On Mayonnaise

An exquisite disquisition on mayonnaise from Metafilter’s Nanukthedog:

Alright. I’ve been a high end cook/chef and I’ve also worked for as ‘Big Mayo’ a company as you can in the world. I know mayo inside and out – from making it in small batches, to mass production and sourcing, to which demographics and some indication as to ‘why’ those demographics buy it.

So first I’ll talk about what mayo has in it and why mayo is what it is. It is the quintessential definition of mise en place. There is not a single unnecessary component to make it what it is. It is culinarily perfect – even if you don’t like it. Mayo, like making wheels for a bicycle, is an art form of craftsmanship. The ingredients are simple: eggs, oil, mustard, lemon, water, salt, and spice. You can make it with a ridiculous amount of junk in it like pureed avocado or sriracha or Frank’s (on edit: Hot Sauce), or chopped up pickles, or vinegar or what have you… but if you can’t combine the first set perfectly – somebody is going to taste ‘something off’ and not be able to figure out what it is.

… Here’s the thing: big mayo makes this stuff surprisingly similar these days. There was a period where they stopped, but unsurprisingly folks have come around to understanding that their changes changed both the flavor profile too much, as well as were things that made people walk away from the product. So mayo has largely gotten back to its roots. I am of the belief that mayo requires egg – which means that you can have avegan mayo substitute, but calling something mayo that is vegan is questionably honest and causes me to raise an eyebrow to your understanding… Don’t get me wrong – I don’t think that there is anything wrong with a vegan product, its just the misleading equivalent of ‘frozen dairy product’ being equated with ice cream. I think you can get some great tastes with sandwich spreads, but thickening salad dressing emulsifications does not make you a mayo without the base ingredients.

… The numbers didn’t lie. Mayo is a full on red-state established food product. Even ‘light mayo’ is red state. If you want to attract blues – Olive Oil and Organic. Everybody loves squeeze mayonnaise. The quantity consumed will be much lower in the blue states, but – full stop – they don’t consume nearly the same quantity.

The fall-off for traditional, store bought mayo purchase is a death curve aligned to the baby boomers, with millennials purchasing some, but then only the ‘squeeze’ form factor for (assumed) sandwiches. For those that bought, the purchase cycle by unit was uniform across demographics, but looking out actual ounces – the older you were the bigger the containers that you bought and the less-healthy your containers were.

More Calories Please

Public Health England now recommends that:

…adults try to limit the calories of their three main meals to 400 for breakfast and 600 each for lunch and dinner […noting] that the remaining calories of the daily guidelines – 2,000 for women and 2,500 for men – are likely to be made up of snacks and drinks.

To achieve this, the government is challenging the food industry to reduce calories in products consumed by families by 20% by 2024. The categories of foods this applies to includes pizzas, ready meals, ready-made sandwiches, meat products and savoury snacks. It’s likely they’ll achieve this target with smaller portion sizes, which fills me with deep and profound sadness.

I successfully weaned myself off unhealthy snacks and drinks several years ago, so I tend to eat bigger portions for my main meals. There is no world in which I want to eat just 1600 calories for breakfast, lunch, and dinner, with the remaining 900 calories on snacks. I can manage this fine when I’m preparing food for myself at home or work, but when I’m on the move it’s clearly going to become harder and harder to find ready-made lunches that have 700-800 calories without adding on crisps or snacks or whatever.

I’m not a nutritionist or a food scientist, but I do know that barely anyone eats the suggested servings for things like breakfast cereal. Just try weighing out 45 grams – it looks like nothing. If retailers have to reduce their portion sizes, I wouldn’t be surprised that people just end up buying multiple portions – a hugely wasteful practice.

Others feel the same way. Susan Jebb, professor of diet and population health at the University of Oxford was interview by the Guardian:

While she welcomed raising calorie-awareness, she noted that the recommendation to eat a total of 1,600 calories for main meals was well below daily levels and assumed people were snacking. “Maybe it is better to have a slightly bigger meal and not to snack,” she said.

Sentience Footprint

I’m confident that in a hundred years, eating meat will be regarded in the negative way we now view racism or sexism – an ugly, demeaning, and unnecessary act. Like smoking, it will simply fall out of fashion because we’ll find better and healthier alternatives, although we’ll still occasionally eat humanely reared-and-killed animals. Note that I still eat meat even though I should know better.

The interesting thing about eating meat is that it encapsulates a multitude of sins. You might worry about its impact on your own health; or perhaps on the environment, given the amount of water and land that a cow requires and the methane greenhouse gases it produces; or of course, on the life and suffering of the animal itself.

