America is full of fat people, right? Well, I've been subject to the same delusion. I think it took hold firmly in 1993, when I drove from Waterloo, Ontario, to Milwaukee, for a few days at GenCon, and for some reason was struck by seeing all the fat people in the streets of Milwaukee. Was that simply observer bias? Who knows. Now I live in California and it's very clear to me that Californians don't tend to be obese, but I still had the meme in my head that "Americans are fat" even if California big cities might be some fitness-oriented exception.

Well, I'm encouraged to hear I might be wrong. It seems this is a common myth, and that other well-off countries have similar weight profiles, as explained by Paul Campos in a TCS interview (I keep reading TCS for exactly this kind of myth-busting material though many articles are more boring). I wish the article provided links supporting the claims, because there were a few quite interesting tidbits from the article:

• The CDC had previously estimated 400,000 deaths from obesity in 2004 but has recently had to revise that figure significantly downward (following CDC links, I see the surgeon general now says an "estimated 300,000 deaths[/year] may be attributable to obesity").
  
• The classification of "overweight" in children is defined as the heaviest 15% of children for a particular age cohort. That would mean nearly a million children are labelled overweight because they're defined that way. According to that methodology, 15% of children in a third-world, famine-wracked country would also be overweight. I can't find evidence at CDC of this methodology but there was a study showing that 16% of teens had been found to be overweight [*].
  
• There's no strong evidence that Type 2 diabetes -- one of the diseases justifying the labeling of obesity as an epidemic -- has increased, let alone having increased due to weight factors.
• The governor of Arkansas has written legislation proposing that body weight index be part of kids' report cards. If I were in Arkansas I would write him personally to indicate my disgust at such misguided and potentially harmful interference in the family.
  

One nasty statistic I picked up myself from the surgeon general's note is that "Overweight adolescents have a 70% chance of becoming overweight or obese adults". Yikes, oh no! But combine that with the claim that 60% of adults are overweight or obese, and a sensible person will see that so-called overweight adolescents would have only a slightly higher (if measurable) tendency than non-overweight adolescents to become overweight adults. Sigh.

I completed another long-term project this weekend -- though it was supposed to take only a couple months, I dawdled for over a year. It's a silk-mohair cardign, pattern and yarn chosen to show off some really marvelous Toulouse-Lautrec buttons I found in Vienna. More details and pictures can be found on my knitting gallery.

One of the projects I've been very involved in at OSAF has been the Cosmo project, an effort to take existing open-source WebDAV server technology and turn it into an easy-to-install, easy-to-administer "sharing server" for use by Chandler -- but also supporting standards that make it easy for any client to share too. Last week, Mitch did a demo with Cosmo in the background for the first time and it went smoothly. Thanks BCM!

I have completed a big long-term hobby project, and given it away (I think it was quite a surprise) so now I can reveal what I was working on: a king-size Amish-inspired wedding quilt. It's at the top of my crafts gallery right now if you want to take a look.

I needed to cancel a credit card I wasn't using, so I called up their toll-free line and tried to navigate their voice-activated automated menu. The automated system repeated all the choices ad nauseum, misheard me, and of course didn't have an option for "Cancel this goddamn card". I tried to say "Cancel card" and "Cancel" hoping that would either trigger some bonus feature or get me out of the system and talking to a human. I tried to say "Talk to representative". At some point the system offered the option of "Help me with something else" but when I said that it misunderstood me. Finally the system gave up on me -- literally said it couldn't seem to help me -- and that it would transfer me to a live person (thank-you).

I got some satisfaction, however, when they asked me why I was cancelling their card. "Because of the automated voice assistance system", I said -- and this was typed in without comment by the operator. I only felt the teensiest bit bad for lying.

I'm concerned about conferences with serious gender imbalances, much like Kathy Sierra and Shelley Powers with their recent posts on ETech. However, it's not ETech I'm thinking of.

To put it bluntly, a knitting conference can be hostile to men. The interaction style is touchy-feely -- women walk right up to strangers and fondle their knitted garments, invading personal space. Although some have said that men aren't discouraged from attending and in fact receive positive attention for being there, this can be of the form of "How nice it is to have men attending", "Is this knitting book for your mother?" (ref) and "I'm sure your wife will love this yarn.

All the instructors at knitting conferences are female, and one wonders if the mostly-female program committee could have something to do with that. In the expos or markets, one finds patterns for shawls, purses and female garments. Although some of the market vendors are male, they are clearly there as "booth bunnies", to attract women to look at the yarn. Some of the male vendors are even pressured to wear demeaning and ridiculous knitted vests.

