3 ms·
> The first is spam and abuse. The problem with a purely "open web" is that there's no solution to those, because of bad actors. The only solution we've found s
by naasking 6y ago
> The first is spam and abuse. The problem with a purely "open web" is that there's no solution to those, because of bad actors. The only solution we've found so far are centralized organizations which can run sophisticated machine learning and hire thousands of moderators.
Not true. It's the only solution when your focus is growth, and so you permit open signups from anywhere needing only something like an email address. This enables brigading, bots and abuse with fake and anonymous accounts.
If a web of trust were involved, you have an introduction chain for every user, and a bad actor revealing themselves casts suspicion on the whole chain. You can revoke the whole chain in one go which is far more effective than picking bad actors and bots off one by one. The investment needed to sneak in is also considerably higher.
So the status quo is not the only solution, it's the lazy solution focused on growing as fast as possible with no regard for the challenges that raises.
- jandrese 6y ago> The second is aggregation and discoverability. What good is it to publish if nobody finds you? Matching what people want to read, with the content they'll like (and not turning it into a firehose) is a really hard problem which, again, open standards basically do nothing for. Relying on a web of trust for signups makes the second problem far worse. If you're only creating a service for your friends and family it works great, if you want to expand it to the general public it is impossible. The encrypted email folks have been trying to get the web of trust to work for almost 30 years without success.
- naasking 6y ago> If you're only creating a service for your friends and family it works great, if you want to expand it to the general public it is impossible. Not impossible at all. Send out invitations and let them pull in their groups, as but one option. GMail was initially invitation only if you recall, and it was crazy popular even then. > The encrypted email folks have been trying to get the web of trust to work for almost 30 years without success. Web of trust has been working fine for all of those 30 years, so I really don't know what you're talking about. What hasn't "worked" is commercial buy-in by email clients and, more importantly, email providers who want to mine your data. People choose the free, easy email options (gmail, yahoo, etc.) without realizing how much personal information they're disclosing. So to claim that the commercial interests who are interested in violating your privacy and mining your information haven't bought into the web of trust is a pretty solid endorsement, not a downside.
- jandrese 6y agoGMail was offering email, a service people already wanted and could access. Email is the first social media platform, Google was just offering a fancy new client to an existing platform. A better example would be Google Wave, or Google Hangouts, or Google+, or Google Talk. The fact you think the Web of Trust is working fine when it is still basically impossible to send an encrypted email to anybody even when you have your client set up properly is somewhat mind boggling to me. A solution is not working if it is impossible to use.
- naasking 6y ago> Google was just offering a fancy new client to an existing platform The point is that invitation-only works just fine if the incentives align. > The fact you think the Web of Trust is working fine when it is still basically impossible to send an encrypted email to anybody even when you have your client set up properly is somewhat mind boggling to me. I used encrypted PGP email from within gmail using a browser addon for years. It was easy, so "impossible to use" is simply false. The incentives to make it easier simply don't exist because commericial interests don't value your privacy or systems that are robust against abuse, so it gets no investment and whatever makes growth easier and faster gets all the investment. The fact remains that the original claim that "the only solution we have" are centralized organizations with moderation and machine learning is simply false. Web of trust is a solution to the broad violation of privacy and the abuse allowed by open, anonymous systems. If people truly valued those things, then they would clamour for software that would make the web of trust easier to use.
- jandrese 6y agoHow many encrypted emails do you receive annually? There have been many post-mortems on the Web of Trust. Few mention "corps can't steal my data" as a failure, the failure is entirely in the impossibility of building the web of trust with enough people to gain critical mass. In fact one danger of the web of trust sometimes mentioned is the way it leaks your social graph just like a social network, only without the privacy protections provided by services like Facebook. https://inversegravity.net/2019/web-of-trust-dead/ https://inversegravity.net/2019/web-of-trust-dead/ https://medium.com/@bblfish/what-are-the-failings-of-pgp-web-of-trust-958e1f62e5b7 https://medium.com/@bblfish/what-are-the-failings-of-pgp-web... https://security.stackexchange.com/questions/68583/is-the-pgp-web-of-trust-keyserver-infrastructure-permanently-broken https://security.stackexchange.com/questions/68583/is-the-pg... https://webcache.googleusercontent.com/search?q=cache:aHYu1jYr0y0J:https://cryptome.org/2013/09/web-trust-sucks.htm+&cd=20&hl=en&ct=clnk&gl=us https://webcache.googleusercontent.com/search?q=cache:aHYu1j... https://cryptography.fandom.com/wiki/Web_of_trust#Web_of_trust_problems https://cryptography.fandom.com/wiki/Web_of_trust#Web_of_tru... I've had my keys published and PGP set up on all my email accounts for years and literally the only encrypted emails I've ever gotten are from Facebook, and only because I went through a fairly arcane process to set that up. I even crypto-sign all of my email as a hint that I can do PGP. I doubt anybody has ever verified my signature on even one piece of email.