From an environmental standpoint, we should be eating far fewer cows and far more chickens, since the latter require less energy input to grow for a given calorie, and therefore (all things being roughly equal) produce less of negative impact. Or we should forget about the chickens and eat sustainably caught-or-farmed fish, which are even more energy efficient and have the smallest carbon footprint.

But what about from a suffering standpoint? You can feed far more people with a single cow than a single chicken, so if we want to reduce the suffering of animals, maybe we should be eating cows. But are cows more sentient than chickens? I don’t know how you measure that. And maybe the environmental impact of a single cow produces more suffering on other sentients than a chicken.

I feel like I’m taking utilitarianism to a place far beyond its ability to survive. I should probably read more Peter Singer.

Serving Sizes

Serving sizes are a joke. Behold this bag of popcorn:


“106 calories per serving,” it proudly proclaims. You quickly do the mental calculation – that’s a mere 5% of your recommended daily allowance! Even better, it’s “wholegrain” and “high in fibre.” You munch on what you consider to be a serving of popcorn, safe in the knowledge that you are surely one of the healthiest and most responsible humans alive.

Some time later, you wonder: exactly how big is a serving? It says it right there on the front, in tiny writing: 20g. But how much is 20g? Is that a lot? It’s a sixth of the bag, which strikes you as being pretty small since you normally have at least a third of a bag at a time. So you weigh it out:


In case it isn’t obvious, 20g of popcorn is not a lot of popcorn. You’d be embarrassed to give that out to kids at Halloween. It is in no way what a normal human would consider to be serving. Of course, neither are most servings.

Why We Go to Starbucks

At Wordstock, where I was improbably on the same (short) bill as Alex Garland and Andy Serkis, I had an interesting conversation about Starbucks.

Why do we go to Starbucks? To get coffee, of course – and maybe to get something and sit down. That’s not an interesting question. What’s interesting is why people who are explicitly unhappy or at least ambivalent about Starbucks continue to go there. There are any number of reasons to dislike Starbucks, from (perhaps) their tax avoidance policies to their mediocre coffee and their cookie-cutter similarity. And yet millions of people still go.

So, why? There’s convenience; if it’s the closest cafe to where you are, that’s a strong incentive. There’s security; if you don’t know the area, you could either go to a non-chain cafe and risk a bad experience, or go to a known quantity. Relatedly, there’s consistency; you know what you get at Starbucks. Their marketing and decor reinforces those latter two qualities; you’re meant to feel at home.

All of these reasons added together seemingly outweigh any misgivings we might have about contributing towards unethical corporate behaviour or the increasing soullessness of our high streets. If we left it there, this would be a rather depressing statement of our values, but I don’t think that’s true. There’s really nothing wrong with people wanting convenience and security and consistency. And if Starbucks provided better service or quality or costs compared to their competition, it’s not at all surprising they succeeded. Yes, we all know their tactics in flooding towns and cities with chains to extinguish the competition and then pulling out; but even today, with the rapid growth over, I still seem to gravitate to chain cafes despite the existence of alternatives.

There’s an independent cafe near where I live called Bread and Bean. It makes very good coffee and it’s in a reasonably decent location. The atmosphere is not amazing and the decor a little spartan, but it’s a pleasant enough place to read a book. I used to go to it fairly frequently until I starting shopping in another nearby area which had at least three chain cafes: Costa Coffee, Starbucks, and Harris and Hoole. I ended up going to the latter, which is, if anything, an uber-Starbucks: even more comfortable and secure.

Harris and Hoole, of course, is 49% owned by Tesco. Most people don’t know this and assume, like I did, that it’s just a really nice independent cafe. It gets the best of both worlds – it’s not Starbucks, yet it exudes the comfort and security that we all apparently want.

In the face of these well-funded chain cafes, it may be wise to simply throw up your hands and declare defeat. I mean, it’s just coffee, after all – who really cares that much about a few million in tax? But it’s not just that. These places are the new public square. It would be better if our squares weren’t all owned by massive corporations, just as it would be better if Twitter and Facebook and Google didn’t control so much of online communication.

I wonder if the recent backlash against Twitter and Facebook in the form of more ad-hoc spaces such as tilde.club and private Discourse forums point the way forward. These spaces are no less secure and consistent than Twitter thanks to the diligent efforts of their founders; they are immune against private ownership since they are open source and capable of being spun up and run on any $10/month server; and they are certainly more comfortable, not because they have fancy features, but because they lack scale and so are more able to adapt to their patrons. Every branch of Starbucks must have the same coffee and decor – that’s a strength, but it’s also a weakness.