What can we do about it? Well, we can be more equal in our language, for a start. Articles like this, although mentioning a few knitting men, are given titles like "The yarn is flying as more women discover the joys of knitting." Vendors like Habu textiles are taking a step in the right direction, offering stainless steel yarn, which is sure to appeal to men. Joe, a male knitter, has some other great ideas, such as making the act of knitting into more of a competition.

Researching Bali, found this:

The once beautiful Bale’s and other buildings have long since fell into a state of disrepair. It was as if the villagers had ‘locked-on’ to the ring of the till from the tourist dollar. This was more than evident when I had a short but sharp conversation with one of the souvenir sellers:

Tourist: I was here 18 years ago and I am surprised at the change in the village.

Seller: (Abrupt tone) What’s wrong with the village?. Now I have a motorbike. We have TV and electricity and... a phone!

It was my point precisely! Too many cultures in this world are decimated by the tourist dollar. Not only that, the intrusion of the modern world has an overall affect upon the social, religious and political aspects of that society. So much so that it literally forces the entire socio-cultural structure to undergo a metamorphosis in order for adaptation. Having said that, tourism is good for the state of the economy in any country. However, when it has a serious affect on the core culture of that country, then it becomes disadvantageous.

Much as I'd love to visit other countries and have them be all picturesque and unique, I can't begrudge a Balinese his motorbike, TV, electricity and phone. I wouldn't give up my conveniences to live in the style of my grandmother, nor would I appreciate pressure to maintain the religion of my ancestors. A tourist like this one would have to stop being a tourist (being one of those intrusions), go back to the country of his own ancestors, give up his own conveniences etc. for me to take those sentiments seriously.

Philippe pointed me to David Berlind's list of Technology's top ten failures. Two of them are directly about calendar standards, asking for the ability to schedule meetings and share calendars. Another two are email related. Good news, David, we're working on the calendaring standards, and calendar and email implementation issues for you, already. Sorry I can't -- personally -- help you with phones and printing problems just yet.

Who is your favorite Philosopher?

Ekr points out that the question "Who's your favorite philosopher" does not have an analogue in all fields. In highly technical fields, you wouldn't ask "Who's your favorite cryptographer" or "Who's your favorite biologist". OTOH it's common for musicians to talk about their favorite pianist or favorite composer, for painters to have a favorite painter, and for writers to be asked about their favorite writer (it's a formula interview question particularly for writers). So I guess that means that in not only asking the question but insisting on it as a valid or important question, David Brooks is implying that philosophers are like artists, their work subject to aesthetic judgements, preferences and stylistic likes and dislikes. I'm not sure that's what he intended in bringing attention to the question.

There's lots of action in the online calendar world. A fair amount of it gets brought to my attention, which is lots of fun.

• Ted pointed me to Mosuki, and I hope to meet some of the creators shortly. It's more oriented towards personal calendars than the other sites but it does allow you to share.
  
• I met Brian Dear a few months ago and heard about EVDB. Now that it's announced, I can talk to others about it too. Scraping sites for event and venue information sounds like a really valuable service.
• Kragen Sitaker showed me a del.icio.us-like calendar site that seemed pretty cool. Rather than try to scrape every page, his demo let the user pick certain text, send it to the demo site and it would turn that text into an event (pulling out date, time and location if it could) on your calendar.
• Brian posts about Upcoming, a public events site with feeds to let you know about event categories you're interested in, and a way to add events you want to your personal calendar. It has a HTTP/iCalendar interface that allows users to synch into iCal. According to del.icio.us, upcoming is the biggest new thing with hundreds of bookmarks.
• So far, Trumba's newly announced OneCalendar gets a rather less enthusiastic response. Trumba consists of several ex-Visio guys according to their press release, and like EVDB they have high-powered investors so expect some noise. Already you can synch with Outlook which has got to be a highly desired feature.
  
• Of course, we are working on some of the same stuff too, only with a federated server approach. When you share your calendar with Chandler you can share it on any WebDAV or CalDAV server (see also recent article). We're working on a WebUI for such shared calendars so that the calendar owner or their friends can view the shared calendar or individual events just by going to the URL in the browser. We may do tagging just like everybody else, too.

So far I don't use any of these, there isn't quite enough "there" there, whether it's users I want to share with or features I need. My calendar is in iCal, stored locally, and I don't even share it with anybody, though I do share one other person's iCal calendar. But I don't think that isolation will last long given the exploding options.

