Hacker News new | comments | show | ask | jobs | submit login
A candy man revolutionized the sushi industry (bloombergquint.com)
172 points by pshaw on Aug 31, 2017 | hide | past | web | favorite | 111 comments

Here's a fun one:

Japanese fully automated restaurant


As a side note, it's funny in that there are many parallels to this fully automated restaurant and this short thought essay that was mentioned on HN a few months ago about the slow and steady future progression of AI: http://marshallbrain.com/manna1.htm

Not _fully_ automated. They may not be "chefs" according to the video, but there are clearly people assembling food.

True - its definitely got a manual component to it. It's interesting to see the sushi rice machine in use in the first half of the video.

What's interesting about restaurant automation is that it could go either way in terms of making existing franchises more of a monopoly or enabling the little guy compete against them.

I tend to actually lean toward the later because the biggest issue with restaurants is managing/coordinating a team of people (interviewing, payroll, scheduling, healthcare, HR, diversity). But if automation can enable the independent restaurant owner to work with a skeleton crew and focus on delivering value and a differentiated product the consumer will win in the long run.

I feel that we as entrepreneurial engineers need to make robots that will help make the independent proprietor capable of competing with the larger franchisors like McDonald. Something like an artisanal grass fed beef hamburger restaurant robot to help the food truck compete against McDonalds.

Here in Austin we have a local chain serving grass-fed beef hamburger that's cheaper than McDonald's by $1.19. The 'combo meal' is $0.50 cheaper at McDonald's, however P. Terry's also makes their fries and everything else from scratch, without any robots. I think they try to take good care of their employees as well.


P. Terry's serves "all-natural" beef, but it is not grass-fed. If you want a grass-fed burger in Austin, try Wholly Cow (which is quite a bit more expensive than McDonald's, as you would expect).

Same thing as Gordon Ramsey serving prepared meals at his restaurants:


Okay, it's "pre"-prepared. Does it taste good? The fact that it's re-heated in plastic bags is more off-putting than it simply being prepared ahead of time...

from what I recall customers where unaware of any difference in taste, so I'd say: success.

And yes, customers were angry when informed it was pre-prepared food (specially being upscale restaurants).

Heck, sounds like the bag could be tossed into a microwave for all I know.

So, those people care more about showing off that they are spending a lot of money, than whether or not the food is better.

> What's interesting about restaurant automation is that it could go either way in terms of making existing franchises more of a monopoly or enabling the little guy compete against them.

Whoever can make his restaurants 0.05% more efficient than the next guy will take over the entire market as scaling up will be so easy.

True. Then maybe it becomes how you define the market. After a while people will crave something beyond burgers and sushi. The wants of mankind are infinite.

The thought/hope is that if individual chef's are freed from having to managing teams of people to handle backend restaurant activities like dish washing, food forecasting, etc. they can start bringing quality differentiated products (Sengalese food, Peruvian food, etc.) to market or existing products to new geographical markets (sushi to rural Midwest).

Sort of how personal computers and now cloud computing enable tech entrepreneurs compete against yesterday's incumbents.

Do you believe that Burger King, McDonalds, and Wendy's are all within 5bpp of productivity of each other? Because none of them have taken over all of the market, not by a wide margin.

I seriously doubt their productivity numbers are anywhere near that tight a spread and yet we've seen no clear monopoly emerge.

I think automation will change the game though. Because the variance that comes from the human factor will be gone. If one chain is a tiny bit better, then every single restaurant will be a tiny bit better. The 0.05% advantage will not drown in the random noise of good and bad employees. And policy updates can happen immediately, the lag from having to retrain will not be there.

One source of variance that will remain is location. The local branch of Very-Slightly-Better-Chain may be poorly situated, which could put them at a disadvantage.

I'm skeptical. This "tiny bit better" might not be noticeable by customers, so it could easily be swamped by better marketing or by having a more convenient location (which will never be consistent).

Also, there is a centralized facility somewhere with people chopping up the fish. It improves efficiency as you need less staff, but it's not exactly automated.

And waitstaff.

I've been to one of those. Setting aside my inner geek, as a foreigner, I really liked that the tablets for ordering were able to switch to English. Also that I was able to eat myself full for about $10 was a nice bonus.

I recently witnessed the latest upgrade to conveyor belt sushi. In this video, you can see the conveyor belts going past the customers, but they still must monitor the sushi going past to pick up their orders.

In a newer restaurant I recently went to, each table has a secondary conveyor belt going right up to the table. After you've placed the order, your plate of sushi rolls to your table like usual, but in addition it then pushes that plate so that it rolls right in front of you on the table.

You didn't watch the video carefully enough. At 0:55, a direct order system is shown, using a secondary conveyor.

The automated restaurant is a chain called Kula Sushi. They have several branches in the US, there's about a half-dozen in southern california.

And Silicon Valley! https://www.yelp.com/biz/kula-revolving-sushi-bar-cupertino Thanks will try this weekend.

Can't help but admire Japanese people who automate everything that can be automated since 1970s! Working with soft and loose material such as rice is a huge PITA for machines, but I'm not totally surprised they eventually found a way to make it work reasonably well. Perhaps this in turn imposed limitations on the water/rice ratio and cooking time, which might compromise the taste.

What is the argument against automation? We have had people putting cans on shelves for thousands of years. Where is the shelf-stocking robot?

Humans are easy to find, fully assembled, break less, and are more easily replaced. We are amazing machines.

Those are reasons not to automate, but they're not arguments against automation. If there is too much automation in an economy (across industries) then there are fewer people available to buy products that are being made. From a Marxian point of view, purchase of automation machinery is part of the cycle of capital crises, and contributes to the tendency for the rate of profit to fall for the employer of automation, because he extracts less surplus value, as machinery embodies dead labour and only has the running cost of 'replenishing' this labour by replacement of its parts or general maintenace of it, these costs coming out of the surplus value extracted anyway.

