A significant proportion of men are red/green colourblind – but modern web design makes little allowance for them. CC-licensed photo by Jam Willem Doormembal 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.

---

There was another post last Friday at the Social Warming Substack: TikTok, Instagram and the lessons never learnt. Free signup.

---

A selection of 10 links for you. It says what? I’m @charlesarthur on Twitter. On Mastodon: https://newsie.social/@charlesarthur. Observations and links welcome.

---

ChatGPT leaking Samsung chip secrets is the tip of the iceberg • EENews Europe

Peter Clarke:

»

Apparently engineers and other workers at many companies are recruiting ChatGPT to work for them, to write software and prepare reports for example. This is sometimes with, and sometimes without, their employers’ approval.

The Digitimes report mentions three specific cases of leaks caused by engineers sharing information with ChatGPT. In one case an engineer uploaded faulty code and asked ChatGPT to find the fault and optimize the software. But as a result the source code became part of ChatGPT’s database and learning materials.

Another case was where ChatGPT was asked to take the minutes of meeting. By default the discussion and exactly who attended the meeting – both confidential – were stored on the ChatGPT database and thus ChatGPT was able to divulge the material to anyone who asked.

As a result of such events, Samsung, SK Hynix, LG and many other companies are scrambling to either ban or draw up guidelines for the use of ChatGPT and other AI chatbot services in the workplace, according to The Korea Times.

The Korea Times seemed to confirm Samsung’s mishaps and said that a message on an in-company bulletin had been posted calling attention to the misuse of ChatGPT. SK Hynix has blocked the use of ChatGPT on its internal computer network and employees must obtain security approval before using ChatGPT, the newspaper said.

The newspaper also quoted Kim Dae-jong, a professor of business administration at Sejong University, saying that the use of ChatGPT in the workplace was spreading.

«

The original report about Samsung workers leaking secrets to ChatGPT (which might want to know them) is said to have appeared in Digitimes. I’ve drawn a blank.
unique link to this extract

---

Samsung forecasts a shocking 96% drop in profits for Q1 2023 • Ars Technica

Ron Amadeo:

»

Samsung’s next quarter is shaping up to be even worse than Samsung’s last quarter, which was already at an eight-year low. The company warned investors today that it’s a shocking 95.8% year-over-year drop in operating profit for Q1 2023. If that expectation holds, this will be the company’s worst quarter since 2009, which dates back to the company’s pre-smartphone era.

Samsung doesn’t have much explanation for the drop other than a weakening economy and lowered demand for chips. Preliminary results have the company making only 600bn won ($450m) in profit for Q1 2023, compared to 14.12trn won in profit ($10.7bn) for Q1 2022.

While phones and TVs are probably Samsung’s biggest consumer-facing products, the company’s nigh-invisible component business makes up most of Samsung’s profits. Components like RAM and NAND storage chips don’t just ship in Samsung products, but also land in most other phones, laptops, desktops, TVs, and other electronics from Samsung’s competitors. A DigiTimes breakdown of Samsung’s business for 2022 has the memory division at 55% of profits, mobile at 22%, and displays at 11%, so Samsung’s profits mostly go up and down with the memory business.

«

Lower chip demand means worse economies of scale. But the worst since 2009? That’s going back a long way. But you can expect that it’s going to come back. Samsung is, in its way, the most determined animal in the technology world.
unique link to this extract

---

DPReview closure: an update • Digital Photography Review

Scott Everett is general manager of DPReview:

»

Dear readers,

We’ve received a lot of questions about what’s next for the site. We hear your concerns about losing the content that has been carefully curated over the years, and want to assure you that the content will remain available as an archive.

We’ve also heard that you need more time to access the site, so we’re going to keep publishing some more stories while we work on archiving.

Thank you to this community and the support you’ve shown us over the years.

«

Hard to know if this is happening because of the outcry, or was planned all along. But you’d think if it had been planned earlier, they would have said it along with the news of the closure.
unique link to this extract

---

Introducing Substack Notes • Substack

Hamish McKenzie:

»

While Notes may look like familiar social media feeds, the key difference is in what you don’t see. The Substack network runs on paid subscriptions, not ads. This changes everything. 

The lifeblood of an ad-based social media feed is attention. In legacy social networks, people get rewarded for creating content that goes viral within the context of the feed, regardless of whether or not people value it, locking readers in a perpetual scroll. Almost all the attendant financial rewards then go to the owner of the platform. 