I wonder what Yahoo, Google and Microsoft are going to do in this space? Yahoo has more of the world's calendar info than any other site at this moment, because of its excellent support for group calendars. What will Yahoo do with that data? If it does not expand its features soon, possibly allowing users to synch up calendar data on the site with their client software, Yahoo will soon see its calendar data decay and vanish. As for the two gorillas, I really don't have any idea whether they will buy, innovate, or crush.

In response to my last post, Mark pointed out that .NET's HTTPClient supports pipelining. Heikki said that Mozilla put in support for this but it isn't turned on normally because so few servers support it. Furthermore, Heikki reports that tests showed only a 7% performance increase when Mozilla uses pipelining. The Mozilla FAQ is a good start for information on this though it doesn't have the data.

The application I had in mind (Chandler) does more synchronization than browsing over HTTP/WebDAV. For example, if the user decides to work offline, and while offline moves a bunch of resources from one collection to another, then goes online again, Chandler would have to issue one MOVE request for each moved resource. With pipelining, Chandler could theoretically fire each request off and wait to start seeing the responses come back in order over the same connection. The only thing limiting throughput is the bandwidth and MOVE requests and responses don't take up that much. Without pipelining, Chandler has to wait for each response before sending the next request. Now this work is limited by latency, multiplied by the number of resources being moved.

Still, I don't consider this proof that pipelining would be useful. It would be great to see more data one of these days -- assuming, of course, that there really is a performance problem in the first place.

Are there any open source HTTP client libraries that support pipelining? Are there any open source libraries that solve an isomorphic problem with a different protocol? So far I'm not turning up anything but this is the kind of thing that is a little hard to Google for.

At the gym last night I was lifting weights and a man started this converstaion.

Man: "What are you going to do with all those muscles? Beat up the boys?"
Me: "Yup."
Man: "Well then, stay away from me.

What am I supposed to make of that? Seriously, I'm asking here.

... So this hat doesn't fit me. The question is, would it fit a guy with a size 7 3/4 head? Joe, you might or might not want to click through :)

There are a whole bunch of "Working Group last calls" going on for email-related technology. Roughly, these are calls for reviews of Internet-Drafts before they are submitted to become standards. I called for reviews for an IMAP Extensions working group document which adds annotations (like properties, or metadata) to emails in an IMAP server. The Lemonade WG has issued last calls on these documents (all last calls on the mailing list and the documents themselves are linked from the Lemonade charter page):

• the Catenate proposal, used to allow clients to more efficiently work on draft messages stored on an IMAP server
• URLAUTH, which lets one user show an attachment to another user without sending them the whole attachment (basically I give you authorization to view my copy on my server)
• Server-to-server requirements for email event notification systems -- intended to allow voice mail and email servers from different vendors work together more smoothly
• The BURL draft, allowing clients to forward an email without downloading and uploading it again as email systems currently require.
  

There are also upcoming improvements to the performance and extensibility of the LIST command which is the heart of IMAP, and ongoing improvements to access control, both coming to a head in IMAPEXT. If you're interested in email protocol and developing standards for those, this is an exciting time for that. I guess that means about 7 of us globally are excited right now :)

I attended a knitting convention this weekend, and learned how to do japanese short rows, double crochet (which is like a cross between knitting and crochet), and twined knitting which is a three centuries old Swedish technique a.k.a. "Tvåändsstickning". The marketplace was also amazing and very tempting and dangerous to my wallet.

The "gallery" at the Stitches conference is very ad-hoc -- it's on the backs of the attendees. Here's a few pics of people I saw.

It's starting to become well-known that well-published suicides inspire more suicides. There's an explanation on everything2.com, but I'm dissatisfied with the references to heroic portrayal as the explanation for the phenomenon. The Wikipedia entry is much more complete, but it still explains copycat suicides by refering to the media "glorifying the deceased." This "reward" theory is simple--that the media encourages copycat suicides by providing attention and glory as a reward for suicide. The standard sources also explain that media coverage may provoke admiration of the deceased in eulogizing him, and by explaining the causes of the suicide and thus implying that suicide is an understandable or normal response to those causal events.

The reward theory has been debunked somewhat in Cialdini's book Influence,
the Power of Persuasion. Some brief reasons to doubt the reward theory:

• Even consistently negative publicity encourages copycat suicides
• Copycat suicide demographics are surprisingly similar to the original suicide (e.g. 35-year old women don't tend to see Kurt Cobain's suicide as one to copy no matter how much glory is supplied)
• Deaths which are not publicized as suicides but which might have been (car and plane accidents) provoke a rash of similar deaths even though there's no glorifying of suicide.
  

