A drought in Taiwan is drying up reservoirs – and if it continues, could hit chip manufacture by summer. CC-licensed photo by http://www.dantw.com on Flickr.
You can sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.
A selection of 9 links for you. Safe journey. I’m @charlesarthur on Twitter. Observations and links welcome.
Too much time thinking about Trump? Preorder Social Warming, my forthcoming book, and distract yourself.
🚨 How Basecamp blew up • Platformer
Casey Newton, following up on his piece last week about how Basecamp got into a tangle. This feels explosive because of the implications around Ryan Singer, who was the CTO, and who has deleted a ton of tweets – and apparently kept posting Breitbart content approvingly in the company Slack:
»
On Friday, employees had their chance to address these issues directly with Fried and his co-founder. What followed was a wrenching discussion that left several employees I spoke with in tears. Thirty minutes after the meeting ended, Fried announced that Basecamp’s longtime head of strategy, Ryan Singer, had been suspended and placed under investigation after he questioned the existence of white supremacy at the company. Over the weekend, Singer — who worked for the company for nearly 18 years, and authored a book about product management for Basecamp called Shape Up: Stop Running in Circles and Ship Work that Matters — resigned.
Within a few hours of the meeting, at least 20 people — more than one-third of Basecamp’s 57 employees — had announced their intention to accept buyouts from the company. And while many of them had been leaning toward resigning in the aftermath of Fried’s original post, the meeting itself pushed several to accelerate their decisions, employees said. The response overwhelmed the founders, who extended the deadline to accept buyouts indefinitely amid an unexpected surge of interest.
This account is based on interviews with six Basecamp employees who were present at the meeting, along with a partial transcript created by employees.
«
Newton is doing terrific work; one of the best around in terms of the contacts and context.
unique link to this extract
This motorcycle airbag vest will stop working if you miss a payment • Vice
Aaron Gordon:
»
Airbag vests are pretty much exactly what they sound like, garments worn by people who undertake exceedingly dangerous personal hobbies in order to slightly reduce the risk of severe bodily harm or death. For example, in 2018 the motorcycle racing circuit MotoGP made airbag vests mandatory.
Since then airbag vests have become steadily cheaper and therefore more popular among recreational riders. One motorcycle apparel company named Klim, for example, sells an airbag vest called the Ai-1 for $400. In the promotional video launching the product, product line manager Jayson Plummer called the vest “a whole new era of a platform where analog meets digital and results in a superior protection story.” Which is an interesting way of framing the fact that the vest includes an additional subscription-based payment option that will block the vest from inflating if the payments don’t go through.
This is possible because the vest includes two components: the vest itself made by Klim and the airbag system including a small black box made by a French company called In&Motion called the “In&Box detection module.” The module has the sensors and computer components that detect a crash and make the bags inflate.
The customer buys the vest for $400 which comes with the module, but then they must download an app and choose how to unlock the module so the vest actually works: either plonk down another $400 to own the whole shebang outright—bringing the total vest cost to $800—or, as Plummer put it in the video, opt for the “subscription-based model” of $12 per month or $120 per year.
«
I guess it figures out when you start your ride whether you’re paid up or not. Not clear whether it tells you, though.
Quite a method for extracting money from people. The always-connected, always-paying economy.
unique link to this extract
Timepieces that tell you how you are • Gazettabyte
Roy Rubenstein with some info about how the glucose monitoring from Rockley Photonics (mentioned yesterday) might work:
»
The technique underpinning smartwatch monitoring has the long title of non-invasive diffuse reflective spectroscopy.
Light at different wavelengths penetrates the skin and is scattered by blood vessels and cells and the interstitial fluid in between. The reflected light is analysed using spectroscopy to glean medical insights.
The smartwatch uses a green LED since blood haemoglobin has a good light absorption at that wavelength. “Effectively, what is being measured is the expansion and contraction of the blood vessels,” says [Rockley Photonics CEO Andrew] Rickman. “It is measuring the amount of light that is absorbed by the change of the volume of blood.”
It doesn’t stop there. Using a red LED and extending it into the infrared range, the blood oxygenation level is measured using the ratio of oxygenated (bright red) and unoxygenated (darker red) haemoglobin. “The ratio of the two wavelengths that you get back is proportional to the blood oxygen level,” says Rickman.
The visible range can also detect bilirubin, a yellow-orange bile pigment associated with jaundice. “But that is pretty much it,” says Rickman. “All the other thousands of constituents, if they have absorption peaks, are swamped in the visual range by haemoglobin.”
