Under a new rule, the Associated Press won’t name people arrested for minor offences. You may be able to think of one. CC-licensed photo by Andrew Feinberg 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 10 links for you. Face the camera. I’m @charlesarthur on Twitter. Observations and links welcome.
Obligatory book promo for
Social Warming, my forthcoming book. Out June 24. (Premature bookjaculations still welcomed.)
Apple pre-installed apps would be banned under antitrust package • Bloomberg
Rebecca Kern:
»
Apple Inc. would be prohibited from pre-installing its own apps on Apple devices under antitrust reform legislation introduced last week, said Democratic Representative David Cicilline, who is leading a push to pass new regulations for U.S. technology companies.
Cicilline told reporters Wednesday that a proposal prohibiting tech platforms from giving an advantage to their own products over those of competitors would mean Apple can’t ship devices with pre-installed apps on its iOS operating platform.
“It would be equally easy to download the other five apps as the Apple one so they’re not using their market dominance to favor their own products and services,” the Rhode Island Democrat said.
The proposal is part of a package of bipartisan bills that would impose significant new constraints on how tech companies operate, restricting acquisitions and forcing them to exit some businesses. The House Judiciary Committee will mark up the five bills in a hearing next week, Representative Jerrold Nadler of New York, the committee’s chairman, said.
Cicilline said the self-preferencing prohibition would also apply to Amazon.com Inc.’s Prime subscription service because it disadvantages some sellers who rely on the e-commerce platform.
When asked whether Microsoft Corp., which was subject to an epic antitrust case in the 1990s, would be subject to the measures, Cicilline said it would be up to the Justice Department and the Federal Trade Commission to make that determination.
«
Totally bonkers. Here’s what would actually happen: you’d go to buy a phone in a carrier store (as many people do) and they’d preload their own App Store and a ton of intrusive tracking junk apps.
Is there any evidence that people who are not developers are dissatisfied with how Google and Apple set things up right now? If so, I haven’t seen it.
unique link to this extract
Facebook to begin testing ads inside Oculus virtual reality headsets • CNBC
Salvador Rodriguez:
»
Facebook on Wednesday announced that it will begin testing advertisements that will appear within the company’s Oculus virtual reality headsets.
In May, the company said that it would begin running ads within the Oculus mobile app, but the announcement on Wednesday is the first time the social media company says it will show ads within its VR headsets.
The Oculus headset ads will first appear in the shooter game Blaston from Resolution Games. Ads will also begin appearing in two other Oculus apps over the coming weeks, Facebook said.
Oculus headset ads could be a significant step for Facebook, which derives more than 97% of its overall revenue from advertisements. Currently, those ads are primarily shown to users within the company’s Facebook and Instagram social networks.
Facebook also said these ads could provide new ways for software developers to generate revenue.
The ads will follow Facebook’s advertising principles and give users the same controls they have on Facebook. This includes the ability to hide specific ads or hide those from specific advertisers. Users can also select “Why am I seeing this ad?” to access more information about the ads they are shown.
«
Why are you seeing this ad? Because, Smith, if you want a picture of the future, imagine an ad being shown on two screens inches from your eyes – forever. Definitely, as Aaron Levie of Box noted wryly, “what VR was missing for mass adoption.”
unique link to this extract
Covid doom predictions that never happened • Noahpinion
Noah Smith:
»
Unfortunately, many of the bad predictions about COVID-19 came true. The people who saw cases ramping up exponentially, and warned that this was going to be a mass death event, were right, while the people who minimized the threat and waved it away were wrong. And a lot of people are dead because we didn’t listen to the former.
But economic predictions are a different story. When unemployment spiked to Great Depression levels in the early days of lockdown, it seemed to me — and to many, many others — like this downturn was destined to turn into a decade of mass economic hardship. Fortunately, that was completely off the mark! I got it very wrong and Paul Krugman got it right — with no financial crisis and no big overhang of debt, the economy simply wasn’t destined for a repeat of 2008-12. Though the recovery has proven bumpy thus far, but most economists still forecast a relatively swift return to the pre-pandemic growth trend.
«
Pretty much everything that was feared economically didn’t happen – except for (only US?) universities, where roughly 1 in 8 jobs vanished and enrollments for next year are down. Smith admits that Paul Krugman called it correctly, perceiving that this wasn’t a financial crisis, and there was no debt overhang, so things could snap back quickly.
unique link to this extract
AP says it will no longer name suspects in minor crimes • Associated Press
David Bauder:
»
The Associated Press said Tuesday it will no longer run the names of people charged with minor crimes, out of concern that such stories can have a long, damaging afterlife on the internet that can make it hard for individuals to move on with their lives.
In so doing, one of the world’s biggest newsgathering organizations has waded into a debate over an issue that wasn’t of much concern before the rise of search engines, when finding information on people often required going through yellowed newspaper clippings.
Often, the AP will publish a minor story — say, about a person arrested for stripping naked and dancing drunkenly atop a bar — that will hold some brief interest regionally or even nationally and be forgotten the next day.
