The idea of “SSLs” as security certificates for online transactions is a very old one. It was around when the first “intellectual property” was created back in the 1800s. It has been around since the late 1800s. Since then it went through several revisions, and it’s still evolving.
In its current form, it’s a combination of a “private” certificate with a “public” certificate, and a “client-side” certificate for transactions on a server with a private key. The private key is stored in a private cache, which is used to allow SSL to be used for connectionless transactions.
A client-side certificate is a certificate that is issued by an agent to a client, and then stored by that client in the client’s certificate store, as long as the client is online. The private key is available from the user, and is used to sign the certificate.
If anyone’s a party-slick, you should have a public certificate for a transaction on a server.
This is the most common form of a server, and it’s the most common form of a client-side certificate that I’ve ever seen. But it doesn’t really matter too much what type of certificate you’re using in the server.
The problem is not that the client is storing your private key, but that the server is storing your private key. You might think the server is storing your private key because the client is using a certificate from a CA, or is using a certificate from a certificate authority, but the problem is that the server is not accepting the client certificate.
There is no problem in storing your private key in the server, but if the client is using a certificate from a CA, then suddenly the server is storing your private key and the CA is not. There are two ways to fix this. One is to use a CA that requires client authentication and a certificate from the CA. The other way to fix this is to use a cert for which you have the private key.
The reason this is so hard to fix is that the CA can’t be trusted. In the past the user was often able to access the client certificate from a remote server which caused the CA to crash and crash. Now the CA has to run a test for security and that test is now impossible. The CA is not trusted by the server, so the user must either be forced to trust the client or run a test for security.
The most obvious way to fix this is to use a certificate for the certificate chain. The server can be used as a security certificate by the client. So, if you’re the client, you have to run a test for certificate chain security with the certificate chain. We want to make sure the CA isn’t running your certificate chain. That means we can’t use a certificate chain for server certificates.
I do agree that it would be nice if the server could be used as a security certificate, but I also think it would be nice to only have to trust that certificate as the client.