What Rockley has done is extend the light’s spectral to measure absorption peaks that otherwise are dwarfed by water and haemoglobin. “We are addressing the visible range and extending it into the infrared range, getting much more accuracy using laser technology compared to LEDs which opens up a whole range of things,” says Rickman.
To do this, Rockley has used its silicon photonics expertise to shrink a benchtop spectrometer to the size of a chip.
«
Related: magistrates can now get people to wear “alcohol tags” in sentencing for offenders whose crimes were “influenced by alcohol”. (Thanks Adewale Adetugbo for the Rockley link, Joel D for the tagging link.)
unique link to this extract
Taiwan drought may worsen global component shortage • Counterpoint Research
Brady Wang:
»
A stable and quality source of water is essential for semiconductor production. However, Taiwan is currently suffering from its worst drought in 56 years due to less than usual rainfall during the past year. The main sources of water in Taiwan are (1) the plum rains that occur in spring and summer when hot and cold air meet, (2) the heavy rainfall from typhoons in summer, and (3) the light rainfall in the mountains from the northeast monsoon in fall and winter. The proportions here are about 12%, 39% and 6% respectively.
Taiwan usually receives 7-9 typhoons every year. However, only one typhoon landed in Taiwan in 2020. To make matters worse, last winter and spring’s rainfall was heavily deficient, causing a shortage of water in Taiwan. The country is topographically divided by the 3,000-metre-high Central Mountain Range, which separates Taiwan’s eastern and western parts. The rains brought by the northeast monsoon in autumn and winter are mostly concentrated in the eastern and northern catchment areas, which means abundant rainfall for Draco, though it is of limited help to the Central and Tainan science parks. Therefore, water shortage becomes a serious problem for Taiwan’s technology industry in 2021. It may also have a serious impact on the global supply chain.
The Taiwanese government has taken many measures to address the water shortage problem, including transferring water between reservoirs, stopping water supply for agriculture, reducing water supply for households, drilling groundwater wells, and desalinating seawater. Besides, industrial users, including semiconductor manufacturers, have been asked to reduce their water consumption. TSMC, for example, has significantly increased the water recycling rate. The water level in the northern reservoirs has reached a multi-year low, though still sufficient for the continued use by Hsinchu Science Park (HSP).
However, the average effective water storage of the reservoirs supplying the Central Taiwan Science Park (CTSP) and Southern Taiwan Science Park (STSP) on April 30 was only 8.9% and 14.3% respectively (Exhibit 2). According to Counterpoint estimates, if there is no heavy rainfall or the rainfall does not fall in the catchment area, CTSP will face a water outage in July and STSP around August.
«
Apple is holding the web back with ‘uniquely underpowered’ iOS browsers, says Google engineer • WCCFTech
Furqan Shahid:
»
In a blog post, [Google Chrome engineer] Alex [Russel] talks about how the WebKit and iOS browsers are “Uniquely Underpowered” compared to the other modern browsers. He claims that Apple “consistently” delays new features for its browsers that “hold the key to unlocking whole categories of experiences on the web.”
»
Apple’s iOS browser (Safari) and engine (WebKit) are uniquely under-powered. Consistent delays in the delivery of important features ensure the web can never be a credible alternative to its proprietary tools and App Store.
«
Alex has cited an example of this by mentioning Stadia along with other cloud gaming services. Apple did not allow those services to be available on the App Store and pushed them to use the web instead, which required Apple to allow gamepad APIs so controllers can be used with these new web apps. That is a function that other browsers have offered for a long time except on iOS. But Apple still held back:
»
Suppose Apple had implemented WebRTC and the Gamepad API in a timely way. Who can say if the game streaming revolution now taking place might have happened sooner? It’s possible that Amazon Luna, NVIDIA GeForce NOW, Google Stadia, and Microsoft xCloud could have been built years earlier.
It’s also possible that APIs delivered on every other platform, but not yet available on any iOS browser (because Apple), may hold the key to unlocking whole categories of experiences on the web.
«
«
Russel’s post is quite complicated, and does accept that there’s little to choose between the browsers that have any significant share. He also allows that Chrome lacks some of the things that Safari has.
unique link to this extract
Don’t buy into Facebook’s ad-tracking pressure on iOS 14.5 • WIRED
Brian Barrett on why you can ignore Facebook’s weepy popups suggesting that letting it track you keeps the site “free of charge”:
»
“There are some types of ads, mostly retargeting, that will be harder to display, since now Facebook wouldn’t know who visited an app, put an item in the shopping cart, etc.,” says Ron Berman, a marketing professor at University of Pennsylvania’s Wharton School. He notes that Facebook will also have a harder time demonstrating that product sales were tied to specific ads, given the limitations on what information can now flow across sites and apps.
