Profile cover photo
Profile photo
Jean-Baptiste “JBQ” Quéru
45,521 followers -
Traveler, Skier, Cyclist, Photographer, Eater, Drinker, Driver, Commuter, Board Game Player, Video Gamer, Astronomer, Painter, Knitter.
Traveler, Skier, Cyclist, Photographer, Eater, Drinker, Driver, Commuter, Board Game Player, Video Gamer, Astronomer, Painter, Knitter.

45,521 followers
About
Jean-Baptiste's interests
View all
Jean-Baptiste's posts

Cautious / Curious

For a variety of reasons, I've been recently looking at US politics much more than I had had in the past.

So far, US politics hadn't made much sense to me. While I understand the mechanics, I had been struggling to understand people's motivations to vote for one side or the other, as those didn't seem to make sense. The confusion went as far as my own position, where I didn't actually understand why I would tend to align with one specific side (disclaimer: I can't vote in the US, I'm not a US citizen).

After reading a lot on the subject, and after simplifying the matter down to a level where I could grasp it, I've reached a tiny bit of clarity. The underlying observation is that people fall into two main categories, which I call the cautious and the curious.

On the cautious side, people focus primarily on the potential risks that come with any situation, which is a form of pessimism.

On the curious side however, people focus on the potential benefits that come with any situation, which is a form of optimism.

I'm going to assume that many of my followers here are interested in technology, so I'm going to use an XKCD example: https://xkcd.com/627/ ("Tech Support Cheat Sheet"). This XKCD flowchart shows the way of thinking in the curious side of the world, especially the part that amounts to "Pick a menu item or button at random, click it. Repeat until you've tried them all." The cautious side of the world, by contrast, is afraid of breaking things, and has a much harder time trying new things.

I can think of my own family here. I can look at my mother and her siblings, where my mother is curious, while the other siblings are on the cautious side, especially the older sister. I can think of my paternal grandmother, who has a computer but is very much afraid of breaking it. I can think of my late maternal grandmother, who wouldn't let us connect a computer to her TV for fear of damaging it.

On the cautious side, it is natural to try to do things they way they've always been done, since doing things differently introduces potential risks. On the curious side, it is natural to try new things, since that introduces potential benefits. Inherently, cautious people are conservative, and curious people are progressive.

From there, we can make a number of guesses. As an example, when faced with an unexplained situation, the curious are more likely to explore, to look for an explanation, to understand what gains can be achieved and whether those gains are repeatable. In a similar situation, the cautious are likely to retreat, in order to avoid triggering any bad results, and to find an explanation that doesn't require to interact with the situation. In other words, in order to fit the unexplained into their world view, the cautious are more likely to resort to religion, whereas the curious are more likely to resort to science.

On the cautious side, which prefers to do things they way they've always been done, there's a natural inclination to respect the authority of leaders. That includes respecting parents and older family members: after all, the simple fact that they're still alive shows that they've been able to avoid risks. In general, the role of leaders is to protect people from risk, including by preventing behaviors that leaders believe might be risky.

Also on the cautious side, there's a logical trust that binds small groups together, which creates strong loyalty within those groups: under the instinct that interactions with strangers might carry risks, people bind in groups within which the interactions have proven not to trigger risks. Once a group is established, it creates a notion of insiders and outsiders, where the outsiders are assumed to be sources of risks.

Still on the cautious side, respect and loyalty somewhat combine and evolve into a notion of sanctity: sanctity is a line that defines some common traits and common symbols that represent or symbolize the authority and the group. The fear of risks prevents from exploring the other side of that line, which is then assumed to be a slippery slope of risks, and the line itself must be protected in order to keep risks away.

On the curious side, the natural inclination is to explore, to research new things, to improve lives, to look for new possibilities. Inherently, the process of trying new thing is a solitary endeavor, an individual process, since that process can't directly rely on accumulated experience. In order for a curious person to be able to drive their individual progress, they need a fair society around them that will support their explorations. That notion of universal fairness is far broader than the one that naturally emerges within loyalty groups on the cautious side.

