Start up: Radiohead v Spotify, Facebook’s news problem, Google’s dark links, Sinclair’s calculator genius, and more

Excitement over web video views compared to TV is overdone, mainly because they measure very different things. Photo by x-ray delta one on Flickr.

You can now 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. Use them wisely. I’m charlesarthur on Twitter. Observations and links welcome.

Radiohead’s new album isn’t on Spotify. So what? Spotify doesn’t need it • The Guardian

Eamonn Forde brandishes data, damn him:


Burn the Witch has been played just short of 3m times on Spotify since its release last week. Pretty impressive, until you note that the service has more than 100 million users. Given it’s Radiohead, the people who played the track will have done so multiple times. So we can estimate that the people playing Burn the Witch make up less than 1% of Spotify’s total user base. Helpfully, Spotify also makes public other metrics. Radiohead have 4.7 million monthly listeners on Spotify, and are the 196th most played act. That means that 95.3% of Spotify’s users have literally no interest in Radiohead. A Moon Shaped Pool not being there could potentially irk 4.7 million people, but actually Radiohead do not appear on the radars of the vast majority of Spotify’s users.

A look at how the band’s catalogue has performed on Spotify – considering that they will have chalked up multiple plays from users – reveals this fact in stark terms. Cumulative plays for Pablo Honey run to 119.1m, with 103m of those streams being for Creep, a statistic bound to cheer up Thom Yorke. For The Bends, it’s 102.2m, for OK Computer it’s 159.1m, for Kid A it’s 61.5m, for Amnesiac it’s 34.7m, for Hail to the Thief it’s 49.3m and for The King of Limbs it’s 45.5m. Spotify users’ interest is clearly in the deep back catalogue of the band (or, to put it another way, in their most accessible music).


Yes, it is calling out for a graph. Here you go.

Radiohead album plays on Spotify, graphed

(Would still like a statement from Radiohead on why the album isn’t, but the singles are, on Spotify.)
link to this extract

Internet video views is a 100% bullshit metric • Gawker

Kevin Draper:


The TV ratings Nielsen reports aren’t concurrent viewers, but rather “average minute audience,” which is exactly what it sounds like. It measures the average audience watching across each minute of the show.

If BuzzFeed’s watermelon video had been measured the way a TV show is, its viewership would’ve been closer to zero than the 807,000 it trumpeted to advertisers. Viewership started off low and took 45 minutes to build to that 807,000, and few people watched the entire video; many tuned in for five or 10 minute blocks at the end. Facebook’s metrics also wildly inflate the number of people watching a given video, as they count somebody as a viewer once they have been watching for just three seconds, and by default Facebook videos autoplay as you scroll to them in your feed.

The conflation of digital and traditional viewership metrics has gotten under the skin of TV people, and for good reason. If advertisers can be hoodwinked into believing that a sizable number of people are actually watching things on Facebook Live, they will direct their money online, where the ad rates are much, much lower than they are on TV. The thing here is that the TV people are right—even serious online video hits deliver numbers that would barely register if measured the same way TV programming is.


In other news: “Youtube says its primetime audience is bigger than the top 10 TV shows combined”. Let’s see if you can work this one out unaided.
link to this extract

The real reason Apple made the Apple Watch • TIME

Tim Bajarin:


If you look at Apple’s current health initiatives, many are focused on helping people record data of all types and get it securely to their healthcare providers. Apple also has projects related to healthcare records, management and interaction between the doctor and patient with a goal of making the patient-doctor relationship more fruitful and less frustrating.

I have long been observing these key moves around healthcare, which accelerated after Jobs’ death. It seems clear that Apple’s management has now and will continue to have a major focus on bridging the gap between a person and their healthcare providers. I believe Apple is on a mission to improve the overall health of its customers as well as that of the healthcare system, a task Jobs gave them before he died. And while Apple’s products define Jobs’ legacy, it may turn out that his and Apple’s greatest contribution may be to bring greater order to the fragmented healthcare world.