By contrast, the lifeblood of a subscription network is the money paid to people who are doing worthy work within it. Here, people get rewarded for respecting the trust and attention of their audiences. The ultimate goal on this platform is to convert casual readers into paying subscribers. In this system, the vast majority of the financial rewards go to the creators of the content.

As we develop Notes, we will focus on building a system that lets people control the contours and boundaries of their subscription universe so that it is easy to keep trolls out and even easier to let valuable contributors in. The goal here is not to create a perfectly sanitized information environment, but to set the conditions for constructive discussion where there is enough common ground to seek understanding while holding onto the worthwhile tension needed for great art and new ideas. It won’t feel like the social media we know today.

Many of us have grown so used to talk of hellsites and doomscrolling—while wondering if social media is driving us mad—that we have forgotten that the internet can be good.

«

Which is why Elon Musk has thrown all pretence of “free speech” out of the window (not that he hasn’t many times already) and first blocked Substack links from being posted, and then saw them labelled as “malware”. Perhaps that will have changed by now. Who knows.
unique link to this extract

---

Time set for national mobile phone emergency alert test • BBC News

Brian Wheeler:

»

A siren will go off on nearly every smartphone in the UK on Sunday 23 April, the government has announced.

The 10 seconds of sound and vibration at 15:00 BST will test a new emergency alerts system.
The test had originally been planned for the early evening but was moved to avoid clashing with an FA Cup semi-final, which kicks off at 16:30.

The government was also keen to avoid a clash with the London Marathon, which starts at 09:30 on that Sunday.

The alert system will be used to warn of extreme weather events, such as flash floods or wildfires. It could also be used during terror incidents or civil defence emergencies if the UK was under attack.
The minister in charge of the system, Oliver Dowden, said it would be used only in situations where there was an immediate risk to life.

«

Such a diary clash! Marathon and the FA Cup final. When is there time to momentarily terrify all the population?
unique link to this extract

---

Who has contributed most to global warming? • Sustainability By Numbers

Hannah Ritchie:

»

we can look at how much warming each country has contributed to date. This is the sum of warming caused by emissions of all three gases, and from all sources.

You can see these contributions in the map – or use our interactive chart to explore the data in more detail. If you’re screaming that this should be in per capita terms, I hear you and address this at the end.

The rankings of countries probably won’t surprise you. Countries with large populations such as China and India are among the top contributors. The United States and the European Union, with long histories of fossil fuel burning and rich lifestyles, are also up there.

The US has caused 0.28°C of warming, followed by China at 0.2°C, and the EU at 0.17°C.

As a share of total warming, that puts the US at 17%, China at 12%, and the EU at 10%.

A final note: if you’re looking at the warming caused by your country and thinking “that’s so tiny, why do we even bother?” you might want to read my recent article on why ‘negligible emitters’ really do matter.

«

The per capita question is, as she points out, complicated because how do you measure population, which keeps changing? Today’s number isn’t the same as the number when the emissions were happening.
unique link to this extract

---

Supercentenarian and remarkable age records exhibit patterns indicative of clerical errors and pension fraud • bioRxiv

Saul Justin Newman:

»

The observation of individuals attaining remarkable ages, and their concentration into geographic sub-regions or ‘blue zones’, has generated considerable scientific interest. Proposed drivers of remarkable longevity include high vegetable intake, strong social connections, and genetic markers.

Here, we reveal new predictors of remarkable longevity and ‘supercentenarian’ status. In the United States supercentenarian status is predicted by the absence of vital registration. In the UK, Italy, Japan, and France remarkable longevity is instead predicted by regional poverty, old-age poverty, material deprivation, low incomes, high crime rates, a remote region of birth, worse health, and fewer 90+ year old people.

In addition, supercentenarian birthdates are concentrated on the first of the month and days divisible by five: patterns indicative of widespread fraud and error.

«

In other words, those people over 100? Often just faked. The full PDF is fun towards the end (para 745 onward) when it poses the question of how you’d determine how old someone is when they’re really old.
unique link to this extract

---

This is what it looks like to be colourblind • The Verge

Andy Baio:

»

About 8% of men, roughly 1 in 12, have some form of colour vision deficiency. (It’s hereditary, so figures will vary from region to region.) My mom’s colour vision is even worse than mine, which is very unusual: only about 0.5% of women globally are colourblind, about 1 in 200.

I’ve had a lot of conversations about my colourblindness with people who aren’t colorblind. (Pro tip: when you meet a colourblind person, don’t repeatedly point to things and ask what colour they are.) It seems like the very idea of colourblindness is hard for them to visualize. 

