The sad thing is if the clipper chip was ubiquitous today that could have killed the ability for spammers to spoof numbers. Some days I can get a dozen robo calls, each with a different spoofed number. People that want non-backdoored crypto could still do so.

I'm still blown away at how hard it is to get people to use any encryption, even people who work in infosec/etc. If nothing else, 2020 was a great year for the uptick in using crypto to communicate with non-technical fam and friends.

Robo calls should be filtered at the network level. The Clipper Chip would have done nothing to prevent this.

There is actually encryption technology[1] available that would solve this or at least make it traceable and blockable. But here we are.

[1] https://en.wikipedia.org/wiki/STIR/SHAKEN

Edit: Also the political situation at the time was that if Clipper was adopted, all other encryption technology would be outlawed. It was a very scary time.

The key word here being "should", and they are clearly not filtering at the network level. I can tell, I still get spoofed robocalls... The clipper chip provided end users the ability to cryptographically prove callers were who they claimed to be independent of the network operator, which is a key need of a crypto system.

Robocalling with spoofed ANI has been a problem long enough that I think we can safely say the network operator is 100% complicit with this activity now. The phone company could track down kids war-dialing blocks of numbers in the 1980's in order to make sure they were not telemarketers not paying higher telemarketing fees. Do people really think the phone company is not getting a cut of these robospoofers?

You don't need a clipper chip to trace nuisance calls. You need an authenticated network that records the origin of calls, and hopefully passes that through when calls are forwarded and makes it simple and worthwhile to report nuisance calls and aggregates those reports and takes meaningful action against the origins.

Anyway, we're getting shaken/stir or whatever RealSoonNow(TM), so we'll probably have better CallerID. I don't think it'll be enough to solve the problem, without a reporting mechanism, but I guess we'll see in the next couple years.

"You don't need a clipper chip to trace nuisance calls. You need an authenticated network that records the origin of calls,"

So where is this magical solution you speak of? We're all products that the phone company sells to telemarketers, there is no incentive for them to give us these tools now.

Again, the clipper chip would have made it difficult for anyone to pretend to be someone they are not, cryptographic signature is part of the deal to ensure you are talking to who you think you are talking to.

Would you want the same capability on the Internet? If not, what makes Internet communication different from POTS communication?

Internet connections have a more-or-less reliable source IP (spoofing does occasionally happen and we take measures to mitigate it), IP ranges have owners and if an address is consistently used for abuse (attacks, spam email, ...) then people do report this to the owner of that address?

> IP ranges have owners and if an address is consistently used for abuse (attacks, spam email, ...)

IP ranges have registrants. Servers usually are assigned IP addresses temporarily by the registrant, or by someone the registrant has assigned the block of addresses to by some other means. This means differs from region to region. In the US the responsible party is ARIN, and registrants can reassign addresses using a database called SWIP. In the EU both registrants and their partners use the same database called RIPE. I have never registered addresses in other regions.

> then people do report this to the owner of that address?

Yes. Registrant again, but yes. And if you don't get satisfaction, then you can (and should) escalate all the way to the region's authority.

Yes.

Email has a "from" address that the SMTP protocol (the thing that often carries emails) that is separate from the IP address of the sending server, the exact same way "CALLER-ID" is a separate field from the billing address (called ANI, or Automatic Network Identification) in telephone networks.

Requiring that ANI match CALLER-ID would break many voicemail systems and call-forwarding systems. They can be fixed, but it will be expensive to do so.

Similarly, requiring the email "from" address agree with the SMTP server's responsibilities breaks mailing lists and email-forwarding, and again, this can be fixed, but it is proving expensive to do so.

I would want to require:

1) That the source IP address be from a range 'controlled' by the operator of a given AS. This might be proven, semi Out of Band, for a duration by a PKI challenge with a key representing that authority. (Allows distributed services on the same IP anywhere. Fulfillment can be asynchronous if UDP.) Edit: This would also be how to securely claim an AS route; you might still need a link authorized for this level of service.

2) The mentioned feedback mechanism should be coupled with 'do not forward me anything from IP || net/mask for X time' via a similar mechanism. A reason might be provided, reasons of clear abuse MAY be aggregated and used to isolate misbehaving hosts / networks.

> Some days I can get a dozen robo calls, each with a different spoofed number.