The fragmented American healthcare system, I think you mean.
link to this extract

The real problem with Facebook and the news • Stratechery

Ben Thompson on yesterday’s claim by one (apparently disaffected right-leaning) curator for Facebook’s Trending News feed: This, then, is the deep irony of this controversy:


Facebook is receiving a huge amount of criticism for allegedly biasing the news via the empowerment of a team of human curators to make editorial decisions, as opposed to relying on what was previously thought to be an algorithm; it is an algorithm, though — the algorithm that powers the News Feed, with the goal of driving engagement — that is arguably doing more damage to our politics than the most biased human editor ever could. The fact of the matter is that, on the part of Facebook people actually see — the News Feed, not Trending News — conservatives see conservative stories, and liberals see liberal ones; the middle of the road is as hard to find as a viable business model for journalism (these things are not disconnected).

Indeed, one could make the argument that an authoritative news module from Facebook would actually be a civil benefit: at least we would all be starting from a common set of facts. What is far more damaging — and far more engaging, and thus lucrative for Facebook — is all of us in our own virtual neighborhoods of our own making, liking opinions that tell us we’re right instead of engaging with viewpoints that make us question our assumptions.


Which of course makes one wonder what an anti-News Feed, which intentionally dropped some stories about things you don’t agree with, would feel like. Jarring? Wrong? Annoying?
link to this extract

190 Android apps infected with malware discovered on the Google Play Store • Softpedia

Catalin Cimpanu:


Researchers spotted the malware-infected apps towards the end of April, but only recently have these apps been removed. The Russian security firm says the apps contained a version of the malware identified as Android.Click.95.

According to their analysis of the malware’s mode of operation, Android.Click waits for six hours after the user installs it as part of an infected app.

After the six hours pass, the malware forcibly loads a URL in the user’s browser, which contains scareware-like messages that tell the user his system or his battery has problems.

To fix his issues, the user has to download another app. In the cases they’ve observed, Dr.Web researchers say the malware redirected users back to the Google Play Store to download these second-stage apps.

“For each download, fraudsters receive interest under the terms of affiliate advertising agreements,” Dr.Web researchers explained. “It explains why Android.Click.95 is so much widespread—the cybercriminals try to make as much profit as they can from these downloads.”


To be precise, “Android.Click.95 opens the fraudulent website every 2 minutes from the moment when it has started functioning, making it irritating to use the infected device. At that, the maximum number of visits of the fraudulent website is limited to 1000 times.” Seems to be Russian apps (judging from the analysis) but likely to spread.
link to this extract

The odds are you won’t know when to quit • Tim Harford


My favourite study of loss aversion concerns players of the TV game show Deal or No Deal, in which players must periodically decide whether to keep gambling or accept an offer from the mysterious “Banker” to buy them out of the game. In one notorious Dutch episode, a contestant named Frank was offered €75,000 to stop; he kept playing and lost his next gamble. The Banker’s next offer was just €2,400, which was actually a fair offer. But at that point loss aversion kicked in. With the lost €75,000 in mind, Frank refused all further deals, kept gambling and kept losing. He eventually won just €10.

A study of Deal or No Deal by behavioural economists including Thierry Post and Richard Thaler found that while Frank’s fate was spectacular, his behaviour was statistically typical. People hate to quit if they feel they’re losing…

…I was struck by a recent FT article by equity analyst Daniel Davies describing how a portfolio based on expert research recommendations would tend to do badly, but if the same portfolio had a “stop-loss” rule that simply jettisoned stocks after a 10 per cent loss, it would tend to do very well. The stop-loss rule cancelled out the instinctive tendency to hold on stubbornly to losers. Yet Warren Buffett seems to do very well by buying and holding.


Harford then has three suggestions for how not to cling on to losses. But psychology is hard to beat.
link to this extract

Researchers say computer screens change how you think about what you read • The Washington Post

Andrea Peterson:


You probably spend a lot of time staring at screens – but all that computer time may be making you miss the big picture, new research has found.

Reading something on a screen – as opposed to a printout – causes people to home in on details and but not broader ideas, according to a new article by Geoff Kaufman. a professor at Carnegie Mellon, and Mary Flanagan, a professor at Dartmouth. 

“Digital screens almost seem to create a sort of tunnel vision where you’re focusing on just the information you’re getting this moment, not the broader context,” Kaufman said.

The article is based on a series of studies involving a total of more than 300 participants that were carried out while the two researchers worked together at Tiltfactor, a Dartmouth game design lab.


