Saturday, April 30, 2011


We've just left Taketomi island and are moving on to Iriomote. The weather is uncooperative but other than that the trip has been great so far. The food is good, people are friendly and the Habu have been conspicuous only by their absence.

Thursday, April 28, 2011

Fold Your Own Book Cover

We got up at 04:40 this morning - hopefully1 - and we've already arrived on Ishigaki island in Okinawa as you read this. We're probably taking a walk around town, looking for lunch and getting a few last-minute things before we take the ferry to nearby Taketomi island where we'll stay for the next couple of nights.

No summer vacation is complete without a shady spot, a glass of beer and a paperback. I feel fully confident in Okinawas ability to supply the shade and the beer (Orion is pretty good), but a paperback I have to actually pick out and bring myself. I'm starting on "Kasha" by Miabe Miyuki; I read her "Riyuu" a couple of years ago, and found it very difficult, but impossible to put down. Now, my Japanese has progressed since then, and Kasha is supposedly an easier read, but Riyuu took me a year, and I expect Kasha will take me months at least. More so now that I read the morning paper on the train, with less time for a novel.

Paperbacks aren't the most resilient of things, and they tend to fall apart when you tote them around in a bag for weeks and months. A book cover is a good way to protect the deathless prose, and not just from wear and tear but also from prying eyes; you can leave your fellow commuters guessing whether that thick volume is Harry Potter or Finnegan's Wake (as if there's any doubt, cover or not).

Book Cover

Kasha, by Miyubi Miyake, and a smallish furoshiki (about 45x45 cm I think). I got this in Kyoto; I like how the two sides are printed in different colors. Any somewhat thick and stiff cloth would do - a cotton sheet might be a little too soft and flimsy to stay in shape, while starched cloth or oilcloth would probably make a very durable cover.

I like furoshiki, those all-purpose squares of cloth you can use to wrap and carry things in. I wrap my lighting stuff in a large blue furoshiki that doubles as a great backdrop. A small furoshiki makes a great book cover. I saw this book cover idea explained in the store I got the furoshiki, and added a step to make the cover show only one side of the cloth.

Book Cover

1) Put the cloth, back side up (our cover will show the blue side) on a flat surface. The vertical size of the cloth needs to be a bit less than three times the vertical size of the book - 2.5 to 2.8 times or so. So, we fold up the bottom edge to make it the right size. You can just fold by hand - especially when you practice - but I like to actually iron each fold to make the cover solid and tight-fitting.

Book Cover

2) Now, the width of the cloth needs to be enough for the entire cover - both sides and the spine - plus at least 4-5cm but less than the side width, for each side. The cloth here is a bit too wide for this book, so we'll fold it to size.

Find where the book will lie in the center of the cloth (fold the cloth in half and place the book in the center, then fold back). We fold in each side enough so that the edge of the cloth - one cm or so at least - will end up under the book itself. Do one side by eye, then measure how much you folded and fold the same amount on the other side. This extra fold will make the book cover an even color.

If your cloth is not wide enough to do this, then don't worry and just skip this step. The front part of the cover will show the opposite side of the cloth. If the cloth is the same color on both sides it won't matter at all, and contrasting colors can make for a cool effect.

Book Cover

3) Here's the tricky fold. Put the book in the center with the cloth folded over it. Then fold back the overhanging cloth on each side back over the book. The edge of each fold should not be flush with the book edge, but should stick out just a few mm - 3-6mm or so - from the edge of the book.

How is this tricky? The amount it needs to stick out depends on the cloth you use, how firmly you want to book to rest in the cover, the thickness of the book bindings, how tightly you do your folds and other things. If it sticks out a lot the cover will be loose. If it sticks out too little, the book won't fit. You may need to redo this a few times to get it just right. You might also want to make this just a little crooked, with the bottom slightly narrower than the top, to make step 6 easier.

Oh, and the extra bit of cloth tucked under the book in step 2 tends to want to slip out, so this is one fold you might want to iron or press firmly. You can use a couple of paperclips to keep it in place until step 5 if you want. If you never did step 2 you don't need to worry about this of course.

Book Cover

4) Now for the vertical folds. Use the book as a measure to fold the top part down. The top part you fold down should not reach all the way down to the bottom edge of the book. Instead, make the part you fold down and the part that is left below the bottom edge of the book about equal in size. Since we made the vertical size a bit less than three times the book height in step 1, this will give you a bit of free space between the folded down top edge and the bottom edge of the book. This is good.

Book Cover

5) Fold up the bottom part to match the book bottom edge. You might want to leave just a mm or two of extra vertical space to make it easy to insert the book.

