Commonplace

I subscribed to Fastmail’s newsletter and was surprised to see they use Mailchimp to send them. But it turns out they create temporary email aliases for each subscriber every time they send an email.

Each time we send a newsletter, we create new temporary aliases for everyone to ensure we respect any changes in your newsletter preferences and keep your personal addresses private from our external email marketing tool (Mailchimp).

The temporary address will then expire, meaning you’ll disappear without a trace on Mailchimp and never receive any mail sent to that alias again.

At the moment, newsletters are sent 2–10 times per year. Each time a new newsletter is sent, a new list of aliases will be imported, reflecting any new additions or subtractions to users on the list.

It’s impressive – and cool – to see privacy-conscious companies walking the walk.

Cooler still, they don’t even store generated aliases:

It’s actually not too much effort! We generate time-limited aliases by hashing together your real address + expiration time + some randomness. When we get the mail, we reverse that process to look up your account and deliver it.

(We don’t even store the generated aliases!)

This all points to a possible future in which social-media giants like Facebook may soon be past their long stretch of dominance. They’ll continue to chase new engagement models, leaving behind the protection of their social graphs, and in doing so eventually succumb to the new competitive pressures this introduces. TikTok, of course, is subject to these same pressures, so in this future it, too, will eventually fade. The app’s energetic embrace of shallowness makes it more likely, in the long term, to become the answer to a trivia question than a sustained cultural force. In the wake churned by these sinkings will arise new entertainments and new models for distraction, but also innovative new apps and methods for expression and interaction.

In this prediction, I find optimism. If TikTok acts as the poison pill that finally cripples the digital dictators that for so long subjugated the web 2.0 revolution, we just might be left with more breathing room for smaller, more authentic, more human online engagements.

Food for thought (and hope).

There’s just one little problem with personalisation: it doesn’t make any sense. We believe the case against personalisation is significantly stronger than the case for it.

Most personalisation efforts are powered by third-party data. Marketers infer who customers are based on their browsing behavior. So how good is that third-party data? It must be extremely good, if you’re claiming to understand buyers on a “personal level”.

Spoiler alert: it’s not. Most third-party data is, to put it politely, garbage.

In an academic study from MIT and Melbourne Business School, researchers decided to test the accuracy of third-party marketing data. So, how accurate is gender targeting? It’s accurate 42.3% of the time. How accurate is age targeting? It’s accurate between 4% and 44% of the time. And those are the numbers for the leading global data brokers.

Many enterprise technology companies spend millions of dollars ‘hyper-targeting’ IT decision makers (ITDMs) using third-party data. But if we get gender wrong more often than 50% of the time, what percentage of ITDMs do you think are actually ITDMs, according to the research?

Do you want to guess? It’s 14.3%. And for ‘senior ITDMs’, that number drops to 7.5%.

A report by Reveal and The Markup found that “Facebook is collecting ultra-sensitive personal data about abortion seekers and enabling anti-abortion organizations to use that data as a tool to target and influence people online…

No surprises there!

From the Washington Post:

The company (Google) received nearly 150,000 requests for user data from US law enforcement in the first half of 2021…and it handed over information on users in 78% of those cases. An estimated 26 states are expected to ban or heavily restrict abortion, and prosecutors will almost certainly go to tech companies, such as Google and Facebook…to seek the evidence they need to charge people who help provide the procedure.

Hoffman’s summary neatly wraps it up:

For years, those who couldn’t see beyond their own noses couldn’t understand how “I have nothing to hide” was so fucking stupid. In an environment in which marketers know everything about us and governments try to know everything about us, everyone has something to hide. We just don’t know what it is.

The notion of surveillance advertising being perpetrated by ad platforms via social interactions is a myth: there is no omniscient social media entity spying on its users and hoarding their interaction data to power ads targeting.

This piece argues that surveillance advertising is a myth because:

1. Social media’s engagement algorithms (i.e. user likes a video) don’t interact with advertiser’s info (user buys product and that’s fed back to the social media platform)
2. It’s advertisers that do the surveilling, not social media platforms

But sending data about a user’s interactions on third-party sites back to social media platforms for ads targeting is precisely the ‘surveillance’ part of ‘surveillance advertising’. Especially when you consider how that data is used for other advertisers – that a user hasn’t interacted with – through lookalike audiences etc.

The piece mention’s the W3C definition of tracking:

Tracking is the collection of data regarding a particular user’s activity across multiple distinct contexts and the retention, use, or sharing of data derived from that activity outside the context in which it occurred.

