3 ms·

There is a way to stop fingerprinting. That way is serving pages via distributed network (over a WoT or torrent-like thing). All these other ways do is give pe

There is a way to stop fingerprinting. That way is serving pages via distributed network (over a WoT or torrent-like thing). All these other ways do is give people the illusion that they're safe from being tracked, when the reality is that they're tracked just the same, but by fewer people so the data is more valuable. This means that the money is centralizing around the actors with the most inexplicable methods of tracking; which are almost always the worst actors. I hate it too, even though I'm not blameless. It's impossible to compete without a level playing field, and that playing field needs to be technically enforced, because otherwise we get region shopping and advertising / analytics models that push people to create intractable mechanisms so they can paper over how tracking fed into it. For example, imagine a world where I'm bidding to show an ad to a visitor of nytimes.com. Now, I may not track the user, but if anyone is, they can incorporate what they know and sell that traffic back to me on a CPA model. All I see is the incoming traffic. I don't track anyone (wink, wink) but there is no difference. In the long run this will either be solved one way or another, and all these online surveillance capitalism companies will crash and burn. Either we get a web with technical guarantees or we get a balkanized internet where every state makes their own weird laws about what is allowed or not.

newscracker 8y ago
> In the long run this will either be solved one way or another, and all these online surveillance capitalism companies will crash and burn. I sincerely hope that happens without causing more harm to people. It also seems to be a long way away.
- kabacha 8y ago
  https://en.wikipedia.org/wiki/Accelerationism https://en.wikipedia.org/wiki/Accelerationism
scottlocklin 8y ago
I'm pretty sure sticking it on a distributed network is potentially much worse unless it is specifically designed to prevent tracking.
mindslight 8y ago
Protocol wise, basic shared VPNs will stop most everything short of a semi-global passive adversary. The problem is running hostile code on your own machine, coupled with browser makers thinking it is a generally fantastic idea to allow that hostile code to access a whole slew of security-sensitive information. Sure, VPN won't repudiate the region bullshit, and can even be outright blocked. But if adoption rose to the point websites didn't want to lose the traffic, those would diminish. However, I do agree a non-immediate user-centric protocol is sorely needed though, especially to stop those global snooping adversaries.
- richjdsmith 8y ago
  >and can even be outright blocked. That's my biggest fear. Yesterday I woke up to find out that Three (a big telco provider in the UK) had blocked Mullvad's API path as 'adult content'. I had to physically go to a store and verify that I was an adult to be able to use my VPN - the very same VPN I use because I have so little trust in the UK Gov (I'm not from here and am not staying long term, but hate that the society reminds me of 1984 regularly),
  - mindslight 8y ago
    I was referring to blocking by websites, as there is actually some competition between them. For example, a few major ecommerce sites frustrate or outright block traffic from my VPS, so I've simply stopping visiting them rather than poking holes. A similar dynamic is present for ISPs wherein if most people expect "Internet access" to mean something that works with VPNs, then ISPs can't block it [0]. But that's a harder state to reach as there's much less competition between ISPs. A commerce site wouldn't want to forgo 10% of their business, but a government/quasi-government has no problem demonizing 10% of their subjects. Which is why we ultimately do need non-real-time protocols and namespaces for the bulk of communication. [0] Nagging systems like the UK would still exist, but they couldn't progress further to outright banning it.

newscracker 8y ago

> In the long run this will either be solved one way or another, and all these online surveillance capitalism companies will crash and burn. I sincerely hope that happens without causing more harm to people. It also seems to be a long way away.

kabacha 8y ago
https://en.wikipedia.org/wiki/Accelerationism https://en.wikipedia.org/wiki/Accelerationism

scottlocklin 8y ago

I'm pretty sure sticking it on a distributed network is potentially much worse unless it is specifically designed to prevent tracking.

mindslight 8y ago

Protocol wise, basic shared VPNs will stop most everything short of a semi-global passive adversary. The problem is running hostile code on your own machine, coupled with browser makers thinking it is a generally fantastic idea to allow that hostile code to access a whole slew of security-sensitive information. Sure, VPN won't repudiate the region bullshit, and can even be outright blocked. But if adoption rose to the point websites didn't want to lose the traffic, those would diminish. However, I do agree a non-immediate user-centric protocol is sorely needed though, especially to stop those global snooping adversaries.

richjdsmith 8y ago
>and can even be outright blocked. That's my biggest fear. Yesterday I woke up to find out that Three (a big telco provider in the UK) had blocked Mullvad's API path as 'adult content'. I had to physically go to a store and verify that I was an adult to be able to use my VPN - the very same VPN I use because I have so little trust in the UK Gov (I'm not from here and am not staying long term, but hate that the society reminds me of 1984 regularly),
- mindslight 8y ago
  I was referring to blocking by websites, as there is actually some competition between them. For example, a few major ecommerce sites frustrate or outright block traffic from my VPS, so I've simply stopping visiting them rather than poking holes. A similar dynamic is present for ISPs wherein if most people expect "Internet access" to mean something that works with VPNs, then ISPs can't block it [0]. But that's a harder state to reach as there's much less competition between ISPs. A commerce site wouldn't want to forgo 10% of their business, but a government/quasi-government has no problem demonizing 10% of their subjects. Which is why we ultimately do need non-real-time protocols and namespaces for the bulk of communication. [0] Nagging systems like the UK would still exist, but they couldn't progress further to outright banning it.