Quite coincidentally, I was thinking about this this afternoon — given that URLs are by definition unique, and that we already have a system in place for allocating domains, the groundwork is already in place.

To my mind, though, there’s a missing link in this approach — while a unique URL / OpenID style system would allow for authentication and validate authority, how would one ensure that it matched a real world identity?

What’s to stop me fabricating a fake online identity, other than some cybersquatting legislation and rules? On the flip-side, does it actually matter?

Josh, I’m really surprised to hear that last line come out of your mouth (fingers?). It’s only to geeks that it could make sense that “identity is seeking a URL.” To most people, my Mom for instance, an URL is a place on the web, not a person and certainly not “me.” Overloading URLs with something else is going to be insanely confusing to people who are still typing http://www.yahoo.com in Google’s search box. It’s not what we’ve asked them to learn for the last ten years and it doesn’t mirror the real world or, ahem, real social behavior.

Good point, Andrew — what’s your take on i-names? They resolve the URL problem by, um, not being URLs, but have the same uniqueness properties.

andrew…I actually don’t know what I meant by that. ..certainly our identity is multi-faceted and complex…I was referring to the tendency for schemes that map a single person’s identity to a URL (like OpenID).

Maybe I’ll take it out so as to not confuse…

OpenID introduces a single point of failure to web-based accounts in order to solve a complete non-problem. Just one account will need to be cracked or exposed by corrupt employees to gain login privileges and view private information on any linked account. I’m surprised that it’s gaining so much political traction so soon after breaches like 2005’s ChoicePoint debacle. Why are so many smart people jumping on this solution the utter non-problem of “lots of accounts”, when software like OS X Keychain provides a SECURE, LOCAL approach?

Some possible unintended consequences of OpenID: LittleguyWebsite starts an OpenID service, but gets compromised; how does this affect ID consumers who may want to know they now have a wolf in sheep’s clothing? SomeBigCo decides not to accept OpenID’s from LittleguyWebsite for this reason; running your own OpenID server is useless because it will be treated like a photocopied drivers license. Users in need of “identity” start using SomeBigCo for their OpenID’s; SomeBigCo knows about your SomeBigCo accounts, your OtherBigCo accounts, and all your LittleCo accounts, and sits on a goldmine of valuable consumer data its shareholders demand to see monetized. OtherBigCo is unhappy that SomeBigCo knows its users’ account details, so it refuses to accept SomeBigCo’s OpenID’s and insists on just its own. This is the exact situation we’re in now, except a ton of energy and effort has been wasted to get there.

Basically, I don’t believe that Google, Yahoo!, and Microsoft are going to end up accepting any OpenID’s but their own. This isn’t even a problem because **it’s a good thing** that all these companies can’t reliably correlate my identity across services.

Goog point. But remember that nobody really owns a domain. Everybody is just renting domains from the different official providers. Tying your personal or business information to something rented may be not such a good idea. But anyway, it will be a good thing.

Michal, you bring up a very interesting point. I actually wondered about that myself, and figured that the OpenID stuff would be rather low-end…non-critical applications that don’t involve credit cards and the like.

I think you should publish your considerations on a wider forum, though. It’s definitely good food for thought.

I don’t beleive that it will work better than what we have today working. Everything becomes disappointing in 1-2 years as it also becomes a target for spam and all such stuff.

Josh: you’re right, I expanded on my comment here: http://mike.teczno.com/notes/openid-again.html

Not sure what a more appropriate wider forum is. =)

This is a lot like what the semantic web people are always going on about with RDF – relationships between people and places and things and organizations mapped by URLs.

Linking your ID to a URL is hard to imagine for most people. Email addresses are also unique, services that could map email addresses to IDs might do the trick.

Why not integrate identity transparency into authority rankings? A comments made by a tangible “real” person, vouched for by others, has more authority than someone posting psuedonymously, who has more authority than someone posting anonymously.
Someone who makes a statement and is willing to stand behind it should be incentivized.

What about Peter Parkes’ question about inames, in his post above? Looks to the non-specialist like this is an idea that’s already been worked-through at inames.net. No?

I hope to see this settled in an inter-operable way soon, because I want to see the “reputation economy” (like Cory Doctorow’s “Whuffie”) begin to catch hold, so that I can get a general opinion on the credibility, reliability and general legitimacy of a company, person, location etc. And if that’s what Whuffie Beta will look like, then the release version would also allow me to weight the accumulated score for something based on how well the ‘voters’ know their stuff, and whether they have similar values to mine. So if the entity is a Mexican restaurant, for instance, the score I use might give more weight to the opinions of Mexicans, and chefs, and other people who like spicy food, and less to the opinions of people who think highly of the food at McDonalds. OR if the entity is a person, I might give more weight to his/her neighbors, and less to people whom I dislike.

This will obviously work best once there are many, many people expressing their opinions of everything all of the time, in a way that feeds the tabulator mechanism instantly and automatically. Pieces of it are already “out there”, but we need this universal identifier before it can really begin to jell.

DVD-Cloner IV is software that makes high-quality 1:1 DVD backup copies with the ability to decode CSS encryptions and regional codes completely.

WEB HOSTING USA |Cheap Hosting Plan, Cheap Hosting Package, Managed hosting services at hostinghelpline.com

Choosing the right web hosting company is not an easy task but hostinghelpline.com has made this now so easy to choose the right cheap hosting package for your web solution. Hostinghelpline.com provides the consumer with an objective way to find the right hosting plan without any problem, and provides web hosting companies with a level playing field to differentiate themselves in a competitive market. Our directory provides user the best way to find cheap hosting plans with respect to all basic features including price, disk storage, mail accounts, data transfer and location. The directory offers a wide range of plans including latest technology such as Microsoft ASP.NET, PHP, Perl, Python, IIS6, SQL Server, Microsoft frontpage, and ODBC and control panel.
Explore http://www.hostinghelpline.com provides you articles covering all aspects of web hosting alongside reviews with web hosting players, a section of the site covering detail of web hosting providers with ratings and about their hosting triumphs.

Thanks

esaudi.info

web designing company