But the name of the person arrested will live on forever online, even if the charges are dropped or the person is acquitted, said John Daniszewski, AP’s vice president for standards. And that can hurt someone’s ability to get a job, join a club or run for office years later.
The AP, in a directive sent out to its journalists across the country, said it will no longer name suspects or transmit photographs of them in brief stories about minor crimes when there is little chance the organization will cover the case beyond the initial arrest.
«
A form of the “right to be forgotten”, which has been implemented in Europe since May 2014. Good to see that the AP has finally realised how this “internet” thing works, though.
unique link to this extract
Creating a trustworthy reviews experience • About Amazon
Amazon says it blocked more than 200 million fake reviews before they even got onto its sites, but:
»
Due to our continued improvements in detection of fake reviews and connections between bad-actor buying and selling accounts, we have seen an increasing trend of bad actors attempting to solicit fake reviews outside Amazon, particularly via social media services. Some use social media services on their own; in other cases, they hire a third-party service provider to perpetrate this activity on their behalf. However, bad actors regularly try to take this transaction outside Amazon to obscure our ability to detect their activity and the relationship between the multiple accounts committing or benefiting from this abuse. As a result, we use a number of techniques, including advanced machine learning, to try to detect groups of connected entities—customer accounts, selling accounts, products, brands, and more. However, it’s also clear that this is an industry-wide battle, and we need to work together to make faster progress.
When we detect fake reviews that may have been perpetrated outside Amazon, we regularly report the activity to the social media company where it occurred. In the first three months of 2020, we reported more than 300 groups to social media companies, who then took a median time of 45 days to shut down those groups from using their service to perpetrate abuse. In the first three months of 2021, we reported more than 1,000 such groups, with social media services taking a median time of five days to take them down. While we appreciate that some social media companies have become much faster at responding, to address this problem at scale, it is imperative for social media companies to invest adequately in proactive controls to detect and enforce fake reviews ahead of our reporting the issue to them.
«
Basically blaming Facebook, without naming Facebook. Part of the big problem is that the “fake” reviews often come from people who have effectively been bribed to give a good review. (Possibly the motive behind the removal of a few brands from Amazon recently.)
unique link to this extract
The COVID-19 lab leak theory is short on evidence and long on guesswork • Foreign Policy
Justin Ling:
»
The lab leak theory says the furin cleavage site, a tiny enzyme dangling from the virus, is key to understanding the novel coronavirus’s origin.
Goldstein agrees. But, he said, that cleavage site actually points toward the virus’s natural origin.
“You cannot, in a normal cell culture, maintain the furin cleavage site,” he told me. When the COVID-19 virus is replicated in a cell culture in a lab, he said, the furin cleavage tends to delete itself. A peer-reviewed paper, published in late April in Nature, noted that habit and identified seven other papers that found a similar deletion.
So if researchers were using traditional methods and their preferred cell lines to try to force the virus to replicate, mutate, and change, the furin cleavage site would likely disappear.
The gain-of-function proponents say this furin site is too well adapted for humans to be an accident. But Goldstein said the opposite is true. The cleavage site is imperfect, so odd, that it could have only been a freak of nature. “No virologist would use that cleavage site,” he said.
It is possible to replicate the virus in a lab while preserving the cleavage site, Goldstein added, but it would “require doing things differently than everyone does them.” And, crucially, it would require them choosing cell cultures that replicate the virus more slowly.
So the researchers would have had to make a series of inefficient and strange decisions to preserve a tiny, novel, odd enzyme. Indeed, the researchers at Imperial College London behind the April Nature article found that the addition of four amino acids in the virus’s spike protein “occurred during its emergence from an animal reservoir and created a suboptimal furin [cleavage site].” Another study published in January in Stem Cell Research demonstrated how these furin sites naturally evolve in many coronaviruses.
«
Sure that we’re going to see the Daily Mail and NY Post and WSJ write up that research real soon now. (Thanks G for the link.) We all became epidemiologists, and now we’re all becoming virology genomic experts.
unique link to this extract
Segmentation faults: how machine learning trains us to appear insane to one another • doxa
Jon Stokes:
»
Imagine a world where the following things are true:
1. Big Tech platforms make money by micro-targeting ads to their users, so that ads that are more accurately and narrowly tailored are more valuable to platforms and advertisers than ads that are more general.
2. Following on the above, the more fine-grained segments you can slice your audience into, the better you can service the long tail of advertisers. So there’s an ad-driven market for audience segmentation that the platforms want to meet.
3. Advertising works, and in general, a human’s behaviors, preferences, values, and even basic tenets of their worldview can be modified by the media they’re exposed to.
4. Our society is getting more fragmented and polarized, as existing groups splinter apart online and new, often smaller groups and clusters form around different ideas, claims, worldviews, and identity characteristics.I understand there are legitimate objections to each point above, but just go with it for a moment. Imagine that the above accurately describes our present social reality in 2021.