That definition of tracking sounds awfully like the data being sent to social media companies from third-parties. Tracking is another word for surveillance.

Surveillance. Advertising.

The piece also argues that the spheres of misinformation and on-platform engagement are unrelated to data fed back to social media platforms. But, as anyone who’s read Christopher Wylie’s Mindf*ck – a detailed account of Cambridge Analytica’s activities – this data has literally been used to anticipate behaviour at a population level and use that to manipulate elections and referendums.

Great tip from John Gruber, and something all iPhone users should know:

The first is hard-locking. When you hard-lock your iPhone or iPad, it enters a mode that requires the device passcode to unlock. With recent iPhones and iPads, you enter this mode the same way that you turn off the device: by pressing and holding the power button and either of the volume buttons for about two seconds.

[On iPhone 8 and later] you can also do the same thing by quickly pressing the side button alone five times.

Tracking is the collection of data regarding a particular user’s activity across multiple distinct contexts and the retention, use, or sharing of data derived from that activity outside the context in which it occurred. A context is a set of resources that are controlled by the same party or jointly controlled by a set of parties.

Good to see this defined by the W3C.

Command + Shift + 5, go to options and select location. Game changer

Sheryl Sandberg announced this month that she’s resigning from Facebook—now called Meta—to focus on her philanthropy. Her work there is done.

During her 14 years at the company, she’s done so much damage to our society that we may never recover. The simple truth is that you cannot simultaneously dedicate yourself to making untold fortunes for a giant corporation and to championing a social good.

This may not be a pivot to data justive warrior, but this get-rich-working-at-horrendous-tech-co before using those funds for conscience-clearing philanthropy has more than a hint of Maria Farrell’s Prodigal Techbro about it.

A website is a file or bundle of files living on a server somewhere. A server is a computer that’s always connected to the internet, so that when someone types your URL in, the server will offer up your website. Usually you have to pay for a server. You also have to pay for a domain name, which is an understandable piece of language that points to an IP. An IP is a string of numbers that is an address to your server.

Links (rendered default blue and underlined—they’re the hypertext “HT” in HTML) are the oxygen of the web. Not all websites have links, but all links connect to other webpages, within the same site or elsewhere.

What a wonderful explanation.

Today more than ever, we need individuals rather than corporations to guide the web’s future. The web is called the web because its vitality depends on just that—an interconnected web of individual nodes breathing life into a vast network. This web needs to actually work for people instead of being powered by a small handful of big corporations—like Facebook/Instagram, Twitter, and Google.

(Emphasis my own)

I couldn’t agree more.

This whole article is a fanastic look at what a website is, what it can be and why it’s important for the web to be diverse.

The notion that if a company has built a business model on top of privacy-invasive surveillance advertising, they have a right to continue doing so, seems to have taken particular root in Germany.

I’ll go back to my analogy: it’s like pawn shops suing to keep the police from cracking down on a wave of burglaries.

Too right, and it’s not just that the surveillance advertising is unethical – the data has often been collected illegally.

Centuries of pre-internet advertising prove that tracking isn’t necessary for advertising to work…

Daring Fireball is a fantastic example of this.

…but no one is arguing that tracking isn’t effective.

Effective at driving huge profits perhaps. There is growing evidence that demonstrates tracking ads aren’t that effective at their core activity.

From Augustine Fou (emphasis my own):

A 2019 study showed that a single targeting parameter, gender, derived from anonymous website visitation patterns was only 42% accurate, worse than random. If you did no targeting at all, and just did “spray and pray” with your digital ads, you’d at least hit one of the genders 50% of the time. When two parameters were taken into account – age and gender – the accuracy dropped as low as 12%. That’s like 9 times out of 10, those targeting parameters were wrong. And advertisers paid extra to make their digital marketing worse, not better.

Then there’s Bob Hoffman’s infamous Programmatic Poop Funnel. That chart showed how only three cents out of every dollar spent on programmatic ads is seen by humans.

And we can’t forget Tim Hwang’s Subprime Attention Crisis:

…the accuracy [of data profiles used for advertising] was often extremely poor. The most accurate sets still featured inaccuracies about 10% of consumers, with the worst having nearly 85% of the data about consumers wrong.

There are many more examples.

But I wonder how many clients would be happy to continue spending on surveillance ads given how inaccurate and expensive they are?