What we need is an open source cafe blueprint; something that amateurs and semi-pros can get up and running without a massive effort, yet still provides good service and good coffee. This may, in fact, be completely impossible due to the significant differences between running a website and running a physical cafe – but of course, “software is eating the world”. Maybe we can semi-automate all the legal and financial shenanigans involved in getting a space and running a cafe; maybe we can create a framework to source all the equipment and consumables required.

And if you can do that, you can probably adapt it (with great effort) for restaurants and shops and whatever. It would be a massive effort requiring close co-operation with national and local governments and banks; probably on the scale of writing something like Unix, really.

Update: Via his related blog post, John Willshire tells me that 95% of all cafes in Australia are independently owned using some of the tactics mentioned above. Turns out we don’t need to write Unix again – hurray!

Sweet sweet corn

One of the things I love about going abroad is the fact that the food is always cheaper and better (at the same price) than London. Sure, London has good food – if you can afford the money and time to check it out. When you’re on holiday, not only does the rest of the entire planet offer cheaper food, but you have more time to appreciate it. So when I went to Toronto for a week, I sampled a rather large number of restaurants, pretty much all of which were good.

However, I’m a little sad that I only got to eat one ear of corn when I was out there. For reasons that are still unknown to me, corn in North America is far sweeter and tastier than our so-called ‘sweetcorn’ in the UK. I know for certain that most of the corn I eat in North America is grown fairly locally, whereas UK sweetcorn seems to come from all over the place, including England. But while I have never had a disappointing ear of corn in North America, I count myself astoundingly lucky if I have a decent bit of corn in England.

For some time – before I first travelled to the US – I thought that UK sweetcorn was how real corn tasted: not bad, but on reflection, not particularly sweet. This changed when I travelled. I then mused that perhaps the corn plants I’m eating are just a different species, or maybe the corn was harmed by the way it was transported.

Like a good scientist, I investigated the latter possibility only a few hours ago. In the interests of furthering human knowledge, I have published my findings below:

Is the taste of UK sweetcorn harmed by the way it is transported? (to be published in Nature 439:7208)

Introduction: Most sweetcorn in the UK is bought from supermarkets, who typically package their sweetcorn in film-wrapped packages that are likely to have been in transit for several days or even longer. They may have also undergone additional treatment during the packaging and transportation process, and other treatments associated with the mass production of food. In contrast, the sweetcorn the experimenter (Adrian) has tested in North America has typically been ‘raw’ ears of corn, unpackaged; this type is also available in the UK, but this experimenter has not tasted it.

This experiment will compare the taste of UK ‘raw’ corn to UK supermarket corn.

Hypothesis: The taste of sweetcorn is harmed by packaging, treatment and distribution processes associated with supermarkets, but not associated with ‘raw’ corn bought from healthy shops, etc.


Sweetcorn bought from ‘Fresh & Wild’ (wholly owned by ‘Whole Foods Market’)
Heating panel

Method: The green bits on the sweetcorn were removed. The sweetcorn was split in two, in order to fit it in the pot, and then both halves were submerged in water. The pot was placed on the heating panel and a pinch of salt added. Heat was applied until the water began to boil. This heat was maintained for three to four minutes, until which point the sweetcorn was removed from the pot and left to cool for five minutes. Before eating, margarine and salt was applied.

Results: From a qualitative perspective, the sweetcorn tasted significant less sweet and ‘tasty’ than North American corn. It tasted essentially the same as supermarket-bought sweetcorn, although perhaps slightly better due to the method of cooking.

Discussion: It appears that UK corn is significantly worse than North American corn, no matter where it is bought from or how it is distributed. However, the sample size in this experiment is small (n=1) and it is possible that there are other sources of sweetcorn in the UK that are better. Even so, it would be expected that a retailer such as ‘Fresh & Wild’ would source tasty corn, meaning that if tasty corn is indeed available in the UK, it is baffling that F&W would not sell it.

Conclusion: Based on the very small sample size of this experiment, it seems that it is not possible to get ‘good’ corn in the UK. Still, additional investigation is required. Increasing the sample size and acquiring different sources of corn is vital. Secondly, it is advisable that the experimenter travels back to North America to perform additional taste tests on their corn. Funding for this travel will be applied for to UK research councils in the near future.