Book Cover

6) Tuck in the corners from the latest fold under the tabs from the previous fold. Getting this all nice and flat can be a little fiddly. The topmost part of the top fold may be a little wide so you get some extra cloth with nowhere to go. You can sort of adjust the horizontal folds in step 3 a little to account for this (make it ever so slightly narrower toward the bottom), or just iron the extra cloth into a crease; it won't be visible anyhow.

Book Cover

7) Now, slide the book binding into the pockets created by the previous folds on each side. If you got the tricky fold in step 3 right, the book should fit perfectly. If it's too tight the book won't close properly; with too much margin it'll be a little loose. The vertical folds may need adjusting too, though that's much easier. If it fits well you may want to remove the book and iron the cover flat, to make it really smooth and tight.

Book Cover

We're done! Neat and tidy, and it protects the book surprisingly well. The cloth is thickest around the front edge and corners that need the most protection. The cloth will soak up light spills and will keep grime, breadcrumbs, marmalade and other stuff from the book. The front-facing folds make pockets where you can tuck in a pencil or a small notebook2. And when you're done with the book you simply wash the cloth and reuse for the next one. A furoshiki this size will do anything from a small paperback up to a large hardcover.

Book Cover

#1 I remember a time, not that many years ago, when four-thirty was still the night before, not the morning after. I'm getting old I guess.

#2 or an electronic dictionary, a cell phone, business cards, subway pass, really flat ham sandwiches (use a rolling pin, go light on the mustard), wads of cash, small screwdriver, tissues, fishing lures, film rolls, filters, condoms, medicine, dried persimmons...

Tuesday, April 26, 2011

Monday, April 25, 2011

(on our way) To Okinawa

Golden Week is coming up, and we're going on a real, true, no-work-at-all-honest holiday in Okinawa. How real? I don't bring my laptop, that's how real it is. I will bring only my phone, a change of clothes and enough cameras to club an elephant.

We're flying directly to Ishigaki island, and we'll mostly be staying on Taketomi island and Iriomote island. These islands form the westernmost part of Japan, closer to Taiwan than to Okinawa island and are about 2.5 hours away from Osaka.

We're going to stay in small lodgings, eat local food, read, sleep and drink beer. Probably spend a fair amount of time walking on the beach, and perhaps try a little snorkelling. Should be sleepy and relaxing. Unless we're stung by box jellyfish or blue-ringed octopus, bitten by habu snakes or brown recluse spiders, gored by moray eels or sharks, washed away in underwater currents, or hit by a typhoon, earthquake, tsunami or flash floods1.

Our plan, other than dodging the venomous parts of the animal kingdom, is to spend two nights on Taketomi, then three nights on Iriomote where we'll try to catch a jungle cruise or guided tour, and finish with one night back on Ishigaki before we return to Osaka. I take one of my vacation days on Thursday so we can leave the day before the Golden Week holidays begin; the flight is cheaper and easier to book, and we'll avoid much of the holiday rush.

My main concern (apart from survival) really is what cameras and what film to bring along. I'll bring my Pentax K10D for casual shots, and I got a waterproof Fuji single-use camera for the beach. The Yashica TLR has been my travel camera, but the winder is getting too unreliable (frames now sometimes run into each other and it has jammed a couple of times). The Pentax 67 is great, but it's pretty heavy and bulky to travel with; I'd have to get a separate backpack. In the end I decided to support the Japanese post-earthquake industry2 and bought a Fuji GF670, a camera I have lusted for ever since it was announced a few years ago; I'll write more about it at a later time.

As for film I'm torn. On one hand I prefer black and white over color. On the other, if there's any place that cries out for color it's Okinawa (the warning patterns on poisonous predators are very colorful if nothing else). On the third hand, I do bring a digital camera along that shoots color. On the fourth (all hands on deck!), it's hard to beat a big medium format negative for clarity and dynamic range. In the end I'll bring rolls of both Kodak Ektar 100 and Ilford Delta 400, and decide later. I'll probably add a roll of Astia slide film too, just for the pure pleasure of seeing MF slides on a light table.

The flight leaves early on Thursday morning - really early; we'll have to get up before five in the morning to catch the first subway train. Here's hoping we don't sleep late. It'd be embarrassing if we did.


#1 Don't google for dangers in Okinawa. I'm sorry I did. I was looking forward to a great holiday, but now I just want to lock myself up in the center of a brightly-lit bunker with a sharp stick and a can of industrial-strength animal repellant. After a few hours on the internet I'm amazed Okinawans even manage to reach adulthood before the local wildlife kills them off, never mind the islands having the longest-lived people on the planet.