I live in the EU and have literally never gotten a robo call in my life. It's a political problem, not a technical one.

The US government intentionally made it inconvenient and infeasible to include encryption at a key time in technological development, and I think it had lasting consequences.

Yes, if the internet community had acceded to their demands to use the clipper chip, that would have been one route to them lowering the barriers they had put in place to encryption. They also could have just recognized that widespread encryption was in the national interest without the clipper chip, instead of deciding it was opposed to it.

Yes, the only munition available on a t-shirt!

Did PGP did that too? I know DeCSS did this. Did they borrow the idea from PGP?

Yes. http://www.loundy.com/Roadside_T-Shirt.html

And AACS as well.

I remember buying some FreeBSD disks from Walnut Creek CDROM in the 90s, these shipped to my home in Mexico. When they Arrived, apparently Kerberos was not included in the distribution because "it was not possible to export weapons outside of the USA". 12 year old me was completely puzzled at what that was about.

The government thinks it still is.

PGP's biggest weakness was that it was too early. There was no normal user accessible software available. No regular person was going to establish the web of trust or use command line utilities.

So now we have easy, strong encryption and the keys are controlled by...someone. Definitely not the user, though.

Funny story - even Phil Zimmerman can't use PGP: https://twitter.com/josephbonneau/status/638772283713060864 So maybe it is just a hard problem.

No, PGP's weakness is that the Web of Trust is an unworkable solution for the general population for key exchange. It works fine for a you and your circle of crypto nerds, but as a general solution it's impossible.

IMHO this is a case where perfect was the enemy of good. Many possible solutions were rejected because there was a possibility that someone could MITM your first contact, even though in the real world this is unlikely. The key registries were almost a solution but none of them ever gained enough traction to be a default solution and mail clients were strangely hesitant to incorporate them even when they did implement PGP.

PGP was always half of the solution. Sadly they never figured out the other half. Microsoft almost got it working with Exchange, but even then you usually it only works on a single domain at a time. You can't use encrypt an email to someone at a different company even if they are using Exchange.

Web of Trust is a perfectly fine system but UI-wise a total disaster. There's also the reality that what most people need most of the time is a web of trust that includes government agencies, banks and other major institutions.

For the vast majority of useful communications for users, your web of trust ideally goes You -> Government ID Agency <- Recipient. And that's fine - that's what people actually need. People get scammed or deceived by faking those credentials.

Having to physically meet people to exchange long hex strings was never going to scale.

No but that's the point: you do have to physically go to the DMV, or the bank, quite frequently. And for most people, the important point to point communications are interactions you want validated against those entities.

Web of Trust mixes two kinds of trust.

If my mom gives me her public key, I trust that it is her public key. If my mom signs some other person's public key - I really don't know how much I trust that. Trusting that somebody is who they say they are is not the same as trusting them to properly vouch for another person.

This is still a UI problem though more then any other: Web of Trust stored and it could have displayed the actual verification chains, and set up some decent defaults based on that - i.e. "Government", "Bank", "Personally Verified", "Friend of a Friend" - all of this would be easy to communicate via what keys signed what exactly who you were dealing with.

This even leads to a logical DNS integration: debian.org advertises the core group of keys which should verify people on that address via DNS, and it shows up as "DNS-only" or something.

Good crypto frequently undermines itself by trying to be adversarial to the whole concept of government or big companies (look at TLS - it succeeded because it's the antithesis of this) but those the primary users and coordinators that can drive adoption.

I have this complaint with Signal right now: if Signal wanted a legitimate funding source, they should sell a "verified Signal" service to let companies subscribe to use Signal as an alternative to SMS providers - my father wants to do this for his small business right now, to replace the SMS bulk sender and be able to send larger files to people securely.

Instead we've got whatever cryptocurrency ridiculousness.

This is the reason Frendster failed. Friending someone automatically made you a friend with all of their friends, even the ones that were jerks or you had never met.

Notably, in recent hosted versions of Exchange that has been resolved since they can use Azure AD to authenticate across tenants, and if you don’t have Office 365 you can make a temp account in their tenant when you click on the link to see the email. It works fine (unless you’re still on-prem for email).

I fundamentally disagree with you that web of trust is hard.

There are much harder problems solved in easy to use ways, there’s just very little financial gain for this problem.