Cialdini's model to explain copycat suicide is more sophisticated than the simple "reward" model -- Cialdini calls the effect "social proof" and it goes beyond just suicide to explain how people copy the actions of those they consider similar enough to be role models. So even if the suicide is criticized unanimously, someone who feels similar (same preference in music, perhaps) will find actions to emulate. In fact, the criticism from the dis-similar people may strengthen the bond felt towards the role model. The bond may be strengthened by the copycat feeling that society is against both himself and the original suicide and provoke feelings of solidarity. Thus, a society that reacts to some unusual person's suicide by criticizing their unusual taste (goth clothes, punk music, or playing dungeons and dragons) may in fact create a self-fulfilling prophecy in calling that taste dangerous and then see a spate of suicides in a population sharing that taste. There's also some evidence (again you can reference Cialdini who discusses Goethe and his hero suicide Werther) that even fictional suicides inspire copycats.

Dorothy Parker seems to have intuitively understood this. I recently read Bobbed Hair and Bathtub Gin, a look at literary culture in the Twenties by examining the lives of Dorothy Parker, Edna St. Vincent Millay, Edna Ferber and Zelda Fitzgerald. Dorothy attempted suicide several times in her life, each time via a different approach, and damaged her own health and reputation. In her short story "Big Blonde", Dorothy wrote:

There was no settled, shocked moment when she first thought of killing herself; it seemed to her as if the idea had always been with her. She pounced upon all the accounts of suicides in the newspapers. There was an epidemic of self-killings -- or maybe it was just that she searched for the stories of them so eagerly that she found many. To read of them roused reassurance in her; she felt a cozy solidarity with the big company of the voluntary dead.

The subtlety of the copycat suicide effect probably means there's no easy solution. It doesn't work to have the media to portray suicides negatively -- that doesn't prevent the social proof phenomenon and may even strengthen it. We probably don't want to instruct the media to entirely suppress news of suicides. Even if we did that, there's still fiction and possibly other art (music lyrics?) -- and even if we censor art there's still deaths-which-might-have-been-suicides.

Despite these subtleties, most journalism codes around discussing suicide in the media focus on the glorification aspect. E.g.

a licensee must not broadcast a program which depicts suicide favourably or as a means of achieving a desired result (ref)

The WHO report from 2000 also talks of glorification and acceptance of suicide as an understandable response, but it goes even further, stating that "certain types of coverage may help to prevent imitation of the suicidal behaviour" (p 6). However, I've not yet seen evidence of that, and I worry that is merely wishful thinking. If Cialdini's model is closer to being correct then the very type of coverage that the WHO report suggests may do more to encourage copycat suicides by mourning the deceased, providing details of their life and families, and providing "risk indicators and warning signs" which can trigger the role model effect.

Update: I've never updated a Wikipedia page before, but this seemed a good time to try. I added the paragraph on Cialdini's social proof model.

I got annoyed on the drive to work today, listening to NPR's Morning Edition program. The show was covering California's new display hardware tax and recycling program. I guess "coverage" always means looking for flaws, but this was silly.

Here's the facts: California is adding an electronics recycling fee, which retailers apply based on size of display, from $6 to $10 per display. This includes laptops, monitors and TVs.

Morning Edition added the factoid that California pays recyclers $0.48 per pound to process the displays, to chew them up and sorting the resulting bits into plastic, glass, etc. From this factoid and the fee rates, Morning Edition concluded that the tax would certainly not pay for disposal. They gave an example of a 53 lb monitor, which would be taxed at $10 at time of purchase but would cost $25.44 to recycle.

With the information provided so far in the program, I could see immediately several flaws in their argument. First, not all monitors are that heavy. People buy 5-10 pound laptops, pay $6 or $8 fee at purchase time, and those only cost $2.40 to $4.80 to dispose of. Second, there's time-value of money: the $10 paid for the 53 lb monitor is paid years before that monitor has to be recycled, and if the state saved that money it could accrue interest.

Even if you say that this year's recycle fee income should cover this year's disposal costs (rather than allocate the fee to one item over its life) it's entirely possible that the program does pay for display recycling. It works if people buy more lightweight displays (for which they overpay on the recycle fee) than heavy displays (for which they underpay). At the same time, it also works if people buy more displays per year than they dispose of -- also a reasonable possibility.

This kind of thing in the media bothers me more and more. It makes me think "How stupid do they think I am?"

Today was the first Interop event for CalDAV. Sadly, I am not there, having cancelled a trip due to a cold. But I'm excited to hear about preliminary successes. It's only been a couple months since people started developing code to this spec, so this is great progress!