Finally, on the curious side, while the instinct goes toward trying new things, there's also an understanding that there are inherent risks in such endeavors, and as a result there's an imperative to always consider the possibility of harm to others any time something new is attempted, since society obviously can't accept reckless behaviors. Society can only accept new behaviors is those new behaviors are built within a framework of care and avoidance of harm to others. That notion of universal care and avoidance of harm is far broader than the one that naturally emerges within loyalty groups on the cautious side.

Those traits, taken together, paint a deeper picture. The cautious side, which is the conservative side as we've discussed earlier, expects authority, loyalty, and sanctity, as means to stay away from risks. The curious side, on the other hand, relies instinctively on fairness and care in order to facilitate new things. Those values that are important to the curious side are the inherent values of liberals, and at that level of abstraction the other side is authoritarian.

In my mind, that's the basic foundation of US politics, which crosses a progression of pairs of contrasting concepts: cautious - curious, optimistic - pessimistic, conservative - progressive, liberal - authoritarian. It certainly helps me frame my own vision of US politics, since in my mind the words "conservative" and "liberal" themselves aren't inherently antonyms, and we've just established that they're different levels of social and political abstraction. In a context of trying to find a balance, maybe a middle ground, it's probably worth it to frame the discussions as conservative / progressive or as liberal / authoritarian.

From here, we can look at a number of dimensions that arise in politics and other domains under the lens of the curious - cautious divide and the various related aspects discussed above.

On the curious side, because of both the natural inclination toward science and the need to consider humankind a collection of individuals, it's fairly natural to rely on statistics to guide decisions. On the cautious side, the focus on negative outcomes and the higher trust placed in a small group means that it's more intuitive to focus on anecdotes. We saw that divide a few decades ago when seat belts appeared in cars: statistically, seat belts save lives, whereas anecdotally, there are some cases where seat belts cause injuries. We see that even more when looking at people's behaviors toward flying: optimists point out that, statistically, flying is far safer than driving, but pessimists' minds can't detach from the massive but rare news reports of catastrophic airplane crashes; cautious people's inclination toward respect within the group doesn't help here, since elders remember a time when airplane crashes were more frequent; on top of that, the more uncommon airplane crashes become, the more prominently they get covered in the news. Vaccines is another similar domain: statistically, they're far better than the alternative, but there's a non-zero risk of adverse reactions; we've stopped vaccinating people against smallpox, since the statistical risk of not vaccinating has fallen to zero, whereas the risk of a vaccine is never quite strictly zero. One last similar domain is health care: proponents of recent health care reforms point out that the number of insured people has increased, while adversaries single out that some people have seen their premiums increase.

Feminism can be looked at under the lens of that divide. The cautious side looks under a perspective of hierarchical authority, where increasing women's role threatens the established hierarchies. The curious side tries to increase fairness between men and women. The cautious side fears the risks that women could be exposed to as they leave the group. The curious side worries about the harm that women can suffer if they can't speak for themselves. The cautious side sees women's role in having children as sacred. It's no wonder that it's a contentious subject in politics.

Attitudes toward guns fit within some of those patterns. Risk-averse cautious people see guns as an effective method of protection, which can be entrusted to authority figures who will protect both the group and its sacred values, and which are only intended to be used against outsiders, whereas the other side gets fixated on the harm that guns can cause, as a statistically significant cause of preventable deaths in the country.

The Black Lives Matter movement is also interesting to look at. The liberal side sees clear issues of fairness and care. The authoritarian side sees an attack on the respect that is owed to law enforcement agencies, which compounds in white conservative circles with a lack of loyalty toward blacks and with a treatment of whiteness as sanctity.

It's revealing to look at the way people's political views change with age: over time, more and more of people's curiosity ends up with no tangible benefits, while more and more unsuspected risks end up causing harm, such that the sense of caution grows and people's political views tend to shift toward the conservative side. That's why younger voters and consistently more progressive.

