Live data from Hacker News

Introducing BrowserID: A better way to sign in

identity.mozilla.com

31–40 of 188 posts

Re: Introducing BrowserID: A better way to sign in

#31
post #10

This seems to encourage using the same credentials everywhere which I think most agree is a "bad idea." If BrowserID is compromised, the attackers have access to all the sites where I use browser id, right?

The same could be said about OpenID and OAuth providers like Twitter, Facebook, Google, etc. The idea of having a centralized authentication provider is definitely not a new one. There are many advantages to using 3rd party authentication providers. The biggest security advantage 3rd party auth providers have is that they do security right and often they implement their authentication service using a commonly defined protocol. Many websites that choose to roll their own systems make many security mistakes and many people use the same credentials on many sites. So if any website is not secure authentication a user's login information is likely compromised across multiple websites because people generally use the same credentials.

Re: Introducing BrowserID: A better way to sign in

#32
post #22
post #4

How is it different from OpenID, apart from it's not decentralized?

It actually is decentralized, in the sense that anyone can implement a BrowserID provider.

Anyone can, but from the looks of it, a site chooses to trust one at a time.

Re: Introducing BrowserID: A better way to sign in

#34
One huge problem: Email address != identity.

I should be able to change my email address (and/or email hosting provider) without changing my identity on a bazillion sites around the internet. Facebook got this right from the beginning. Google is sort-of getting this, although the chasm between Google Accounts and Google Apps Accounts makes this really messy.

Really this product should be called BrowserEmailAddress, not BrowserID. It doesn't serve identity.

Re: Introducing BrowserID: A better way to sign in

#36
post #22

Earlier quoted context omitted.

It actually is decentralized, in the sense that anyone can implement a BrowserID provider.

Anyone can , but from the looks of it, a site chooses to trust one at a time.

There are two separate things here. Using the terminology from http://lloyd.io/how-browserid-works , you have:

1. Primary Identity Authority. This is a host (i.e. the one in your email address) that supports BrowserId. This is fully decentralized.

2. Implementation Provider/Secondary Identity Authority. For now the site has to choose one to trust, but once your browser has support it becomes the IP, and when your host becames a PIA there's no need for an SIA.

Re: Introducing BrowserID: A better way to sign in

#37

One huge problem: Email address != identity. I should be able to change my email address (and/or email hosting provider) without changing my identity on a bazillion sites around the internet. Facebook got this right from the beginning. Google is sort-of getting this, although the chasm between Google Accounts and Google Apps Accounts makes this really messy. Really this product should be called BrowserEmailAddress, n…

I agree that email address != identity, but nothing would stop a site that uses BrowserID from allowing a user to change the email address that they use on that site.

It's very similar to the countless existing services that rely on email for identity... you'd just have to verify ownership of the new email address (usually through a confirmation email).

Re: Introducing BrowserID: A better way to sign in

#38
post #23

What I'd really want to see is public-key authentication for website. Let me upload my public key when I create an account on a website, and let the browser interact with my ssh-agent to authenticate.

This already exists, but the UX is terrible and there's a chicken-and-egg problem in getting it fixed in every (deployed) browser.

http://www.gnegg.ch/2008/05/why-is-nobody-using-ssl-client-c...

Re: Introducing BrowserID: A better way to sign in

#40
post #4

How is it different from OpenID, apart from it's not decentralized?

This lets users sign in with an existing email address, so they don't need any new sevice or identifier to remember. It's decentralized; Mozilla has a service for web developers for convenience, but any site can implement the protocol itself instead (or use another provider). And it's designed to let browsers handle the login flow in the future, simplifying login and account creation for end-users.

Most of the oauth services I used allowed me to create an account with ... an e-mail and a password.

Literally the same thing, here.

Post reply on HN