I was on my way into Marks and Spencers foodcourt today when I was stopped dead in my tracks by the sight of self-service checkouts. I know these checkouts are not uncommon in the US but I’d never seen them anywhere in this country until now. For the uninformed, self-service checkouts allow customers at shops and supermarkets to check out their own goods and pack them into bags themselves. Security is ensured both by placing the bag racks on a balance that cross-checks the weight of the goods in them with those that youv’e scanned, and by the presence of a highly suspicious supervisor.

Theoretically, self-service checkouts are a good solution for the shop and the customer because they drastically reduce labour costs and decrease checkout time. I put the latter part of the equation to the test today.

The self-service checkouts are about a metre and a half long and consist of a LCD touchscreen, a barcode detector and a bag rack (among other things). When you stand in front of the checkout, you can either start passing your items through immediately or press ‘Start’ on the touchscreen, which I did and launched the machine into an excessively polite and cheerful introduction into how to use the checkout. It really is as simple as you’d imagine – you scan your items and put them into the bags. As you scan items, their information and cost appear on the screen. I did have a problem in one item not appearing when I scanned it but that may have been due to it being in the wrong orientation.

When you’ve finished, you press a button on the screen and choose a payment option – credit or debit cards, or cash. Cash can be inserted into the machine, and credit cards can be swiped. There’s a signature panel on the side of the machine which you have to use in the latter case, and at the end you get a receipt printed. Throughout all of this, the instructions continue via videos and audio.

The entire process took rather longer than it would’ve if I went to a normal ‘assisted checkout’ due to the novelty of the experience but I imagine I could cut the time down quite a bit on a second visit. I doubt that I could ever be faster than an experienced checkout person but I could give a good shot at beating many of the slower ones I’ve encountered. The argument could be made that this means self-service checkouts would in fact not save the majority of people’s time, but that’s neglecting the fact that you could fit more self-service checkouts in the same space as normal ones, thus reducing queue length.

I doubt that shops or supermarkets will ever eliminate normal ‘assisted’ checkouts simply because there are some people who will not be able to use (or intensely dislike using) self-service checkouts. This suits me fine – the slow folks can go with the normal checkouts, and the quicker, more froody people can go with the self-service ones. This will save a lot of time, I expect. Of course, it’s only a matter of time before they have to replace the whole lot with RFID scanner gateways.


The great thing about not regularly drinking caffeinated beverages is that when you do, you get a real kick out of it. I basically never drink anything that contains any caffeine – no tea, no Coke and definitely no coffee. Every so often though, when I drink just a single cup of tea, I’ll feel noticeably stimulated and full of energy. It even stops me from going to sleep during lectures.

I don’t intend to make a habit of drinking tea though; clearly its power is dependent on its sparing use. In fact, I would say there’s a negative exponential correlation between the frequency of tea drinking and its power; the longer I wait, the more powerful its punch. I can only imagine that the same is true (but moreso) with coffee. Since I’ve never drunk a full cup of coffee in my life, the energies waiting to be unleashed are surely beyond the ken of anyone on this world. No doubt, when it finally becomes necessary for me to harness this awesome power, it will be for some great and desperate task that my country, indeed, Earth itself, requires from me…

GM Spin

I posted a comment in this MetaFilter thread about GM crops, on how research in the area is often misrepresented by the anti-GM lobby. Case in point: it was claimed that the Bt toxin pesticide might actually benefit some pests, meaning that transgenic Bt plants could be utterly counterproductive. In reality, the research data has no grounds for such a conclusion.

Some thoughts on King Street

King Street is a long road, probably a couple of miles in length, that forms the heart of Newtown, a district south of central Sydney. The rather imaginatively named Newtown was the third town after Sydney and Paramatta to be founded in Australia, and now it’s been completely subsumed by Sydney.

I’m reminded of places like Cowley Road in Oxford when I walk down King Street; while a little run-down in appearance, almost all of the shops are conducting brisk business and notably there are around two hundred restaurants dotting its length, from all over the world. Japanese restaurants, Thai, Singaporean, Chinese, Italian, fast food, sandwich shops, Vietnamese, all sorts. I’m told that the Thai restaurants in particular are engaged in a sort of punning Cold War, each having increasingly awful names such as ‘Thai Tanic’, ‘Thai Me Down’ and so on. The other shops on the street haven’t been spared either, with one bedclothes shop named ‘Holy Sheet!’

The friends I’m staying with have been here for over a decade and originally they intended to visit every restaurant on King Street from one end to the other. Unfortunately, unless they went out every night (and of course they didn’t) by the time they got any appreciable distance down the street, there’d already have been a fair amount of turnover in the restaurants open anyway.