End Users

End Users are normal humans. I'm thinking in terms of users of desktops and laptops, but maybe that's too narrow.

Some of them will go out of their way to embrace non-monopoly solutions (e.g. Mozilla, Linux), for a variety of reasons. This will probably be ~5-10% of the population unless something dramatic happens. Most will muddle along with whatever appears easiest, writing off certain semi-catastrophes (spam, loss of data, credit card theft which results mainly in paperwork as vendor/issuer eat loss) as the unavoidable side-effects of computer use, regardless of whether this is accurate.

Most of them will end up with some basic personal data inside Passport, because Microsoft is requiring Passport usage for so many purposes.

If Passport helps them (a) avoid having to fill out lots of registration forms for Member Sites and (b) avoid having to remember IDs and passwords for those Member Sites, they will be happier. At least if that's free.

But will they pick one site over another based on saving registration time?

Or will Member Sites adopting Passport simply see fewer abandoned shopping carts? Why are there so many abandoned carts right now? What do users do instead? Go buy at another site? Go buy at a physical location (hard to consider that "easier")? Order by phone? Avoid the purchase altogether?

Ranked requirements for this group for a Distributed Membership System:

• avoid having to remember id/pw for every site (without depending on Web Cookie-s or always using same id/pw) (Single Sign-On)

• reduce hassle of re-typing registration data at every site (do current browsers handle this OK already? it's a browser-per-user model, but not bad... hmm, maybe MS will make this only work with Passport in the future)

• avoid getting confused by interface (at reference server, at each member site)

• avoid having to worry about keeping data current on every site (oops, they wanted to tell me about a problem but my phone number had changed).

• be able to change reference-server (service provider) with minimal hassle (and still be able to update records)

• Privacy

  • don't end up with my email distributed everywhere for spam

  • control/approve what fields a given site asks for from my profile ("why does this site need to know my credit card number?")

  • avoid grouping together of data from multiple sources (e.g. cross-site aggregation of my habits: I don't want my insurance site to know that I read a lot of AIDS articles on my healthcare site) (is that a technical issue or a site-usage-agreement (EULA) issue? the risks in a single global userID are big)

Do current browser features provide a chunk of these benefits? Is a lot of the value from this system delivered by browser features which (a) remember and let you select-via-pulldown values you've typed into HTML form fields of the same name, even on other servers, and (b) remember your id/pw for each individual server?

Some obvious shortcomings:

• doesn't solve problem of juggling array of ids/pws (if this is the only deal killer, then should we design a system to solve only this problem, for the sake of simplicity?)

• every browser has to support those features

• data is tied to browser, so hosed if (a) switch browsers or (b) use another machine (but this doesn't seem to apply (yet) to the vast majority of users)

• sites have to use "popular" field names (which I think they'd have to under any architecture we propose, anyway, right?)

• I think this feature is generally disabled for SSL forms, but I could be wrong.