#2 It's a convenient excuse, yes, but with a basis in reality. The quake and tsunami and the consequent power supply problems have hit the manufacturing industry very hard, and people are understandably reluctant to spend, which is making things worse. The slump makes it even more difficult for industry in Tohoku to rebuild, and one of the things the badly hit communities need is a return of work so that people can stay and begin rebuilding their lives.

So yes, I bought the camera mainly because I really wanted it, and we're going on holiday because we want to. But in a small way both help alleviate some of the economic aftereffects of the disaster. My comment is mostly tongue-in-cheek, but not only so.

Friday, April 22, 2011

Opinion Polls

I've been pushing this text around for a week or more; it's still too wordy by far but it has resisted all my attempts to tighten it up properly. Better post it now than push it around my plate for another week or more.

The latest polls for the Kan administration are out. They're disastrous. This is not exactly a surprise. His government has been indecisive and wavering, and an earthquake and tsunami of historical magnitude with a nuclear breakdown as a chaser doesn't help. After only ten months in office , there's mounting calls for him to step down. Take a look at this figure:

Approval ratings

Approval ratings over time.

Notice a trend? Japan has seen its share of less than amazingly competent prime ministers lately - neither Abe nor Hatoyama will ever grace any political top-ten list - but there's something more to this. One or two dud leaders is normal. Five leaders in a row is not1. The leaders range from the muscular nationalist right (Abe) to moderate, almost social democratic center (Kan). Would you believe the next PM - whoever it is - will avoid the same fate?

I don't think so. People who become prime ministers aren't stupid or devoid of talent, by and large. They need plenty of skills and political savvy to ever get into contention in the first place2. But in Japan, at this point, you could elect the genetically cloned love-child of Mahatma Gandhi, Machiavelli and Julius Caesar, and they'd crash and burn at the polls within six month of assuming the Japanese premiership.

I find it frightening that Japanese governments are so weak that opinion polls can actually topple them. I find it strange than nobody else seems to think it is. Think about it - they're media polls. They're not votes and have no legal force. They're anything but impartial; you can get wildly different results depending on who you ask, and how you ask them. It's no surprise that conservative media polls find support for conservative causes, progressive polls find support for progressive causes, and liberal polls for liberal causes.

A government could just shrug and ignore them. And yet, this agenda-driven competition for whoever screams the loudest - a mass media version of extremist gangs roaming the streets, fighting each other - is allowed to set the agenda and to save or topple elected officials. That is frightening. It is mob rule through media.

I don't know what other countries regularly allow a sitting government to topple - not just call an election, but actually remove the sitting prime minister - simply because opinion polls look bad. If anything, the ruling party would normally circle the wagons and protect their leader - and their handle on power - and take a bet that they'll improve things in time for the next election.

But the major Japanese parties aren't really parties in the normal sense. They don't have a unifying ideology or commonality of purpose. They're more like electoral umbrella organizations that lets its members collect contributions and campaign under one name. Both the DPJ and the LDP house wildly divergent political groups within their ranks. The Ozawa group and the Kan group within the DPJ, for instance, have less in common than either have with similar groups in the opposition parties. They're really separate parties in all but name.

So when Kan gets into trouble with the opinion polls, the reaction of the Ozawa faction is not to help and support the party standard bearer. They don't see one of their own under attack; they see a political enemy weakened and will try to use it to bring him down. A Kan government is not "their" government, despite having the same party label, and they see themselves as being in opposition to it. That both groups happen to belong to the DPJ is irrelevant. The same goes for other factions of course, in LDP and DPJ alike.

But why are polls so dismally low? And why are no parties benefiting from this discontent? A recurring refrain from opinion polls is that the government - whoever it is - is passive and unable to to take decisive action to resolve important problems. There is a well-founded belief that the government is sitting on its hands while problems grow worse over time. The opposition is hurt by this too; the public doesn't believe they'll do any better.

The problem is that taking decisive action means making impopular decisions. The current system makes it all but impossible.

Any impopular decision will have an effect in the polls. If it hits a particular group or industry you'll lose their donations and their campaign assistance, and so you'll lose the support of party members that represent them. The parties are weak, with little leverage over their members, and parliamentarians are dependent on outside help to win or retain their seats. They have plenty to gain and little to lose by helping a donor against the wishes of their party.