But you need not look much further than Facebook’s most recent quarterly earnings report, released last week, to see that iOS 14.5 seems unlikely to push the company toward any kind of precipice. The company took in over $26bn of revenue in the first three months of 2021, and its net income of $9.5bn nearly doubled that of the same period a year ago. It has over $64bn of cash and equivalents on hand. It’s doing just fine. Even if every single iOS 14.5 user opts out of tracking, Facebook will still have Android devices aplenty from which to squeeze profits.
It’s also not as if tracking prevention makes ads go away entirely. It arguably makes them less relevant. People may not click on them as often, which makes them less valuable, and outside analysts have predicted that Apple’s new policy will show up in Facebook’s bottom line. “We’ve seen estimates ranging from about a 2% to a 7% impairment of Facebook’s ad revenues this year and that range seems plausible to us, especially at the low end,” says Nicole Perrin, a principal analyst at eMarketer.
However, she adds, the company is expected to increase its ad revenue overall despite App Tracking Transparency. As WIRED’s Gilad Edelman has noted before, when third-party data disappears, companies that hold more first-party data have an edge. That’s Google, and that’s Facebook.
«
Trump launches new communications platform months after Twitter, Facebook ban • Fox News
Brooke Singman:
»
Former President Trump on Tuesday launched a communications platform, which will eventually give him the ability to communicate directly with his followers, after months of being banned from sites like Twitter and Facebook.
The platform, “From the Desk of Donald J. Trump” appears on http://www.DonaldJTrump.com/desk.
«
2003 called, and would like to point out that the “new communications platform” is known as a “blog”. (Meanwhile, at 1030 EST/ 1530 BST, the Facebook Oversight Board will announce its decision on whether Trump should be allowed back on Facebook. The broad expectation I’m seeing is that the FOB will say he should be. Divisive and polarising, algorithmically fuelled: social warming in action.)
unique link to this extract
Teens, tech and mental health: Oxford study finds no link – BBC News
Zoe Kleinman:
»
There remains “little association” between technology use and mental-health problems, a study of more than 430,000 10 to 15-year-olds suggests.
The Oxford Internet Institute compared TV viewing, social-media and device use with feelings of depression, suicidal tendencies and behavioural problems. It found a small drop in association between depression and social-media use and TV viewing, from 1991 to 2019. There was a small rise in that between emotional issues and social-media use.
“We couldn’t tell the difference between social-media impact and mental health in 2010 and 2019,” study co-author Prof Andrew Przybylski. said. “We’re not saying that fewer happy people use more social media.
“We’re saying that the connection is not getting stronger.”And this was a warning to regulators and lawmakers focusing on commonly held beliefs about the harmful effects of technology on young people’s mental health. Participants, in the US and UK, graded their own feelings using set questions with sliding scale responses. And they were asked about the duration of social-media or device activity but not more specifically how they had spent that time.
The paper is published in the journal Clinical Psychological Science.
«
I read a lot of these sorts of papers in preparing my book, and they’re very contradictory. I also spoke to Przybylski, who is generally dubious about studies that have suggested these links – there have been quite a few, and some books, strongly pushing the idea. One common problem these studies run up against is that kids use different devices: boys usually play video games (which makes them happy) while girls use social networks (and don’t seem to be happier).
unique link to this extract
The loneliness of the modern office team member • Financial Times
Pilita Clark:
»
Every other week or so, a number emerges somewhere in the world that I find both understandable and troubling.
It is the percentage of people who consistently say they don’t want to go back to working full-time in the office. Nearly 60% of British workers said this was how they felt back in September last year and also in March this year, even though more than a third of the UK population had had at least one Covid jab by then.
In the US, the share of workers who would prefer to keep working remotely as much as possible went from 35% in September to 44% in January. More recent European research found 97% of people who have been at home would prefer to stay there for at least part of the week once their offices reopen.
Since I am one of the millions thrilled to be liberated from a rushed commute and the tedium of presenteeism, these findings seem utterly rational. But they are also worrying because there is a gloomier reason that even well-paid, valued people in lofty jobs may be in no rush to go back to the office: long before the outbreak, they were lonely.
Their relationships with people in the office felt shallow. Worse, their sense of isolation may have had less to do with their personal lives than the way their work in teams was organised.
«
The heavy implication of course being that you’re doing a job that can be done from home. What about delivery drivers? Warehouse workers? People who answer telephones on switchboards? Perhaps I haven’t looked, but I’d like to know what proportion of jobs can and cannot be done remotely. It seems relevant.
unique link to this extract
Errata, corrigenda and ai no corrida: none notified
The Overspill: when there's more that I want to say 