There's a wonderful little essay by Ernest Mandel on the effects of automation on the amount of wage labour employed in his own time (the 80s), why this occurs, and how capitalism deals with automation versus how a Socialist mode of production might do so: https://www.marxists.org/archive/mandel/1985/xx/future.html

Edit: a relevant little quote follows.

"Japanese socialists [7] have tried to study the effects of new technologies especially on the automobile industry. Also stressing qualitative aspects of the changes (loss of skills, increase in accidents, emergence of new layers of workers and of skills etc.), the authors find a reduction of shop floor workers of around 10% at the most highly ‘robotised’ automobile plant in Japan, Nissan’s Myrayama Plant, between September 1974 and January 1982, accompanied however, by small increases in white-collar personnel. Even the Japanese ‘company unions’ seem worried by these developments, ‘life-long employment’ still the rule in Japan notwithstanding (Japan Economic Journal, February 21, 1984)."

Why does this remind me of the famous Henry Ford quote about doubling his employees pay, so they could afford to buy the cars they made?

This reminds me of that time I asked in an economics class how you make money. In the case of Ford he pays his workers who turn around and give it back to him? So you need money coming in from external sources right ie. Other countries.

What I was asking about is the process of taking something that costs x and charging 3x to make the profit. Where does the 2X come from doesn't that result in debt. I understand you're paying for time that you didn't spend/resources. But it's like swishing water around in a bucket right without external input?

Probably a dumb question but also doesn't help my professor barely spoke English and I stuttered over the word "Inevitably" haha

> Where does the 2X come from doesn't that result in debt.

I think you should read Marx's Capital. It is devoted to answering that question: the generation of surplus value and its consequences in society. As you suggest, such system is not stable, explaining capitalism's tendency to crisis.

Can't recall off hand if he was a good or bad guy "marxist" so I'm wondering about that book is that with his bias and if he was bad... I don't know I suppose the system has to make sense right? Like I agree with the doctor charging more for his services versus a janitor sweeping the floor. But yeah the product buying part.

You could probably map this out.

If X is the united states, and Y is another country. All of the money in business/laborers would eventually go to the top (business) because of the "profit" or 2x but then how does the worker not end up in debt. Unless the worker did the same thing you know building services that multiplied their worth instead of hourly rate, even hourly rate though if you had high wage.

I don't know I tried reading Adam Smith's Wealth of Nations man that was a hard book to read.

I'm not at all an economist, and I expect I'll make some mistakes here, but I can think of two things that might be relevant to your question here. First is https://en.wikipedia.org/wiki/Deflation#Scarcity_of_official... which is that as you can produce more value for lower cost, with nothing else compensating, then things become cheaper. It's the value of things that's "made" in Ford's factory, not the money. If, hypothetically, you had a completely fixed money supply, then you wouldn't get money from nowhere; instead the price of everything else in the market just goes down. There are some reasons to not prefer deflation, so to counteract this the government prints more money. For another example (I think?), look at Bitcoin, where there's been massive deflation because the growth in value has far outpaced the rate of new bitcoins added to the economy by mining.

I'm less certain about the second thing, but I think that debt (and banks?) increase the (effective?) money supply. If you borrow money from a bank (or debt directly from a company for goods or services) to build a new house, then you've acquired a valuable new asset, the laborers you paid acquired new currency, so the total economy can exceed the number of bills that have been printed and are physically in circulation.

There's a lot there that I'm not sure about, and there's plenty of further fascinating questions that follow on from there that I don't know anything about, but I hope that helps you understand some things about how production and the money supply work!

From a bank's perspective, a loan is an investment/asset, so should probably be a part of what is called M2 or M3 money.

EDIT: Wikipedia seems to agree: https://en.wikipedia.org/wiki/Money_supply#Fractional-reserv...

> Whenever a bank gives out a loan in a fractional-reserve banking system, a new sum of money is created. This new type of money is what makes up the non-M0 components in the M1-M3 statistics.

Is bitcoin undergoing deflation? (asking because I'm not sure) it's going to be worth $5,000 soon it looks like I thought deflation is the opposite of that where there's so much it's worthless.

I just assume we get money from other countries. And provided we (people in the US) accept dollars from each other then it keeps going.

Yeah it's confusing I'm definitely not an economist myself haha.

edit: thanks

It comes from economic specialization and the benefit of trade.

Whenever you draw a supply and demand graph, traditionally as a big 'X', draw a horizontal line from the left of the intersection of supply and demand. The area above that line and under the demand curve is the consumer benefit of trade. The area below that line and above the supply curve is the producer benefit of trade. That is the surplus produced just by the act of going to the market instead of trying to do everything yourself.

Ford's employees build more cars than they could possibly use themselves. Ford's customers have to bring something to trade. The existence of money lubricates the trade process. One customer may, after following the cash flows, do plumbing repairs in all the houses of all the employees, and in the Ford-owned buildings. It is easier for that plumber to fix a thousand toilets than to build one car. It is easier for the factory employees to build one car than to fix ten toilets. So if the plumber trades 100 toilet repairs for one car, everybody wins. The plumber, wanting the car, gets one with 10% the effort, and the workers get 10x more toilet repairs with the same effort. Ford himself decides the proportion of toilet repairs that go to him, and how many go to each worker.

The money price of goods and services continuously rebalance to reflect the desires of customers and the capabilities of producers.

In modern times, the owners of robotic capital are essentially allocating themselves--by means of adjusting money payments--tens of thousands of toilet repairs per year, and maybe half of one repair per year to each of their human employees. The robots, of course, don't poop. This just doesn't work. The owner can't possibly break that many toilets, and the employees are forced to repair their own toilets in addition to their regular jobs whenever they break more than once every two years, which reduces the amount of surplus produced by trade between specialists. You have to pay people enough for what they do or make to pay for the things they want, or you can't support economic specialization.