Since I work in the domain of computer technology, it's insightful to think of my own industry along those dimensions. Computing is very much on the curious side of the divide, where survival requires to constantly do new things. In fact, the primary source of initial funding is called venture capital for a reason, as a warning to risk-averse investors that hey should stay far far away. As such, it's no surprise that technology companies tend to lean toward the liberal side. The industry is filled with the moribund shells of companies that didn't see progress as a key to survival. IBM said there was only a worldwide market for 5 computers, and eventually DEC happened. DEC couldn't imagine the notion of a home computer, and ultimately Microsoft happened. Microsoft claimed that Apple was dead, and the iPhone happened. I won't be surprised if Apple's ultimately failed assumption ends up being that smartphones are a luxury. Within the tech world, the bias toward curiosity is harmful, since the industry keeps underestimating the difficulties that some users will have to adapt to any change.

Fragments of sentences that are said by conservatives and in turn trigger sharp reactions in liberals, such as "I'm not racist, I have black friends" or "As a father, as a husband," take a new meaning with the dimensions we've established above. One side puts a high value on fairness, and gets offended at the notion that the care for a vulnerable population gets trivialized as an anecdotal relationship. The other side puts a high value on group loyalty and therefore cares much more for people within the group, and gets offended at the notion that outsiders need to be treated with the same consideration as insiders.

Finally, there's a grab bag of notions that the right advances and that the left rejects, such as the homosexual agenda, the coastal elites, or other similar theories. The difference in opinion about those comes from the different perspectives around groups and hierarchies. The cautious side grows notions of belonging to groups and of conflicts between groups that the naturally individualist views of the curious side don't take into account. At the same time, the cautious side is accustomed to a hierarchical notion of authority, where the most likely relationship between two people or groups has one dominating the other; in that perspective, it's easy to picture oneself as the dominated side when you're clearly not the dominant side, even if the intended nature of the relationship is simply coexistence.

Personally, framing the partisan gap in US politics as an issue of curious vs cautious helped me. It also shows me that the path to repair the cracks in our society will require that everyone think about authority, fairness, loyalty, care and sanctity, even though only a few of those are dominant for most people.

Post has attachment
Inverted price curve

This is a scary article: http://www.latimes.com/projects/la-fi-electricity-capacity/

One implication of this article is that the utilities are guaranteed that they can recover certain fixed costs from consumer bills. In other words, a part of the electricity price goes up when demand goes down.

On the other hand, there's still some market logic where lower demand lowers the cost of production, by allowing utilities to shut down the most expensive sources.

This feels very risky to me: with more and more houses using rooftop solar, there's mathematically a possibility that the price curve could become inverted, i.e. that electricity could become more and more expensive as there's less demand for it.

That would end up being very regressive, since people who can afford the upfront cost of solar panels would have such panels, while people who can't afford panels would end up having to pay the increasing utility rates. I don't actually expect that CPUC will let California get into such a death spiral, and sadly I have to expect instead that California taxpayers will eventually have to step in and bail the utilities out.


Oroville evacuation

If you live in the region downstream from the Oroville Dam that is affected by the evacuation order, if you're currently in the Bay Area or in the San Joaquin Valley, if you can't find a place to stay, please leave a comment. I have a guest bedroom available, with the possibility of turning an office into a second guest bedroom.

Update, 2/12 8:30pm: The lake has come back down below the level of the auxiliary spillway, such that there doesn't seem to be any immediate risk of flooding.

Update, 2/12 10:10:30pm: The evacuation order remains in place for the foreseeable future. My offer still stands.

(Reshares intentionally disabled so that I can update the post if/when I run out of room).

Post has attachment
US Manufacturing vs Trade

This chart (h/t +Riff Zifnab) tells me a lot about the last 25 years of US manufacturing. Caveat: the scale for the blue curve doesn't start at zero.

Here's what I read in it, at a high level:

-The US didn't lose manufacturing to trade, since manufacturing went up after trade agreements were signed in 1994 and 2001.

-US workers didn't lose jobs to immigrants, since that would typically result in an increase in the workforce (more workers, paid less, individually less productive). Instead, the jobs truly disappeared.

At a slightly more detailed level, here's the timeline I can reconstruct from that chart:

-1994, NAFTA. Within 2-3 years, US manufacturing output increases as a result of the newly opened markets. The extra profit is invested into robots, which makes more economic sense when the output is expected to be durably higher. 5 to 7 years after the increase in output, the investments cause workers to get laid off as their jobs are replaced by robots.