The unelected, unaccountable state bureaucracy is as powerful as the elected government; the administration may propose a new directive, but if the ministries are opposed it won't ever get implemented. Many ministries are very close to the industry, such as agriculture, energy and construction, that they're supposed to regulate. Senior administrators get an exit and pension plan through amakudari, and in return the ministries work as protectors as much as regulators. Any adverse legislation, such as deregulation of the agricultural sector or tightening safety rules, are fiercely opposed by the ministries. Most administrations have tried to weaken those cosy ties by abolishing amakudari, but as the ministries oppose it that has proven impossible as well.

The administration, the elected parties and the bureaucracy are the three main actors, all with more or less divergent interests. And in Japan, the administration is the weakest of those three. Which means any decision has to find favour with all three entities, with public opinion and with industry interests. This effectively means that impopular or painful decisions can not be made, no matter how necessary, no matter how beneficial they may be further down the road.

Replacing Kan as prime minister will not change anything. Replacing DPJ with LDP - or with Minna no To or any other party - will not change anything either. And every failed administration will weaken the power base for the next administration in turn.

This deadlock is unsustainable. So what will break it? I have no idea. And I suspect I won't like the answer when we find out.


#1 Koizumi, before Abe, was exceptional, with a rare gift for public relations, but also lucky to have a captive party and a period of economic recovery. Previous PMs, such as Mori, reverted to similar trends we see here. I decided to look at just the last five as they're all of the same generation and worked in a similar political and economic climate, and comparable polling data becomes difficult to find for much earlier administrations.

#2 Abe and Hatoyama were chosen largely for who they are, not for any ability of their own. Abe was the compromise candidate, with an impeccable family background and with solid support from an important right-wing faction within the LDP. Hatoyama was elected party leader, not prime minister. He is a founder of and basically paid for the establishment of the DPJ, so there was a sense that he'd earned a shot as party leader.

Tuesday, April 19, 2011

Pocky - Mikado

Pocky is a snack made by Glico here in Osaka, a cookie stick dipped in chocolate. It's really, really popular in Japan, and is sold in a number of other countries too. When in Paris last summer we found "Mikado", apparently Pocky by another name. We had to compare.

Mikado and Pocky

Mikado and Pocky. Below a Mikado stick on the top, and Pocky on the bottom.

Both are made by Glico. The Mikado stick is a bit shorter, and the taste was a little different - but we got milk chocolate Mikado, and dark chocolate Pocky so that's no surprise. The cookie seems to be the same. If you get a Pocky craving in France, you can safely look for a Mikado box and rest assured that you're getting the real thing.


The Glico Running Man. The company has had some version of this billboard up on Dōtonbori for almost a century, and it's become one of the iconic images of Osaka.

Tuesday, April 12, 2011


2:46 - Aftershocks: Stories from the Japan Earthquake is a collaborative collection of stories and artworks around the Tohoku disaster. There's pieces by Yoko Ono, William Gibson and many others affected directly or indirectly by the quake. It is on sale right now, and every single yen, cent or penny you pay goes directly to the Japanese Red Cross.

It's good, and it's for a good cause. What are you waiting for?

Monday, April 11, 2011

Picture Post - Uji

I haven't done any picture posts for a long time now; time to change that. Last month I held a seminar at my home department at Kyoto university's Uji campus. I work at NAIST, a fair distance away, so I rarely get to go there.


Uji campus is, well, old-looking. Not "charming ivy-covered medieval buildings"-old, but "peeling paint and crumbling floors in Soviet era-style concrete boxes" kind of old. Nevertheless, in a certain light, and certain angles, some of the buildings can look downright good.

No Humanities Around Here

Uji is mainly home to a number of science research labs, and, evidently, home to few or no linguists or language scholars.

Shadows and Lights

Stairwell at my department. It's amazing how black and white can make things look good.

Not Actually Disgusting

As I wrote earlier, we had surströmming while I was in Kyoto. Here Dr. Honda demonstrates his culinary courage and ability to pose for pictures with a sandwhich without looking silly.

Obaku Station

Obaku station, Uji.

Wednesday, April 6, 2011


It finally smells like spring, and the sakura is flowering right on time too. We are hitting the brief, pleasant few weeks between the raw, wet cold of winter, and the oppressive summer heat. We're not doing hanami this year, but I took a couple of quick phonecam shots of flowering cherry blossoms at work this afternoon.


Cherry Blossoms, Ikoma.

One day somebody will release a phone with an actual good camera in it. One day.

Monday, April 4, 2011

Twitter is a Conversation
- That's Why I Gave Up on It

Cool people use twitter nowadays. I don't like to feel uncool more than anyone else1, so I made myself a twitter account some time ago. After a brief trial I gave up and I no longer use it.

Chicken Feet

Tweet, tweet?