If I’m right, then it seems quite possible that the first items in the list above — i.e., tech platforms have created a lucrative advertising market for an atomized, segmented audience — is leading directly to the last item in the list — i.e, to our increasingly atomized, segmented public.
«
In the “great minds thinking alike” category, I’d offer this as Jon coming up with the same realisation about how (some elements of) social warming emerge as me. Though he chose a different name for the phenomenon.
unique link to this extract
Spotify’s Clubhouse competitor Greenroom is here • The Verge
Ashley Carman:
»
Spotify’s live audio app, Greenroom, formally launched on Wednesday on iOS and Android, marking the company’s first real attempt at creating a social media platform. The social audio app, which is similar to Clubhouse, allows users to host live conversations about sports, music, and culture.
Today’s launch doesn’t come with a marquee creator announcement or specific event planned, but instead, the company is taking the opportunity to encourage people to sign up and figure out how they’ll want to use the app. Some of its core functionality, a person close to the situation says, will eventually make its way to the actual Spotify app, so the team will monitor what happens in Greenroom closely.
The app is built on Locker Room, which Betty Labs created and Spotify acquired in March. That app focused solely on sports content, so users who have been logged on since the start will have to get used to seeing more than just sports talk, which is likely the biggest change. Other noticeable changes to the app are mostly visual. It now has a Spotify green-and-black color scheme, as well as a new logo and font. Functionality-wise, it also now features native recording, which will allow users to save their shows and distribute them as podcasts. (Of course, Spotify owns Anchor, so one could easily imagine shows eventually being natively moved to the creation software for further editing and publishing.)
«
This seems like a better idea than Clubhouse, inasmuch as it’s about sports – and people like talking about sports, which also gives it a structure by default, based on the time of events.
Anyway, remember Clubhouse? If these Statista figures are reliable, it’s not quite dead. Interest is biggest in Asia, while in the US it’s totally fizzled.
unique link to this extract
Cryptocurrency miners bought 700,000 GPUs in Q1 2021 • ExtremeTech
Joel Hruska:
»
According to JPR, component prices for GPUs have increased by as much as 70%. I’m perfectly willing to believe that board manufacturers have boosted their own prices to compensate for that, but the current GPU market remains ludicrously inflated. Component prices may be up 70%, but GPU prices have been up 2.5x – 4x.
The data below does not include the notebook market. “DT PC w/o WS” means “Desktop PC without workstations,” meaning Quadro and Radeon Pro shipments are not included here.
Attach rates — the percentage of PCs that ship with a discrete graphics card — have trended downwards in the desktop market, with a very clear bump towards the end of 2020. Note, however, that even as the graphics attach rate has dropped once more, the total number of AIB (Add-In Boards) being sold has skyrocketed. In other words: we’re not seeing a huge demand spike because OEMs like Dell, HP, and Lenovo are suddenly selling lots of systems with high-end graphics cards. We’re seeing a huge spike in AIB shipments because of cryptocurrency mining. The last time AIB shipments sparked like this was 2017-2018. That time period corresponds to the second cryptocurrency bubble.
JPR estimates that miners bought 700,000 midrange and high-end GPUs in Q1 2021, accounting for about 25% of the AIB market and roughly half a billion dollars in cash. None of that revenue flows to Nvidia or AMD; it’s all being captured by the channel partners. JPR notes that electricity prices in Mongolia are around 4.5 cents per kW, explaining how these farms remain profitable (and why it’s impossible to compete with them if you have to pay residential rates for electricity).
«
Compared to the PC market, of around 70m units per quarter, that… doesn’t feel that big?
unique link to this extract
Now the UK is investigating the Apple and Google duopoly • Android Authority
Hadlee Simons:
»
The UK’s Competition and Markets Authority (CMA) has announced that it’s investigating both Apple and Google over its smartphone operating systems, app stores, and web browsers (h/t: Android Central). More specifically, the authority says it’s checking whether the two companies are stifling competition across a variety of digital markets.
“The CMA is concerned this could lead to reduced innovation across the sector and consumers paying higher prices for devices and apps, or for other goods and services due to higher advertising prices,” read an excerpt of the announcement. This investigation will also look at the power Apple and Google have over businesses like app developers.
This announcement doesn’t name any specific third-party apps or companies, but the likes of Spotify, Tinder owner Match, and Tile have all criticized Apple and Google’s business practices in recent months. These criticisms center on the app store holder’s cut of sales, changes to app store rules, and competing products by platform holders.
Google has already been slapped on the wrist over its Android-related practices. The firm was previously slapped with a $5bn fine by the EU in 2018 for requiring OEMs to bundle specific apps, incentivizing the use of its products, and preventing OEMs from running Android forks. Meanwhile, Apple is currently embroiled in a legal battle with Epic that could have major ramifications for its App Store and iPhone business practices.
«
Views from anyone welcome to July 26, and the CMA would particularly like to hear from app developers. CMA is giving itself 12 months to conclude the study.
unique link to this extract
Errata, corrigenda and ai no corrida: none notified
The Overspill: when there's more that I want to say 