-2001, China in WTO. I can copy-paste the exact same timeline as above: increased output within 2-3 years, followed by a sharp drop in manufacturing employment 5-7 years later.

Because the output rises before employment plummets, I'm dismissing the theory that trade causes extra competition and that automation is the only way to remain competitive: in that theory, output should fall after 2-3 years and recover after another 5-7 years, and that's not what we're seeing here.

Looking ahead, we can predict what could happen after trying to tighten existing trade links. Output would drop as a result of losing those extra markets. The result of that drop in output would be a new wave of job losses, as the investments in automation wouldn't be undone. That's encouraged by accounting rules, where the amortization of the automation capex would have to be accelerated, possibly as a writeoff, which would be reported as a massive loss.

My gut feeling: if we close US trade borders, we will end up losing manufacturing jobs, not gaining them. At the same time, the price of manufactured items for which we have no significant production capacity left in the US would go up, which would hurt US consumers, and that would especially hit consumer electronics.

Powerwall 2 vs Whole House?

I've ordered a solar system from +SolarCity a few days ago. My wife and I had wanted to wait until the +Tesla Powerwall 2 was available, so that we could implement our own load shifting.

I was surprised to hear from my SolarCity salesperson that the Powerwall 2 can only power a single circuit in the house. Our electricity usage that would benefit from load shifting is split between HVAC, kitchen, car charging, TV, and we can't reduce that to a single circuit.

Needless to say, this was disappointing. Is the single-circuit restriction accurate, given that Powerwall 1 looked like it was able to connect on the DC side of the solar system and from there power the whole house? Does any of my followers already have a Powerwall 2? What's your experience with it?

Manufacturing and High-Tech

I've been asking myself a lot about the state of manufacturing, and what made US manufacturing so strong a few generations ago.

I think I've been able to figure two aspects out.

One aspect was the historical conjuncture as a result of WW2: most other countries that had had a manufacturing past had been devastated by the war: the infrastructure was damaged, the factories were destroyed, and many working-age men had been killed. By comparison, the US was left relatively unscathed, and in fact had built a massive industrial capacity during the war, as well as trained a lot of extra (female) workers. From that point of view, the US was the world leader in low-cost manufacturing. Conditions where a war affects the whole world but not the US are unlikely to ever come back, both because enough industrialized countries have nuclear weapons to act as a deterrent against a world war, and because manufacturing industries are far more scattered around then world than they were 80 years ago and would therefore be unlikely to all get destroyed in a world war.

The other aspect is that manufacturing was the high-tech of the time. Consumers were buying fridges, washing machines, cars, TVs, all of which were high-tech items at the time. A lot of the work toward modern electronics happened between about 1965 and 1975, and that ended up taking over as the high-tech industry of the time, which is where US leadership shifted. Later on, where we are now, the highest layer in the technology stack is software. Really, looking at history, the US has been the hotbed of high-technology for a century, but the incarnation of high-technology has evolved over time, and we cannot assume that the US dominance over a given industry can be long-lived.

What I see, therefore, is that manufacturing in the 1950s was at a crossroads that is unlikely to occur again. The highest technology at the time was still being manufactured, such that only the US could run such manufacturing. At the same time, the US had the cheapest labor of any industrialized country. Those two conditions aren't likely to come back, not even individually, and most probably not together. The US cannot regain its dominance in manufacturing, because manufacturing is now both too easy and too widespread.

Chevrolet Volt

Last June, I bought a 2016 Volt. I figured I'd share a few thoughts about it. For clarity, I think about it as an electric car that can fall back to being a hybrid car, and I'll write that way.

For background, I had been thinking for a while that my next small car would be at least a hybrid. For the rare cases when I don't take public transit, my drive to work is 54 miles each way (87 km), with the ability to charge at work (but no guarantee). Slightly more often, I drive to the nearest BART station, 29 miles each way (46 km), and I can't charge there. I routinely need to drive 15 miles (24 km) in town for various errands.

