2018-08-10

Planning & Strategy.

• In a mix of strange tweets + a somewhat more well-reasoned blog post, Elon Musk announced that he wanted to take Tesla private - sending the stock price up and then down and prompting both an SEC probe and an investor lawsuit. An interesting factor here: If Tesla's stock price isn't over $360 on 2019.03.19, a $920 MM bond will probably need to be repaid in cash - cash that Tesla historically has not had. Note: I am proud to own a very small amount of Tesla stock.
• DARPA is hoping that an older (90-nm; 10-nm chips are common in consumer products today) chip foundry can use "monolithic 3D integration" to make chips that'll continue to boost performance as Moore's law ends.

Making & Manufacturing.

• how_shits_made on Instagram: A really good source of on-the-ground photos/videos from factories, mostly in China. A *big* net positive in my feed for sure.
• A developer in Berkeley has built the first prefab housing in the US that was manufactured in China; it will be leased to UC Berkeley for student housing. "Using prefab material is supposed to be less expensive than building from scratch, [the developer] said....But the savings haven’t been as great as expected, he said. 'Sixty-five to seventy-five-percent of the construction costs are still incurred on the site. In addition to the usual trades, we have crane operators, flagmen, truckers and special inspectors.'” Of particular note: "The cost of trucking to Berkeley from the port of Oakland was more expensive than the cost of shipping from Hong Kong."
• Misumi's application library for mechanical engineers. Lots o' robotic integrated systems here.
• Octopart (a sponsor of this newsletter) is now offering ECAD and MCAD models of "many of the parts" they list. In my experience, high quality documentation is a *big* need in the product development process, and has often driven my own decision making.

Maintenance, Repair & Operations.

• There's a Festival of Maintenance being organized in London in late September, and it sounds awesome.
• A college dropout designed a system for cleaning up ocean plastic, and sent it out into the Pacific ocean for a test last week.
• Second Ave Sagas on the Supplemental Environmental Assessment that the MTA released for the 2019 L train shutdown.

Distribution & Logistics.

• Walmart's plan to pay store employees to deliver packages on their way home has been terminated. As I said a little over a year ago: "[The article] points out that 'about 90 percent of the U.S. population lives within 10 miles of a Wal-Mart,' which is crazy, but I wonder if this particular spin on the traveling salesman problem doesn't introduce unique complications. Also, I live within 10 miles of two or three Wal-Marts, but those are all 40+ minute drives...In other words, distance might not mean as much as you might think."
• Drew on the future of autonomous cars: "I spent a few days last week in a series of thinly-populated Scottish islands that collectively boasted one ATM, three taxis, and almost no cellular service. In a hypothetical future scenario where self-driving cars have achieved a level of global adoption on par with credit cards or mobile phones, what would be happening in these remote places? Would there be five or six cars that busily moved everyone around each island, or would the low costs of operating the vehicles result in a more exuberant "loitering car cloud" that ensured nobody had to wait more than a few minutes for a ride? Well, the islands don't have ATMs or cell reception today, so there's little reason to assume they'd have any self-driving cars in such a scenario, even if most other places did. We often imagine that driverless cars will turn mobility into a hardware-dependent service with minimal marginal costs that is ubiquitously available, but mobility could also just become another version of cellular service, usually good enough everywhere but plagued by small geographic pockets of unavailability, even in large cities ('let me meet you down the block, I'm not getting any mobility right here')."

Inspection, Testing & Analysis.

• Tesla's Model 3 did *not* perform well on an automatic braking test.
• The NYTimes did a very interesting analysis of marathon running time data from Strava, and used it to evaluate the effectiveness of various running shoes - most notably the Nike Vaporfly 4.

Tangents.

• On the myriad difficulties in running porn sites. Customers of porn sites frequently lie to their banks and say their credit card charges are fraudulent. So banks simply refuse to process charges made to porn sites in the first place, forcing porn sites to outsource their billing to shady third parties, which makes customer service really hard and just comes across as scammy. "If anyone wants to make a killing, seriously, build a proper billing system for this industry."

An excellent analysis of how megacities across the world are growing vs. how active their airports are.