I'd mostly forgotten about the whole thing, but I stumbled onto a post about how people who give up on twitter have misunderstood the point. Now, the author is the kind of person who feels that a brief text of 300 words and no pictures is best shown as an embedded Flash applet2 so you'd normally just shrug and ignore him, but he does have a point I want to bring up: Twitter is a conversation. And that, I realized, was the very reason I gave up on it.

What happens when you join is, you start following people. People you know, mostly, online or off. And they talk. And they talk. In brief sentences. All the time. Any one person isn't saying all that much, but follow a dozen people and you get a never-ending stream of brief, disjointed comments strobing across your screen.

You're in a large group of people at a party where everyone sits together but they're no longer all talking about the same thing. A couple of people are talking about work over there; another few discuss politics over here. Someone explains how to make the perfect Chicken Balti while another gets advice on stain removal. Many of them talk to people over the phone as well, so you only hear one side of the conversation.

Great at a party I guess. Not so great when it happens on my computer while at work. When I log in to twitter I invite that group to hang around right next to my desk and loudly chatter away. I lose my concentration and can't focus on my job. So I shut it down for most of my day. But this is real-time conversation; when I log back in I drop right into the middle of dozens of ongoing discussions (or monologues, more often than not), with no good way to catch up. I read "Just make sure it's covered with the oil", and I have no idea if it's a curry recipe, removing ketchup stains, car maintenance or sexual advice.

Seagull hat

Tweet, tweet, twetwetweeeet!

Twitter is brief, free-form, informal, real-time and conversational, and that's exactly why I don't use it. "brief, free-form and informal" really is just saying that it lacks structure. Structure may be constricting and free-form liberating, but structure is actually important for readers to understand what you're saying.

A typical newspaper article follows a specific structure, and as a result we can skim a newspaper extremely effectively. We can read just the headline for the subject, then jump right to the specific details we're interested in. The article structure - and our experience with it - tells us exactly where to find it. Bloggy short-form essays like this one has a well-defined structure too, and scientific papers take structure to an extreme, all to make life easier for the reader3. Twitter streams lack this structure and become hard to follow, especially when you don't follow the thread from the beginning.

The real-time, conversational nature makes it worse. There's no recaps or summaries, no longer, coherent arguments and no background, so catching up becomes all the more difficult. And when you follow several people - and when they talk with several others in turn - a number of disjointed conversations get all jumbled together in a single stream4. You have to continuously follow the chatter to keep track or you'll start from zero all over again whenever you log in.

Twitter isn't the only "conversation tool" out there of course. Most of our communication tools are about conversation. Chat, SMS, twitter, email, online forums, IRC, blogs, multiplayer games, research publications - they're all about people talking to each other, and they all offer some unique combination of speed, reach and longevity.

Chat and SMS have a nearly real-time communication cycle - on the order of seconds - tailored for transient one-to-one conversation. Twitter is just as fast and ephemeral but good for mass communication. Research papers and books are also one-to-many, but very permanent and with a cycle measured in months and years. I guess the classical handwritten long-form letter would qualify as the quintessential one-to-one slow cycle medium. Email is the most long-lived and important electronic communications form we have, and for good reason. It's so very flexible: email supports a communication cycle from minutes up to days or weeks; supports both extremely short and and very long-form messages; it does one-to-one but also group communication; and it lends itself very well to long-term information storage.

Twitter isn't bad. It sits at a particular spot - a combination of speed, longevity and brevity - that proves to be very popular for a lot of people. But no medium is for everybody, and its combination of features just doesn't work well for me.


Not a Tweeter.

#1 You might argue that particular train left the station many years ago. I would not disagree.

#2 You would have thought an "information professional" working for libraries would have the minimal sense not to exclude disabled visitors or visitors using devices without flash, especially when there's nothing to gain. Could he be one of those people who write all their emails as embedded Word-files, just so they can choose their own typeface?

#3 Research papers aren't exactly known to be easy to read, but they are - if you're in the target audience. They are written for a very specific reader: busy scientists that work in the same field and get inundated with new papers every week. They need to quickly and accurately see what the paper is about and determine if the information is useful and relevant for them.

Reading a research paper is a skill; a skill you need to learn and to practice. Once you do know how, you can skim dozens of papers in very little time, and be confident you didn't miss anything relevant. If you don't know how, and you have no background in the field, they can be incomprehensible. Which is OK - you don't learn about a scientific field by reading research papers, any more than you learn how to drive a car by reading manufacturer-issued vehicle maintenance manuals.

#4 "Aha, but IRC and chat rooms are just the same!" Yep - and I've never really taken to IRC either, for much the same reason. And at least with IRC and chat rooms you do see both sides of all conversations.