I wasn't impressed by the Prius PHEV (the Prime didn't exist yet), for its small electric range, which felt too short for my use cases to justify the cost. I also wasn't impressed by the original Volt (which didn't seem to do a good job when driven as a hybrid, and whose EV range was then too short for my usage). Even with the latest version, the Leaf's range was just a bit too short to make me comfortable driving to work, in case I couldn't recharge there or along the way. I had been planning to go toward a Prius C.

My wife suggested to look at the second-generation Volt. Given that I had ruled the first generation out, I didn't expect to be impressed. As it turns out, it looked good enough on paper.

Here are some thoughts about it:

-I find that I like driving an EV much more than I anticipated. It's both very smooth and very responsive (especially when set to Sport mode). This is actually a driver's car, though obviously not a sports car. My overall experience with the drivetrain is very good.

-The factory tires suck, in the same way the factory tires sucked on my Camaro many years ago. They can't deal with the weight of the car under spirited driving. That's a small detail since they can obviously be changed, and there are good summer tires available in the exact same size. Since the gas engine relieves me from range anxiety, I can afford to lose a bit of range in exchange for grip. I haven't done that yet, though.

-The oil in the engine is specially formulated to stay chemically stable for 2 years. At the current pace, my own driving is on track to use less than a tank of gas per year, and I've had the gas engine turn on once for its maintenance cycle, which is what it does when you don't use it enough.

-I get significantly more range from it than its EPA rating: it's rated for 53 miles (85 miles), I typically get about 60 miles (96 km) from a charge, and I have driven as many as 70 miles (112 km) on a single charge. Unlike plain gas cars, the Volt gets a much better mileage in slow traffic than at full freeway speed. I seem to get about 44 mpg (US) on gas (53 mpg UK, 5.3 l / 100km, 19 km / l), better than the EPA-rated 42 mpg.

-I get the battery empty or near-empty at least 3 times a month. I feel that I actually use all the battery capacity that I paid for, which wouldn't be the case quite as much with a plain EV (or at least it'd be very stressful).

-There's something very peaceful about putting the car in the garage, plugging it in, and it refills itself without drama. For day-to-day driving in town, even the trickle charge does the job well, and I haven't felt the need to invest in a high-speed charger. That makes the Volt a very practical option for people who rent a place with a private garage: charging a reasonable range is practical with a household outlet. On the other hand, outside of home, the fact that the high-speed charging only uses half of the capacity of a level-2 charger turns out to be more annoying than I thought, as it means that there's very little value in charging during small errands, and that public chargers that charge by the hour are significantly overpriced. GM shouldn't have cut corners on the level-2 charger, and should have included a true 30-amp circuitry.

-In a way that I hadn't anticipated, the Volt generates no waste heat when driven from its battery, such that in winter the car tends to stay cold inside. That explains why it's got heated seats and a heated wheel: those consume less energy than heating the whole car. I have found that using the AC in summer is less detrimental to the range than using heat in winter.

-The car itself is nothing to write home about. It's somewhat basic, somewhat cramped. The infotainment system mostly gets the job done, though Chevy's MyLink is well behind Ford's Sync in my experience, both in terms of features, in terms of usability, in terms of polish (my pet peeve is that it can't connect to different Bluetooth devices for telephony and for music). Keep in mind that there's no built-in map-based navigation, and getting a map on the car's display requires to use Android Auto or Apple Carplay. Knowing the technology industry too well, I have to assume that those will stop working before the car gives up, such that after a few years I expect not to have maps any longer.

-In terms of assembly, this is the first car I buy that was actually assembled in the US. Of my previous 4 cars (3 Ford and 1 Chevy), 2 had been assembled in Mexico and 2 in Canada. The Mexico-built ones (both Ford) had no assembly problems at all. The Canada-built ones (one of each brand) were mostly OK. This US-built Chevy is a definite notch below, and less than a year after assembly and barely more than 6 months of usage, some parts are already coming apart. This is an anecdote and is probably not representative of all US-built cars, of all US-built Chevys, or even of all Volts, but limited to my personal experience it's clearly visible.

