Live data from Hacker News

One of the Most Alarming Internet Proposals I've Seen

lauren.vortex.com

91–95 of 95 posts

Re: One of the Most Alarming Internet Proposals I've Seen

#91

Earlier quoted context omitted.

HTTP/2 changes the meaning of the http:// scheme. All connections will now be TLS-encrypted. (Edit: Maybe not. See hobohacker below.) http:// means that the endpoint has not been verified using the CA system and is using a self-signed certificate (and is thus trivially vulnerable to a MITM should certificate keys not be checked out-of-band). The purpose is to provide confidentiality to the vast majority of traffic, e…

> http:// means that the endpoint has not been verified using the CA system and is using a self-signed certificate (and is thus trivially vulnerable to a MITM should certificate keys not be independently checked). Right. I understand that. The lack of certificate verification for the http scheme means that ISPs can MITM HTTP traffic with or without this proposal, just like they can with HTTP/1.1. So how does this pro…

The lack of certificate verification for the http scheme means that ISPs can MITM HTTP traffic with or without this proposal, just like they can with HTTP/1.1.

Can they really? At least some of those connections should have the certificates verified out of band, so I'd imagine they would get caught fairly quickly. And then loudly accused of conspiring with the CAs (to stifle competition) or the NSA (to steal your data), or the competitors of whatever site/product was being used, or someone I haven't thought of.

Re: One of the Most Alarming Internet Proposals I've Seen

#92

Earlier quoted context omitted.

Unless Grande and Google eventually decide to do the same thing. Why not, from their perspective?

Frankly, I would be surprised if this was not in Google's ToS from the start, considering that Google's core business model is mining of big data. I suspect there wouldn't be an option to turn it off, not even by paying extra.

Google doesn't need to, your encrypted traffic is to their servers.

Re: One of the Most Alarming Internet Proposals I've Seen

#93
post #62

Earlier quoted context omitted.

I pay AT&T about half that no-DPI price for 12mbit. I wouldn't sweat paying the extra to avoid DPI if I wanted the service.

The point isn't whether one can afford it, but whether it's right .

I agree. But that sad situation looks good from where I'm standing.

Re: One of the Most Alarming Internet Proposals I've Seen

#94
post #79
post #74

Earlier quoted context omitted.

Precisely this is actively being discussed in the W3C WebAppSec WG: http://w3c.github.io/webappsec/specs/subresourceintegrity/ The security folks I talk to are... nervous... about this use of subresource integrity, however.

Can you say any more about what makes them nervous? (What do they think will go wrong?)

Early to say... it's their job to be nervous about new things.

Re: One of the Most Alarming Internet Proposals I've Seen

#95
post #71

Earlier quoted context omitted.

I think the crucial thing to understanding this is this diagram - UserAgent Proxy Server TLS Session #1 TLS Session #2 HTTP which makes it clear that the point is to be able to say "I trust this proxy sufficiently that I'm ok with it acting as an intermediary, but I still don't want my stuff in the clear between me and the proxy or between the proxy and the far end server." So it's about enabling the user to trust -o…

If your definition of "better" is simply "faster", then yes, it is better. If you care about security of your data, end-to-end, this is worse (as others have pointed out because of your ISP being a MitM). In addition, now we are introducing another vector of attack for bad guys to exploit - how enticing does it sound that every ISP becomes a root certificate authority, essentially?

Your ISP could be MitM-ing your plaintext HTTP already, and it's the use cases that we currently use plaintext HTTP for that this is addressing.

If you care about security of your data, end-to-end, then you should probably only use this feature, if at all, with the proxy running on a machine you control - but presumably in that case you currently aren't using plaintext HTTP for anything anyway, so I don't see how it relates.

Post reply on HN