Ford paid his employees more because all of his employees were on average more specialized than the rest of the economy. Assembly-line auto production could not exist without specialization.

It isn't like swishing water around a bucket. It's like ten people adding one bucket of vegetables each to a pot and getting twenty buckets of soup out of it. The whole is more than the sum of its parts. Debt is more like a time machine for making trades happen that could not take place otherwise, as though the vegetables in the soup are all fresh when they go in the pot, but are harvested in different seasons.

Thanks for writing this out. The soup analogy is good. I'm probably dymb/not getting it with regard to how it all starts, who owns that fruit initially.

I mean the system is running already and you just join in.

Yeah alright, I'm not arguing over thinking with a bad brain.

And extremely versatile. When the store is slow, self-check and self-stock robots can't sweep the floor.

The way I see it, that is a waste of a humans potential, to do work as simple and mundane as that.

Wait, what?! Putting cans in shelves for thousands of years? That's just a figure of speech, right? As far as I know, the process is only a couple of centuries old.

For the argument against automatic look no further than the... Japanese:


(Famed animator and film director Miyazaki giving thoughts on AI animation efforts)

> Where is the shelf-stocking robot

In an Amazon warehouse.

I don't really know much about sushi, but this was a fascinating read.

The entire time I kept wondering what the difference would be between the machine made sushi and the stuff made by the guys in the documentary, Jiro Dreams of Sushi[1] Would I be able to tell the difference?

[1] http://www.imdb.com/title/tt1772925/

You can distinguish artisan sushi from mass produced sushi if you pay attention, since they're made in very different ways. Artisan sushi is airier with softer rice. You might be interested in these analyses:



That MRI sushi scan trick is great.

What, you don't bring your MRI to dinner?

I stopped doing this when, in addition to everybody's attention, I started attracting also everybody's cutlery (less of a problem in a sushi restaurant, tho)

When you're in the Jiro levels you're presumably evaluating something other than just the food quality.

Here's a fairer comparison: can you tell the difference between sushi which uses machine-packed rice and sushi prepared by the lowest rung of traditional sushi establishment? It is possible that a hypothetical person would be unable to tell the difference, but if that is true of a hypothetical person, they'd likely also be unable to tell the difference between a Chicken McNugget and a more traditional chicken preparation.

I'm not being a food snob about this, I promise -- they're just two very, very different industrial processes which produces output that shares a portion of the flavor profile / color palette / etc but also has trivially distinguishing characteristics. You don't have to get into any of the semi-mystical claims about artistry or tradition to be able to bucket effectively just based on rice consistency.

Machine sushi rice tends to be packed together, and I think it's usually cooked a little firmer so that it survives machine handling. There's a finesse that good sushi chefs have that hasn't been replicated by a robot yet.

Good sushi rice is somehow lighter, but still cohesive. The flavor balance with the fish is better and more consistent. The fish doesn't tend to fall off as easily.

If you eat a fair amount of sushi, you'll notice the difference. If you don't often? Maybe not.

I've had a few dinners at really good sushi restaurants, small spots with the traditional 2-3 chefs working behind the sushi bar. Those were some of the best meals I ever had.

You would definitely be able to discern the difference. It's a pretty big margin. Having said that, I have eaten plenty of robot sushi, and it's sometimes pretty good.

First: Yes, if you are the kind of person who pays attention to your food, you will be able to easily notice the difference.

Sushi is vinegared rice. There are 2 main types of sushi you will eat in a sushi restaurant: nigiri sushi (formed with the hand) and maki sushi (rolled in a sheet of seaweed). There are other sushi dishes, but you won't get them in a sushi restaurant.

Sushi is made by cooking the rice, simultaneously cooling the cooked rice and reducing the moisture content, cutting in the vinegar/sugar/salt mixture and then finally forming the rice into a shape. Some of these things can be done well in an automated process, some of them can not.

To make very good sushi, you need to understand the rice. Rice is all slightly different and it will contain more or less starch, have more or less bran still attached to the kernel, etc, etc. When you start you need to inspect the rice, wash it appropriately and choose the correct amount of water to cook it with. How well you do this step makes a surprisingly large difference in the end product. If you don't wash off the starch well enough, it will end up gluey. If you don't polish off the bran, you will get rancid flavours. If you don't choose the correct amount of water, you will get mushy rice. The list goes on and on (I'm by no means an expert, but I've failed to make great sushi on enough occasions to at least enumerate some of my failures ;-) ).

Generally speaking, if you buy sushi at a kaiten sushi place -- or even just a place that just isn't very good, they aren't going to get this step right because it takes experience and attention to detail. Big kaiten sushi chains are going to select a forgiving rice. They are going to wash it exactly the same way each time. And they will get a relatively consistent, but not very good result.

The next bit is cooking. This one is perfectly suited to automation. If you get the rice to water ratio correct, modern rice cookers cook rice virtually perfectly. From Jiro Dreams of Sushi, I know they use their jury rigged pressure cooker setup, but TBH, I'm a bit skeptical that it makes all that much difference. He says that he has to do it under pressure because of the rice variety, but I'm not sure why he can't use a normal pressure cooker -- probably because that's just not the way he does things.

Anyway, once the rice is cooked, you need to cool it to a precise temperature while at the same time reducing the humidity. This is really hard to do well. If you get it wrong, you will either get too hard a texture, or more likely mush. However, it's the kind of thing that machines are really good at.

