Supporting newspapers harms your privacy

2021-01-28

I am a Gaurdian reader. Have been since 2006.

I am a Guardian subscriber. Have been, on & off, since they introduced subscriptions.

I still see this banner every time I go to the site. (OK, I would if I didn't have an adblocker.)

---

The point I am clumsily trying to make is that I pay but I don't sign in.

When I don't sign in, I have the option/notion of rejecting all tracking. Once I sign in, all bets are off.

---

Thankfully, the Guardian only has annoying ads and not a paywall. Unlike the other newspapers I subscribe to.

Newspaper I pay for	Can I read without logging in?
The Hindu	No
The New York Times	No
Wall Street Journal	No
Live Mint	Yes
The Guardian	Yes

Why do newspapers want you to sign in?

The obvious, if naive, explanation is that you need to answer are you a subscriber?. This is essentially the AAA problem:

• Authentication
• Authorisation
• Accounting

But the answer to the question above is solely Authorization. You don't really need Authentication or Accounting.

• The question Authentication answers is Who are you?.

• Accounting answers What and how much are you consuming?

Sure, some degree of Accounting may be needed for tiered access and so on, but last I checked none of these sites offer anything of the kind.

So, why do they really need AAA?

I don't know the answer. But I can guess. It's because they really do want the answers to the age old questions:

• Who are you?
• Why are you here?

Either that or they have a bunch of lazy coders.

Because there's a better model

Let me introduce you to Blokada. I have been using it for a couple of years to block trackers and ads on my phone. A few months ago, with the WeChat ban in India, I signed up for Blokada Plus. This gives me access to the toggle button at the bottom of that screen.

Blokada works on my phone, on my web browser and my desktop. And it does not know who I am.

How does it do that?

By asking the only question it needs to, Are you a subscriber?.

And when it does, I provide my unique, randomly generated, 12 character account ID and all is good.

To access the VPN from my desktop, I log into the web interface with, you guessed it, the account ID. Once logged in, I can generate cryptographic client certificates to use with my VPN client.

What's the catch?

In theory, this model is no more immune to tracking than any other.

• Blokada has a payment processor who did receive some of my identity and card information - I didn't choose the BTC payment option.
• The payment processor could have shared this information with Blokada.
• Blokada could link the "private" account ID with the identity information.
• Blokada could, right now, be compiling a dossier on every site and app I use.

BUT...

In practice, it is possible to make the system pretty rugged.

Let's imagine a much more private flow for my beloved Guardian.

• I get tired of the half-page banners/want to encourage better journalism and opt to subscribe.
• The Guardian clears all identifying cookies and session data about me from my browser and their servers.
• I am redirected to a privacy focused payment processor who doubles as an IDP.
  • UPI, this is not.
• The payment processor generates a unique ID and sends it to Guardian Towers who then present it to me and make me promise to keep it safely.
  • This could even be a long, memorable phrase.
• I log into Guardian on all three devices I own with this unique ID.
• The Guardian verifies that the ID is kosher with our SuperPrivatePaymentProcessor.
• I rest easier knowing that Barney Ronay will never know that I read every one of his articles.

It's not that hard. We just need to take this shit seriously.