scene.''

I peered up into a mirror positioned over my head, and she began flashing
pictures. There were images of a Hummer, a mountain bike, a can of Pepsi. Then a
Lincoln Navigator, Martha Stewart, a game of basketball and dozens more
snapshots of everyday consumption. I imagined piloting the Hummer off-road,
playing a game of pickup basketball, swigging the Pepsi. (I was less sure what
to do with Martha Stewart.)

After about 15 minutes, Kilts pulled me out, and I joined him at a bank of
computers. ''Look here,'' he said, pointing to a screen that showed an image of
a brain in cross sections. He pointed to a bright yellow spot on the right side,
in the somatosensory cortex, an area that shows activity when you emulate
sensory experience -- as when I imagined what it would be like to drive a
Hummer. If a marketer finds that his product is producing a response in this
region of the brain, he can conclude that he has not made the immediate,
instinctive sell: even if a consumer has a positive attitude toward the product,
if he has to mentally ''try it out,'' he isn't instantly identifying with it.

Kilts stabbed his finger at another glowing yellow dot near the top of the
brain. It was the magic spot -- the medial prefrontal cortex. If that area is
firing, a consumer isn't deliberating, he said: he's itching to buy. ''At that
point, it's intuitive. You say: 'I'm going to do it. I want it.' ''

he
consuming public has long had an uneasy feeling about scientists who dabble in
marketing. In 1957, Vance Packard wrote ''The Hidden Persuaders,'' a book about
marketing that featured harsh criticism of ''psychology professors turned
merchandisers.'' Marketers, Packard worried, were using the resources of the
social sciences to understand consumers' irrational and emotional urges -- the
better to trick them into increased product consumption. In rabble-rousing
prose, Packard warned about subliminal advertising and cited a famous (though,
it turned out, bogus) study about a movie theater that inserted into a film
several split-second frames urging patrons to drink Coke.

In truth, marketers only wish they had that much control. If anything,
corporations tend to look slightly askance at their admen, because there's not
much convincing evidence that advertising works as well as promised. John
Wanamaker, a department-store magnate in the late 19th century, famously quipped
that half the money he spent on advertising was wasted, but that he didn't know
which half. In their quest for a more respectable methodology -- or perhaps more
important, the appearance of one -- admen have plundered one scientific
technique after another. Demographic studies have profiled customers by
analyzing their age, race or neighborhood; telephone surveys have queried
semi-randomly selected strangers to see how the public at large viewed a
company's product.

Advertising's main tool, of course, has been the focus group, a classic
technique of social science. Marketers in the United States spent more than $1
billion last year on focus groups, the results of which guided about $120
billion in advertising. But focus groups are plagued by a basic flaw of human
psychology: people often do not know their own minds. Joey Reiman is the C.E.O.
of BrightHouse, an Atlanta marketing firm, and a founding partner in the
BrightHouse Institute; over years of producing marketing concepts for companies
like value="Coca-Cola Company" />Coca-Cola and Red Lobster, he has come to the
conclusion that focus groups are ultimately less about gathering hard data and
more about pretending to have concrete justifications for a hugely expensive ad
campaign. ''The sad fact is, people tell you what you want to hear, not what
they really think,'' he says. ''Sometimes there's a focus-group bully, a
loudmouth who's so insistent about his opinion that it influences everyone else.
This is not a science; it's a circus.''

In contrast, M.R.I. scanning offers the promise of concrete facts -- an
unbiased glimpse at a consumer's mind in action. To an M.R.I. machine, you
cannot misrepresent your responses. Your medial prefrontal cortex will start
firing when you see something you adore, even if you claim not to like it.
''Let's say I show you Playboy,'' Kilts says, ''and you go, 'Oh, no, no, no!'
Really? We could tell you actually like it.''

Other neuromarketers have demonstrated that we react to products in ways that
we may not be entirely conscious of. This year, for instance, scientists working
with value="DaimlerChrysler AG" />DaimlerChrysler scanned the brains of a
number of men as they looked at pictures of cars and rated them for
attractiveness. The scientists found that the most popular vehicles -- the idsrc="other-OTC" value="PSEPF"> />Porsche- and Ferrari-style sports cars -- triggered activity in a
section of the brain called the fusiform face area, which governs facial
recognition. ''They were reminded of faces when they looked at the cars,'' says
Henrik Walter, a psychiatrist at the University of Ulm in Germany who ran the
study. ''The lights of the cars look a little like eyes.''

Neuromarketing may also be able to suss out the distinction between
advertisements that people merely like and those that are actually effective --
a difference that can be hard to detect from a focus group. A neuromarketing
study in Australia, for instance, demonstrated that supershort, MTV-style jump
cuts -- indeed, any scenes shorter than two seconds -- aren't as likely to enter
the long-term memory of viewers, however bracing or aesthetically pleasing they
may be.

Still, many scientists are skeptical of neuromarketing. The brain, critics
point out, is still mostly an enigma; just because we can see neurons firing
doesn't mean we always know what the mind is doing. For all their admirable
successes, neuroscientists do not yet have an agreed-upon map of the brain. ''I
keep joking that I could do this idsrc="NYSE" value="Gucci Group NV" />Gucci shoes study, where I'd show
people shoes I think are beautiful, and see whether women like them,'' says
Elizabeth Phelps, a professor of psychology at New York University. ''And I'll
see activity in the brain. I definitely will. But it's not like I've found 'the
shoe center of the brain.''' James Twitchell, a professor of advertising at the
University of Florida, wonders whether neuromarketing isn't just the next stage
of scientific pretense on the part of the advertising industry. ''Remember, you
have to ask the client for millions, millions of dollars,'' he says. ''So you
have to say: 'Trust me. We have data. We've done these neurotests. Go with us,
we know what we're doing.''' Twitchell recently attended an advertising
conference where a marketer discussed neuromarketing. The entire room sat in awe
as the speaker suggested that neuroscience will finally crack open the mind of
the shopper. ''A lot of it is just garbage,'' he says, ''but the garbage is so
powerful.''

In response to his critics, Kilts plans to publish the BrightHouse research
in an accredited academic journal. He insisted to me that his primary allegiance
is to science; BrightHouse's techniques are ''business done in the science
method,'' he said, ''not science done in the business method.'' And as he sat at
his computer, calling up a 3-D picture of a brain, it was hard not to be struck,
at the very least, by the seriousness of his passion. There, on the screen, was
the medial prefrontal cortex, juggling our conscious thinking. There was the
amygdala, governing our fears, buried deep in the brain. These are sights that
he said still inspire in him feelings of wonder. ''When you sit down and you're
watching -- for the first time in the history of mankind -- how we process
complex primary emotions like anger, it's amazing,'' he said. ''You're like,
there, look at that: that's anger, that's pleasure. When you see that roll off
the workstation, you never look back.'' You just keep going, it seems, until you
hit Madison Avenue.

Original Article on nytimes.com