Linked by David Adams on Mon 8th Nov 2010 16:49 UTC, submitted by HAL2001
Thread beginning with comment 449124
To view parent comment, click here.
To read all comments associated with this story, please click here.
To view parent comment, click here.
To read all comments associated with this story, please click here.
Developers are not the problem. It's the CAs and the browser vendors. Producing scary errors on self-signed certs protects absolutely bloody nobody and locks password protection to the SSL racket.
I'll agree that the entire concept of SSL certificate/encryption, with CA's and high prices, is indeed a broken system and a scam to some extent - but the "scary errors" browsers display have a valid purpose.
Given how SSL works, and how users expect it to behave, if you can't verify the certificate you're using belongs to the site you are surfing, you can't know that the encryption keys you're sharing with them haven't been tampered with by a middleman. On a public wifi network, this can be a real threat...
In any case - if I encounter a site with an "untrusted" certificate, and I don't figure it matters for that particular site (read: I'm not revealing personal information to the site), then I'll just accept it anyway.
These days, you can get a free Class 1 cert (unrevokable, single domain)... or a cheap Class 2 verification wildcard cert for like $25/year ($50 for two years) from StartCom:
http://www.startssl.com/
All major browsers accept these... so it's hard to complain about it much.
Edited 2010-11-08 22:46 UTC




Member since:
2005-11-10
The reason so few use SSL on their sites is the brain-dead stupid business model of SSL certs. Tying identity to encryption is a misdirection. Encryption should not require identity. SSL certs are expensive and simply don't prove anything useful.
Developers are not the problem. It's the CAs and the browser vendors. Producing scary errors on self-signed certs protects absolutely bloody nobody and locks password protection to the SSL racket.