The next step is really hard to do well with a machine, I think. While maintaining the correct temperature, you need to mix in the vinegar mixture. The technique you use to mix the vinegar, the speed that you mix it, the amount that you use -- all depends on what the rice is doing that day. It depends on the ambient temperature and humidity. It depends on how well you've washed the rice. It depends on how perfectly you got the water/rice ratio when you were cooking. If you get it wrong, you will range from crumbly, falling apart sushi to glue. Machines do this very poorly, it seems. They uniformly under-vinegar the rice. I think it's because when they are extruding the rice (in the next step), they need the rice to be firmer than you would if you were doing it by hand.

First you need to keep the sushi rice around before use. You can't really have a continuous cooking process, so you need to cook in batches. However, as the rice cools, the starch forms crystals. This causes the rice to taste dry and to have a rough texture. You need to maintain a consistent temperature -- not too high because it will continue cooking, not too low because it will crystalise. You also have to maintain the proper humidity. There is nothing that makes this harder or easier for automation, but it depends on how you are moving the rice from one place to another.

Finally, you have to form the rice. As I said, there are two main kinds of sushi that you will have at a sushi restaurant. The one that many North Americans are most familiar with is maki sushi. This one is dramatically easier to make. You spread the rice on a piece of nori seaweed and roll it up.

To understand how to do this well, you have to understand that rice has a flavour. It is a very subtle flavour, but a flavour nonetheless. Japanese meals are not like western meals. You don't have "main dishes" with fillers like potatoes and bread. Instead, the main dish is rice. Everything else is a "side dish" and is meant to accentuate the flavour of the rice. This is why Japanese food is so delicately flavoured, without strong herbs or spices. It's also why you eat the rice without salt or some kind of sauce.

With maki sushi, you have a piece of roasted nori that adds some crispness and earthiness. Then you have a filling that should accent the rice. The problem is that if the rice is mushed up together, it tastes like glue (umm... well, it literally is glue at that point). To get the flavour of the rice to come out, the most important point is air! Each kernel of rice needs to have air around it. At the same time, you need moisture, so that it has a good texture, sticks together, and creates a kind of sauce in your mouth (mixing the nori, rice and filling together). It's that balance of rice, air, water, filling and nori that is important to getting something that has nice flavour and then dissolves in your mouth, with a satisfying, mouth feeling texture and flavour.

Like I said, maki sushi is dramatically easier to do than nigiri sushi. That's because you spread the rice on the nori. This automatically creates both places for air and water to exist. Then you add the filling and roll up the nori. The nori then protects the sushi from drying out. Machines are extremely good at making maki sushi, IMHO.

Nigiri sushi is much more difficult. With nigiri sushi, you wet you hands (just enough), take a small handful of rice, form the rice into a rectangle, add wasabi (if you are going to do it) and then apply the topping (usually sashimi -- raw, sliced fish). Getting a machine to produce nigiri sushi at all is a triumph of engineering. However, it is uniformly awful. Instead of making a 2 dimensional object and then rolling it (to trap both air and water), you need to make a 3 dimensional object, with water surrounding the rice kernels and air in between. It has to be compressed just right so that it sticks together, while at the same time retaining the air pockets and water pockets. Then you have to apply the topping so that it sticks to the rice, so that when you bite it, it seems like one thing as opposed to falling apart in pieces in your mouth.

Good nigiri sushi seems to melt in your mouth. The flavour of the fish mixes with the flavour of the rice and washes down your throat. There are many different ways nigiri sushi can be bad, but the main ones are "piece of fish sitting on a pile of glue" and "a bunch of random junk that falls apart in your mouth in a sticky, mealy way". Machine made nigiri sushi tends to be the latter.

Having said all that, there are some caveats. The convenience stores sell "onigiri", which are triangular shapes rice balls wrapped in nori, with some kind of filling. They are universally awesome, even though they are made by machine. There are onigiri restaurants where they make everything by hand and they are admittedly better, but not nearly as much as hand made nigiri sushi vs machine made nigiri sushi.

And finally, I've taken some friends to very, very good sushi restaurants in Japan before (I live in a fishing village and we're lucky to have a couple of good sushi restaurants). Usually they are blown away. Occasionally they are, "Meh. Fish on rice. I don't really see the difference other than the price tag". So without knowing you, I can't tell for certain that you would care about the difference.

Thanks for the writeup. What a fantastic read.

You also, implicitly, very well explain the reason why it's sacrilegious to dunk a piece of really good Sushi into a mixture of soy sauce and ersatz wasabi. If you invest so much care into the rice this is an insult to chefs studying the process for years.

I have a question, which you may be able to answer. I'm off to Japan in ten days and would like to try some good sushi, without completely breaking the bank.

The problem with top level places, apart from paying 400EUR for a 30 minute meal (which, admittedly must be out of this world) is to get in in the first place. Web sites (if they have any) are in Japanese and most such places only take reservations by phone, which, for a gajin is pretty much a deal breaker.

Do you have any tips about second tier sushi places, which serve a good product without totally killing off the budget? Where to find them, how to spot them and what to look out for? Any other important tips you can think about for a 3 1/2 weeks trip through Japan? (By train, Tokyo to the South and then up to Hokkaido)

I hope this is not imposing, but that was such an interesting and insightful writeup that I just had to ask.


My friend helped to edit the Eater guide to Tokyo earlier this year so it's quite good.


As for sushi- there are a lot of good places near the Tsukiji fish market. There are cheaper places and more expensive places- you can walk around and look at menus and decide. Some are kaiten, some are not. Near the fish market, you generally get what you pay for- which is to say that you can get a good sushi meal for 5000 yen, a very good meal for double that, and if you want to pay more, you can get much better than that too. In general, the top places are not near the market- they're in Ginza or other neighborhoods.

What an amazing resource.

Thank you very much for the link.