Reading speed is higher on paper too, or at least used to be recorded as such; not sure whether that’s still true in the era of invisible pixels.
link to this extract

Reversing Sinclair’s amazing 1974 calculator hack – half the ROM of the HP-35 • Righto

Ken Shirriff:


In a hotel room in Texas, Clive Sinclair had a big problem. He wanted to sell a cheap scientific calculator that would grab the market from expensive calculators such as the popular HP-35. Hewlett-Packard had taken two years, 20 engineers, and a million dollars to design the HP-35, which used 5 complex chips and sold for $395. Sinclair’s partnership with calculator manufacturer Bowmar had gone nowhere. Now Texas Instruments offered him an inexpensive calculator chip that could barely do four-function math. Could he use this chip to build a $100 scientific calculator?

Texas Instruments’ engineers said this was impossible – their chip only had 3 storage registers, no subroutine calls, and no storage for constants such as π. The ROM storage in the calculator held only 320 instructions, just enough for basic arithmetic. How could they possibly squeeze any scientific functions into this chip?
Fortunately Clive Sinclair, head of Sinclair Radionics, had a secret weapon – programming whiz and math PhD Nigel Searle. In a few days in Texas, they came up with new algorithms and wrote the code for the world’s first single-chip scientific calculator, somehow programming sine, cosine, tangent, arcsine, arccos, arctan, log, and exponentiation into the chip. The engineers at Texas Instruments were amazed.

How did they do it? Up until now it’s been a mystery. But through reverse engineering, I’ve determined the exact algorithms and implemented a simulator that runs the calculator’s actual code. The reverse-engineered code along with my detailed comments is in the window below.


Amazing. The ingenuity of the work in these early systems is inspiring. What’s the equivalent today?
link to this extract

HTC Q1 16A – Family silver • Radio Free Mobile

Richard Windsor:


▪ HTC’s war chest is diminishing fast and it is quickly selling the family silver in order to support the substantial cash burn.
▪ During Q1 16A HTC recognised cash gains of NT$2.1bn from selling fixed assets, NT$6.1bn from selling non-current financial investments and NT$1.4bn from selling current financial assets.
▪ This is how HTC managed to show an increase in cash and cash equivalents (compared to Q4 15A) of NT$3.7bn.
▪ These sales are very likely to be one time in nature meaning that, in reality, HTC is draining its reserves by NT$5.9bn every quarter.
▪ These asset sales have masked what was a dreadful 3 months for HTC and I am concerned that the Vive will not take off in the kind of volumes or soon enough to keep HTC from real trouble.
▪ Consequently, 2016 is likely to be dominated by the agonising decline of its handset business raising the high likelihood of further substantial cash outflows.
▪ Consequently, the most I would pay for HTC would be 1 times its cash balance as I view the Vive as a long shot to rescue the company and it is very likely that the cash balance will continue declining.


HTC’s accounts show a cash balance of NT$41.7bn at the end of Q1. That’s about eight quarters’ worth if it keeps draining cash at that rate.

Or, put another way: at the end of March 2015 HTC had NT$51.7bn of cash and cash equivalents. Ending March 2016, it had NT$39.0bn. That’s a fifth of its cash and equivalents gone in a year – and as Windsor points out, that’s only done by one-off asset sales. It’s burning the furniture to keep warm.
link to this extract

Google is testing a radical change by turning people’s search results black •

Cara McGoogan on what seems to be an A/B test where results links are black, not the web’s usual blue:


If the comments from users on Twitter, Reddit and Google’s forums are any indication, it is not likely that the changes will be rolled out across the board any time soon. 

Other than the red links it uses in China, and different shades of its blue Google has not used other colours for its search results before. 

Google is not the only major technology company to carry out A/B tests on its users. Netflix recently admitted that it tests six different images for many TV and movie titles, and rolls out the one that most viewers click on. 

Facebook has also conducted tests designed to emotionally manipulate users by highlighting positive and negative emotions, while OK Cupid has deliberately matched incompatible people to see the outcomes

There isn’t a blanket way to turn off Google’s A/B testing, but users on a Google Search Help Forum have reported that logging out of their Google account and back in again reverts the links back to blue.  

A Reddit user also reported that disabling “Your searches and browsing activity” in Chrome’s settings turns the links back to blue. To disable the feature go to the Google home page and click on the grid icon in the top right hand corner and select “My Account”.


link to this extract

Errata, corrigenda and ai no corrida:

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.