-There've been 3 recalls on the car already, which isn't too surprising since I got an early build. Chevy's handling of those recalls is very poor, though, as it's very hard to get actual information about the actual issues, and because the dealerships are inflexible in the way they handle those issues. While my most recent Ford has had a couple of recalls as well, I feel that Ford did a far better job managing them.

-For me, the major letdown is the online integration. I can't find a single good thing to say about it. It barely works. The web site is bad. The mobile app is even worse. That someone would think that this is good enough is appalling. While it makes me feel good professionally to know that I'm surrounded by hundreds of people at work who would find this unacceptable, including and especially my CEO, as a user of this car the online systems are so poor that they hurt the overall ownership experience.

Overall, the Volt is satisfying me about as much as I had anticipated, with some areas far exceeding my expectations and others falling well short. With my experience, I would definitely consider a pure EV now, with the knowledge of what to look for, but that's unlikely to be a Chevy unless I can get evidence that they've made some significant improvements in a variety of areas.

Edit: clarified distance and mileage in other units.

Edit: mentioned that charging when renting a home with a garage is a practical option. Then re-edited for clarity.

5 whys and US manufacturing

There's been a lot of chatter recently about the state of US manufacturing, which has lost a significant number of its jobs in the last few decades, and a lot of discussion about the cause of that decline.

A common technique to investigate issues is called "5 Whys", i.e. the notion that the immediate cause of the issue we see might be the consequence of a deeper cause, and that iterating is more likely to identify the root cause, often requiring as many as 5 iterations.

At the first level, the commonly mentioned reasons for the loss of manufacturing jobs are offshoring (i.e. trade) and automation (i.e. gains in productivity).

We shouldn't stop here, though, if we intend to really understand and potentially if we try to affect it: stopping here would create the risk that we aim at a symptom and not at a true cause.

At the second level, we find that the reason for either offshoring or automation is cost: offshoring and automation are cheaper.

Yet one level deeper, there are two possible reasons why companies would try to reduce costs.

The first possible reason for trying to reduce costs is competitiveness: reducing costs might be the only way to get the price to a point where the market will bear it. In other words, market economy, indirectly through 2 steps, tends to cause US manufacturing jobs to disappear.

The other possible reason for trying to reduce costs is profit: lowering costs increases profits, and in a world of capitalism (and especially large-scale anonymous capitalism) increasing profits is a primary goal.

Now, capitalism and market economy are the pillars of the US economy, and there lies the difficulty: the actions that are likely to be effective at helping US manufacturing jobs will either get in the way of market economy by increasing prices higher than they would naturally need to be, or by lowering profits. Neither of those is impossible, of course, and in a number of scenarios they might in fact have beneficial consequences. However, we need to be conscious of one fact: helping US-based manufacturing jobs might require to get in the way of market economy, of capitalism, or of both.

Disclaimer 1: personal opinion, as usual.

Disclaimer 2: I'm really not an expert in that domain, and I'd love to hear from actual experts, for refinements or refutation.

Post has shared content
WWJBQD

Reportedly +Steve Kondik​​​ asks "WWJBQD?" (and I have no reason to doubt +David Ruddock​​​'s reporting).

The answer is easy: put users first. Find an opportunity where there's work to do that'll put users first. I'm doing it at Yahoo, I did it at Google, I had done it before. Putting users first actually simplifies a lot of decisions that would otherwise be too complex because of too many dimensions.

Post has attachment
"Others: 0.3%"

There's a lots of chatting today about recent smartphone sales numbers, but there's one aspect that I'm not seeing discussed much: the "Others" category fell from 2.3% to 0.3% in a year.

For me, as a developer of mobile apps, a lot of decisions are about deciding what I can get away with not supporting. So, I don't actually care much about how many users are using Android 7.x or iOS 10.x within their respective ecosystems (except when evaluating the priority of trying to work around the new bugs introduced by Google and Apple). However, I really care about how many users are still using iOS 7.x / 8.x or Android 4.0 - 4.3, as that actually influences my decisions.

On the side of all that are all the "Others", and a 7x drop year-over-year down to a fraction of a percent tells me that for now I don't need to worry about those right now, and this might remain true for at least a couple of years ahead. That's what web fallbacks are for.
Wait while more posts are being loaded