Despite what many think, I can see most colours! My world isn’t a black-and-white movie. Achromatopsia, or total colourblindness, is much more rare, affecting about 1 in 30,000 people. (Unless you were born on the Pingelap atoll in the South Pacific, where 10% of the population have inherited the gene.) 

Ninety-nine% of colourblind people, like me, have a form of red-green colourblindness. I was born with the most common type, deuteranopia, a genetic mutation that affects the ability of the green-sensitive cones in my eyes to absorb light.

As a result, some hues of green and red look like each other, converging on a muddy brown. Other colours, like shades of purple and blue, bright orange and green, or even pink and grey, can look very similar. People with other kinds of colourblindness will confuse different colours.

For example, at a glance, barring other context clues like texture and toppings, avocado toast and peanut butter toast look pretty much the same to me.

Apparently, this is nauseating to people? That’s my life.

Because red and green are complementary colours opposite one another on the colour wheel, they’ve become the default colours for every designer who wants to represent opposites: true and false, high and low, stop and go.

Inconveniently, these are also the two colours most likely to be mixed up by people with colour vision deficiencies.

I wish every designer in the world understood this and would switch to, say, red and blue for opposing colours. But I know that won’t happen: the cultural meaning is too ingrained.

«

Although in the UK the danger of colourblind (male) electricians was realised after WW2, and mains plug wiring changed from red (live)/ black (neutral)/ green (earth) to brown (L)/ blue (N)/ yellow green (E). (First applied 1969, yet not universal until 2004.)It can be done.
unique link to this extract

---

Hubble sees possible runaway black hole creating a trail of stars • Nasa Hubble Site

»

The universe is so capricious that even the slightest things that might go unnoticed could have profound implications. That’s what happened to Yale astronomer Pieter van Dokkum when he was looking through Hubble Space Telescope images and noticed a suspected blemish that looked like a scratch on photographic film. For Hubble’s electronic cameras, cosmic rays skimming along the detector look like “scratches.”

But once spectroscopy was done on the oddball streak van Dokkum realized it was really a 200,000-light-year-long chain of young blue stars located over halfway across the universe. Van Dokkum and his colleagues believe that it stretches between a runaway monster black hole and the galaxy from which it was ejected.

The black hole must be compressing gas along its wake, which condenses to form stars. Nothing like it has ever been seen anywhere else in the universe before.

«

You can read the preprint paper. Plus there isn’t just one of these: there’s another one heading in the exact opposite direction, doing just the same.
unique link to this extract

---

Google will drop Dropcam and Nest Secure in 2024 • The Verge

Nathan Edwards:

»

Google is ending support for the Dropcam and the Nest Secure home security system in one year, on April 8th, 2024. They are among the few remaining Nest products that haven’t been brought over to Google Home, and their demise hints that the new Google Home app might almost be here. At least, no more than a year away. Surely.

Google is also winding down the last few legacy Works with Nest connections, but not ‘til September 29th.

Existing Dropcam cameras will keep working until April 8th, 2024, after which you won’t be able to access them from the Nest app. To soften the blow, Google’s offering a free indoor wired Nest Cam to Dropcam owners who subscribe to Nest Aware. Nonsubscribers will get a 50 percent-off coupon. The promotion runs until May 7, 2024, so you can keep using your Dropcam until it stops working.

The Dropcam (fka Dropcam HD) came out in 2012, and the Dropcam Pro in 2013. Then, Google bought Nest, and Nest bought Dropcam. In 2015, Google spun Nest out when it formed Alphabet, and for a while, Google and Nest were both making smart home products. Then, Google reabsorbed Nest in 2018, and there’s been a whole lot of messy business trying to integrate Nest products into the Google Home app — and killing off the ones that can’t be integrated.

Now that it’s dropping Dropcam and Nest Secure, the Nest Protect smart smoke alarms are the only Nest App-only devices left, and Google has promised to bring them to the new Google Home app. The updated app has been in public preview since October, and there’s still no firm date, but it must be getting close, right?

«

Ten years seems like a fair run – but it’s hardly as if the need for home security cameras has gone away. It’s all a bit messy: Nest, Dropcam, Google Home? Google’s hardware strategy is still in pieces.
unique link to this extract

---

• Why do social networks drive us a little mad? • Why does angry content seem to dominate what we see? • How much of a role do algorithms play in affecting what we see and do online? • What can we do about it? • Did Facebook have any inkling of what was coming in Myanmar in 2016? Read Social Warming, my latest book, and find answers – and more.

---

Errata, corrigenda and ai no corrida: none notified