I had several inexpensive sushi meals near Tsukiji market and another at a counter in the Ginza train station, all of which beat 99+% of the sushi meals I've had in the US.

We also went to one of those amazing, reservation-needed places, but were arranged by a local there.

I strongly suspect that it's about as hard to find good sushi in Japan as good Italian food in Rome. My advice is to get out of the hotels and onto the streets and don't be afraid or ashamed to eat at a place without English menus. Point at the pictures if you must, but I didn't have a single bad meal experience there.

Thanks! I'll make sure to check out the section around Tsukiji market.

I've gone to several sushi places in large centres, but nothing compares to my favourite place in my home town. It's kind of opposite to my initial gut feeling, but it makes sense in a strange way. In large cities, there are enough people to sustain mediocre restaurants. The top restaurants command really high prices, but you can survive if you are willing to charge less. In small towns, nobody has much money and they all eat at home. If you want to survive, you have to be amazing.

So if you find yourself in a small port town, chances are the local sushi restaurant will be amazing. The other really important thing to realise is that really, really great food is often available in incredibly unassuming places. The restaurant will be 50 years old, will be onto it's third generation of master, and will be falling apart on the outside. But the food will be incredible. So it's super hard to tell where to go.

A couple of things might help. First you should know a few kanji: 営業中 means "open for business". 準備中 means "preparing" (not open at the moment). The easiest way to distinguish it is to look at the first character. If it looks like a fat guy with his hair on fire, that's open :-)

Next, quite a lot of great eating establishments are also drinking establishments. Especially if you want to eat and drink at the same time (which I recommend highly). The thing to look for is 居酒屋 (izakaya -- bar/pub). Not sure how to remember it. Write it on your hand :-). People will be impressed if they see it! Often this will be written on an orange paper lantern outside the establishment. Stay away from things called "pub" or "snack". Those are drinking establishments, but are really hostess clubs and the food is terrible.

Another thing to look for is a noren. Here you can see an image of one [0]. When shops are in business, the noren will be displayed out like that. When they are closed, they will either be taken down, or displayed behind a closed window.

The best thing to do to find good restaurants is to ask for recommendations from the hotel where you are staying. It's important to indicate that you are looking for an actually good place and not one catering to tourists. It may be slightly difficult to communicate that. The main concern is that because you don't speak Japanese, you won't be comfortable in a Japanese establishment -- especially if you can't read the menu. If you can't manage to get an answer from the hotel staff (often they are afraid to make a mistake), the way to go is simply to have courage and wander into likely looking establishments.

Extremely good restaurants don't cater to tourists. They won't have menus with pictures on them. They won't won't won't have menus with English. They won't speak English. They spend all of their time thinking about food, not sales. You have to break down the barrier with your own courage. It'll be fine, don't worry :-) And if it isn't, they will be very polite as they usher you out the door ;-).

Some very quick useful Japanese: When you enter, it's useful to say, "Aite imasuka?", which means "Are you open?". If they cross their arms in an X pattern, it means it's no good. Otherwise it's probably OK :-) If they are willing to seat you they will say, "Nan me sama?" (How many people?). Just hold up the appropriate number of fingers. Again, if it's no good, they will cross their arms in an X pattern.

When ordering, draft beer is "Nama". Sake is "Nihonshu". Something stronger is "Shochu". But you can probably get away with ordering "Whiskey" or "Wine", etc. Carbonated fruit flavoured alcoholic drinks are called "Sawaa" (sour) or "Chuhai". If you want to stay away from alcohol, the mainstay is usually "ooloncha" (oolong tea). You can also order "cora" (cola), etc.

For food, just ask for a suggestion: "Osusume wa nan desu ka?" (What is your recommendation). Whatever they say, respond with "Hai. Onegai shimasu". ("Yes, please") It'll be great. Even if they just asked you a question, by responding with "Yes, please" you will establish that you have no freaking clue what they are saying and that they should just give you food.

As you eat, it's good to smile and remark "Oishii!" (Delicious!). Shop owners are very concerned when foreigners enter because they don't know how to please you. If you are visibly happy, they will also be happy. It diffuses a lot of problems. Usually they will give you a lot of special free food (or sometimes they will give you a lot of special, expensive food that you will pay for ;-) But they will love you!) Unless you have food allergies do not ask for substitutions or customisation!!!!! Japanese restaurants can't deal with this. The server's brain will melt. If you press the subject, they will sadly go back to the kitchen where the chef's brain will melt. After a very long time, they will come back and ask what they can possibly do. If you press the subject, they will probably cry. Don't do it!

If you have an allergy, say "Arerugi nan desu!" (I have an allergy) and try to describe it as best as possible (Best to have it printed in Japanese before you go so you can flash it to them).

Since you were asking about sushi, the kanji for sushi is 寿司. The best sushi restaurants will not put that on a sign because they are the best sushi restaurant in the area and everybody knows it's a sushi restaurant (which is why you need to get a recommendation). Unless you want to try kaiten shushi (conveyor belt sushi -- which is actually quite fun, despite the terrible food) say away from 回転寿司 restaurants. Again, if it has photographs of food, or English menus, it might be good, but it won't be at the top. Also, don't look for modern, glitzy, fancy restaurants. Look for "It seems to have been around since 1950 and they haven't painted the exterior once". But the inside will be nice.

Other than than, just relax, have courage and enjoy your trip! The food here is amazing virtually everywhere. On a 3 1/2 week trip you will see and do a lot of great stuff (I envy you going up to Hokkaido -- especially by train). Keep in mind that every small town is incredibly proud of its local produce and cuisine. They will want to impress you with it. Just take it in and appreciate it. If you do, people will respond with more kindness than you can imagine.

Hope that helps!

[0] - http://media.gettyimages.com/photos/noren-at-sake-shop-at-ed...

Wow! Just wow!

  Hope that helps!
Yes, it does indeed. And just by reading your reply I can hardly wait to head off to Japan.

One of the good things is that I eat essentially everything, absolutely love Japanese cuisine (it's part of the reason for the trip) and don't mind ordering the Menu Surprise, because I anyway don't have a damn clue what I order :)

While I'm aware of some of the cultural dos and don'ts (like never, ever lose it, no matter what, not ever!) your reply is a trove of really useful advise. Like "Oishii!" to indicate happyness (which I will experience in all likeliness, so there's nothing phony about it) or to never, ever ask for a substitute in a restaurant. While I probably wouldn't have done that anyway it's really good advise.

The primer in quick, useful Japanese is also fantastic.

To cut to the tempura: I just PDF'd your reply and will keep it handy as a travel resource, while in Japan.

I can't thank you enough for taking the time and putting that up. It would be great if other Japan visitors stumble over it and find it useful. I know that I do.

Thank you so much!

I just wanted to reiterate how amazing your comments have been here. I honestly feel it's a disservice to you to have your words languish in soon-to-be-forgotten HN threads. I'd love to read a blog of yours, or something along those lines!

Ah, spotted the man who cares for food, lovely write up. As much as I have been exposed to Japanese food, there is nothing on earth that will beat a realy good Ramen. The reaction of many of my fellow travellers who join me on my quest for the perfect Ramen is the same as what they give you: they will not taste / see / respect / enjoy the subtle distinctions. This is not their fault, it is my inner freak that over cares I think. S/He who cares about Ramen, watch the lovely movie "Tampopo".

Don't get me started on ramen!!! I can't tell you the hours and hours and hours I've spent trying to make decent ramen. I've gotten up to about "acceptible" level now :-) When I go into a famous ramen shop, I'm always blown away with what they can do. Lots to learn!

Are you willing to be on that? I'm talking about a double blind test. Two plates (plain names with no marking) with 5 pieces from each place. No identifying marks as well.

Off topic, but how's living in rural Japan? Always been drawn to those areas from photos -- probably need to visit soon!

I absolutely love it. I live in Shizuoka prefecture that has some of the best weather in the country (a bit hot this past week though...). It's beautiful. The food is amazing. It's quiet. In the winter I can walk out to the beach at night and see the milky way. The people are super friendly. There is very little not to like (unless you are a city person, in which case you will probably go stark raving mad with boredom ;-) ).

I highly recommend visiting rural Japan. It's a completely different place than Tokyo/Osaka/etc. It's an easy place to travel to. People are friendly. It's not so expensive. Transportation is good. It's insanely safe.

For living, though, you need to spend some time here and decide if you like Japanese culture. Some people really, really don't. It's quite different living here than visiting here. It suits me down to a T, but I know a lot of people who just can't deal with it.

Thank you for the thoughtful comment! I'm going to go back and read some of your longer posts on your experiences. Perhaps it's time to start learning Japanese...

Wow. Thank you so much for the info! Super informative, really interesting, and written very well.

I can't say I've been to Jiro's restaurant (and though I admire his dedication, the whole thing honestly seems a bit much), but I've been to plenty of good sushi places and eaten plenty of conveyor-belt sushi. As it says in the article, they are just different. When you get a burger at McDonald's, you don't expect it to be the same as one from a good restaurant.

The flavor changes pretty quickly. Non-fresh sushi can become very bland.

If that's the case, might a machine produce a more predictable outcome?

The unpredictability is due to the fish itself. Unless the machine can lab grow bluefin tuna it's going to have the same issues.

That's what freezing is for.

And don't respond with "good sushi fish is never frozen"

The fish is literally frozen when caught.

Also, some states legally require sushi fish to be frozen and thawed.

And for good reasons! Fish meat -- especially large fish like salmon -- are notoriously riddled with worms [0]. Freezing helps kill (most of) them.

[0] https://www.theguardian.com/lifeandstyle/shortcuts/2017/may/...

There’s a difference between that and longer term deep freezing. In Japan you can buy fish that’s never been through the latter at some restaurants.

This. My favorite sushi restaurant was a buffet on University in Berkeley, Edoko. If you got there early everything was fresh, especially the saba; I've never had anything close. Not so good at 9pm.

As a UI developer I feel like I am one of those artisan chefs and I dread the day automated UI will come a knocking

> he could bring the previously elite Japanese dish to the masses

I distinctly remember reading elsewhere that sushi was poor's people street food, basically. That is, it was cheap and certainly no elite.

That's sort of correct. Sushi today is very different than what Japanese people ate hundreds of years ago. Before refrigeration, fish was largely salted or vinegared unless it was eaten fresh. Japanese never ate salmon sushi because Japanese salmon often had worms. It wasn't until Norway began importing salmon that Japanese began eating raw salmon. That was a long/difficult road as I heard from Norwegians who know that story. But now Japan imports a lot of salmon from Norway. Also before the modern era, tuna/maguro was not the revered fish it is today. White meat fish were, and to many sushi chefs, still is the top fish for sushi.

"Conveyor-belt sushi" has such a terrible connotation. The worst of fast food, cardboard cuisine, mindless automation, and glassy-eyed consumers.

That said, the concept of what is effectively an automatic rice-forming machine seems pretty tame and reasonable. Compare dicing an onion by hand vs. using a robot-coupe.

I was particularly struck by the dedication here: it took 5 years to get the machine to "work". Keep that in mind the next time you are building your MVP.

Conveyor-belt sushi might sound bad but it's likely better than you'd expect. It also isn't completely made by machine, at least in the ones I've eaten at in America. It's a nice way to have lunch especially, I like that you can get the exact amount of food that you want, and the immediacy of sitting down and taking right from the conveyor is great. It usually feels like a good value for the cost too.

Looks like one of the sushi making robots costs somewhere around ~$15,000. Would be great if there was one available for the prosumer market at a lower price.

I'm not sure prosumer market exists for sushi making robot.

Tough part about sushi isn't forming rice and wrapping, but rather, cooking and seasoning rice, sourcing and preparing ingredients.

They have a nice clean website, if the product descriptions were absent, then I definitely would not have been able to guess what the machines were used for (http://www.suzumokikou.com/product_type/for-factory/).

> Why had the Japanese government embarked on a policy to limit rice production, effectively paying some farmers to keep their paddy fields idle?

Ah, the classic question asked of every overbearing economically-intervening government.

> Ah, the classic question asked of every overbearing economically-intervening government.


So I come from Slovenia, the land of wine that you've never heard of. One of our things is that everyone with even a little bit of land grows vines and sometimes makes wine. There's at least a few vines growing in people's backyards even if they just eat the grapes.

People with slightly more land have vineyards and sidehustle making wine and selling it. This has two effects, 1) there's a lot of very very crappy wine out there, 2) crappy wine still gets you drunk so people keep buying it. As a result a liter of wine cost about 30 euro cents at one point.

So the government stepped in and put a stop to it. Gotta avoid crashing one of the bigger industries in the country right?

How do you do it? You pay people to cut down their vineyards. Cut it down and we'll give you X0,000 euro.

Now people have a choice, if they're making less than that much by selling their wine, they can take the government subsidy, cut down their vines, and go do something else with their life. This is a lucrative opportunity for all small-time wineries. Many people took it.

As a result, people who were making more than the amount of the subsidy, didn't cut down their vineyards AND they can now sell their wine for more than they could before. Very lucrative situation for them as well.

We now have a stronger wine industry, less crappy wine, and fewer pesticides and stuff making it into nature. Hobby vineyards tend to overspray because they really really really care about every little vine.

We still drink most of the wine ourselves though, so you are not likely to ever hear about Slovenian wine outside our country. My French girlfriend says it's the best she's ever had.

>>you are not likely to ever hear about Slovenian wine outside our country. My French girlfriend says it's the best she's ever had.

Sounds like an opportunity to market an entire country's wine culture. Once you get some recognition internationally, vineyard tours, bed and breakfasts, restaurants, etc with wine tourism.

Your girlfriend is in love... congrats! ;)

I also gave her the most expensivest bottle I could find to taste ;)

And she had to say she likes grandpa’s wine. No other option there.

Shhhh, don't say that, government cannot be good!

You'll scare the libertarians!

personally i'd be outraged if my taxes were used to destroy edible food that could be given away to those in need

No food destroying. This is about stopping grapes from being grown in the first place. Money going straight to taxpayers, too. It’s literally perfect.

It's Broken Window Theory economics. You don't increase the wealth of a nation by destroying things.

It probably would have been better to build some distilleries to buy up all the cheap, crappy wine and turn it into blended fortified wines. That would at least be easier to export.

The problem is you’re not dealing with rational actors. A lot of this wine was somebody’s hobby and they would use it for personal consumption and sell to family only, or maybe to a local restaurant or two. No interest in selling to real businesses/distributors.

Is it really broken windows if you encourage people to avoid producing negative economic output? Making wine is a veeery labor and chemical intensive process.

People doing what they want is never negative economic output. It is positive economic output that probably just isn't getting counted by your favorite metrics.

If you can't sell someone a €2 bottle of wine because they make their own, which costs them €10 a bottle but could probably only sell on the market for €1, that just means they are deriving at least €9 per bottle of entertainment value from participating in the act of winemaking. That wine tastes better to them, because they made it. And that €10 is still going to the producers of winemaking equipment and support services, rather than to other makers of wine.

The economist may say, "That bottle is negative output, because that €8 difference between making your own and buying at market is €8 that could be spent elsewhere!" But that person didn't want to spend it elsewhere. They were perfectly happy burning that €8 on their hobby, just like someone else might pay €8 to see a movie in a theater. Rational actors don't make themselves more miserable so that someone else can make another buck. And you can't say that €8 is wasted because the owner of it didn't spend it on what you wanted them to spend it on.

Burning down your vinyard is actual negative output. The productive resource is gone. There are no grapes. There is no wine. There is no money paid to maintain the vines. There is no money paid for bottles or corks. There is no entertainment value from DIY winemaking. All that remains is the desire to drink wine, which can be fulfilled by someone else, whose wine is now worth relatively more only because it represents a larger slice of a smaller pie.

I don’t think you realize how much labor and arable land goes into wine making. It also creates a lot of alcoholics.

When the normal culture in vast regions of the country is to drink 6 liters of wine per day (yes one person) you have a problem. One way to fix it is to incentivize people to stop hoarding hectoliters of wine in their basements.

My grandpa stopped making wine 5 or os years ago and still has about 1,000 liters of inventory. Now that he drinks less it’s almost def gonna last him for as long as he stays kicking.

How’s that for broken window?

I can't even drink 6L of water per day. I can only assume that young Slovenians must have a routine operation to install a wine bladder somewhere along the gastrointestinal tract.

Can you hear the sloshing as people walk? Do you have to salt the wine to avoid a nationwide epidemic of hyponatremia? Do people actually eat food, or do they just get all their dietary calories from sweeter wines?

Six liters? What the hell? I might have had 750mL of wine total so far this year. How do you even... What do you... No, I literally cannot fathom the depths of this aspect of Slovenian culture.

This has to be an exaggeration, so let's check.... "Slovenians drink on average almost 100L of beer per year." "Economic improvement in Slovenia leads to increased sales of alcoholic beverages." "Slovenia ranks 5th in alcohol consumption in EU." "...between 10.3 and 13.5L of pure alcohol per year." "Slovenia: Could a Country Be Addicted to Alcohol?" "According to official data, every fifth man and every 25th woman in the country is an alcoholic."

The alcoholism rate in the US is maybe 5% to 5.5%. I think it's safe to say I don't understand the problem. At least you're not Estonia, I guess.

Except you've now abandoned your old justication this is good for the economy and are now acting as if you always meant this is good for reducing alcoholism

Bait and switch, motte and bailey

It’s good for both

I suspect that if you account for opportunity costs, the "broken window problem" with Keynesian economics is fixed.

In order to stop grapes from growing, you have to destroy them. Cause -> effect.

Because otherwise, prices crash, running all the small-time-to-medium farmers out of business, sending them to slums in big cities as they seek new opportunities, their farms are bought up by large industrial agricultural firms that can afford to scale and survive the depressed price, and then capture regulation to serve their interests.

It's a tale as old as agriculture...

Damned if you do...

It's funny you should mention lack of centralization causing small/medium farms to sell to big firms.

I just read an article which showed how one of the great side effects of the last half century of US estate taxation (where the government takes ~30-50% of the estates wealth after the original owners dies) resulted is a massive centralization of farms to a few large companies.

Basically farming in the US used to be 90% small/medium sized and largely family owned. Most farmers worked on their farms until they died and they left a decent business for their kids, not a wealthy one, but one that was sustainable to support a medium-large sized family - but also significant enough to cross the estate taxation barrier.

When they die their various children are left with the choice of a) giving 50% of the estate to the government, selling half the farms assets to do so, and working hard on this smaller farm for the rest of their lives (and then to give their kids the same choice) or b) selling it all off to the highest bidder (aka one of the mega-farms), paying off the government tax, and moving to the city.

They ALL pretty much chose b) (or sold half in a) to the big farms) and over enough generations we now have a massive oligopoly of a few companies owning all the farms. And this was in the name of making the rich 'pay their fair share'. While the mega-farm owners pay the same estate tax their estate tax doesn't translate into them giving up their mega farms which are controlled by a broad group of wealthy owners, likely engaging in your typical offshore tax evasion schemes to avoid paying their own.

I also know within a few years of Venezuela centralizing food production it caused massive shortages in order to keep prices lower - well before the drop in oil prices pushed their economy into crisis.

Otherwise I don't know enough about the farming industry to comment.

Do you have a citation for that article, because I'm extremely sceptical estate tax is the main reason.

You have to basically be very wealthy to pay the estate tax. Only one in 700 estates pay federal estate taxes today, and while the top rate is 40%, the average rate paid is 17% (https://americansfortaxfairness.org/tax-fairness-briefing-bo... ). Currently, a married couple needs to have an estate over $11 million before that pay any estate tax. I always thought liberals had very poor marketing with the estate tax: conservatives were very successful framing it as the "death tax" (a lot of people thought everyone had to pay it), while I always thought it should have been called the "aristocracy prevention tax".

It's seems much more likely that smaller family farms wanted to cash out and move to the city because farming is pretty grueling work, a similar reason for urbanization all over the world.

Honestly, that sounds like the estate tax working as intended. We would like to avoid a hereditary class of people who can inherit a valuable asset and an upper-middle class job to go along with it. It sounds great for that beneficiary, but it's not so great for all the other people who aren't born into such circumstances.

When the owner of the property dies the heirs inherit it at the stepped up basis without having to recapture any of the depreciation. Even paying a 50% estate tax this is a substantial financial gain if the farm has been taking the tax credit for depreciation for 15+ years.

Interesting. Can you please share the article?

Perhaps because overuse of the soil leads to desertification and a downward spiral where each individual tries to produce more and more leading to faster desertification?

See: The Dust Bowl.

Are you saying paying companies to remain idle is an ideal resource management solution for certain scenarios?

Reading the Wikipedia on the "Dust Bowl" seems to indicate it was solved by financing soil conservation techniques such as tree planting combined with researching and educating business owners on better practices called "Dryland farming" - which the early farmers seem to have been unfamiliar with in <1930s America. Basically farming evolved to take into consideration the unique environment they were operating in which every business and property owner would naturally be incentivized to follow (something that is not always the case for environment related externalities)... I don't see how farm centralization played any role in solving this problem.

"Are you saying paying companies to remain idle is an ideal resource management solution for certain scenarios?"

Yes. And how is farm centralization the same as paying farmers to not plant?

As prices for a good fall, the producers try to make up for it on volume. This is a feature of capitalism. Everyone acting in their own interest leads to a downward spiral. As a result, there is an exacerbation of harm to an ecosystem.

If you don't believe me, model it on a computer. It's the tragedy of the unmanaged commons. And Capitalism also has this effect under normal circumstances. Look at how much deforestation and desertificaton and extinction of species has happened in the last 50 years.

Half of all deforestation in the last 10,000 years happened during the last 50 years.


A quarter of all farmland has now been affected by desertification.


It's not just the bees - the rate of extinction of species has increased massively in the last 100 years and biodiversity is plummeting around the world. We are turning our ecosystems into farms.


The largest concentrations of CO2 in millions of years are in the last 50 years, and increasing.

This is not scary to many ideological conservatives in love with Capitalism. They wave it off and believe in human ingenuity of the future.

Navel gazing at the nice things Capitalism produces doesn't mean that there are no negative externalities. Everyone is incetivized to sell more and more stuff at the expense of natural ecosystems, third world countries and future generations. Look at lake Baikal as a random but serious example.

But once prices crashed enough, farmers stopped even trying to grow things leaving nothing to hold the soil there and making the dustbowl even worse.

So that is indeed a problem, but a different problem from growing too much.

Perhaps letting the land lie fallow every 7 years was an awesome strategy!

somebody had to say his name thrice

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact