Stream of Consciousness

Given the sorts of “reputation” mechanisms (Technorati, etc.) that are springing up in the blogging community, it seems that it would be valuable to have the capability for blogs to be signed by a non-spoofable entity (person or pseudonym). Does this capability already exist? Does RSS have tags for (e.g.) a PGP signature guaranteeing content and authorship? Even beyond, perhaps blog posts could be encrypted so only authorized entities could read them.

If these features are not available, are they perhaps being included in future developments, e.g. Sam Ruby’s Echo proposal? Or is there some reason I am not aware of that makes this useless in the blogging world?

As long as I’m at it (and given that I’m working on a light weight ‘nym server for the Identity Commons)…

Another thing that will be of use in a fully open and secure blogging world might include the ability to attach a comment to a blog entry under a spawned pseudonym to protect one’s identity (this might require a blog-nym server with a set of mixmaster-like mailing capabilities). This new blog-nym could remain anonymous or itself gain reputation by making other comments and/or having comments attached to it. If at some time the owner of this blog-nym wished to claim ownership of it, s/he would be able to. (On the other hand, if s/he wished to disconnect from it, that would be possible, too, and the blog-nym would perhaps get reaped after a time of inactivity by an automatic process on the blog-nym server.)

Update: 6/30/2003: I floated my question to a mailing list I belong to, and received a negative response.

I am surprised. As we move towards an Identity Commons, one’s reputation will become the currency (ala “whuffie” in Cory’s DAOITMK). PR people feel this today, because it affects their bottom line. But as it is impossible to indefinitely trust anyone with your personal (profile) information, pseudonyms will become the common way to float pieces of information to determine their worth, as well as to build trusted readerships while maintaining anonymity (ala Publius, the author of the Federalist Papers).

I would have thought that the bloggers would be all over this. The security and privacy capabilities of SAML, available to individuals and their nyms (as opposed to simply between e.g. Liberty servers) is what the next step of communications and publishing protocols ought to provide.

First and foremost, there must be a major immediate benefit for the user. Persistent online identity - the long-elusive holy grail of “single sign-on” - offers one such, both for the consumer-oriented and the civic-minded. But how do we get mass acceptance of such a new technology among the civic-minded when it’s the consumer that such services are being built for?

One path is to help the average user with managing the most valuable resource they have: their attention. One way to do this is by creating a personalized news and information service that would provide:

• Less overall quantity of information
• Higher quality (more relevant) information
• From better (trusted) sources

The first online personalized news system was called ‘NewsPeek’, a name I chose as while it provided a peek at the news, there was a clear danger (with a tip-of the hat to Orwell) that he who owned the profiles would wield the power of Big Brother (shades of Microsoft’s Passport). Security issues aside (for the moment) here’s a brief list of some of NewsPeek’s benefits:

• Easy to find:
  • Sources (people, texts, articles, via links or web-of-trust)
  • More (or less) about the current subject
  • Supporting (or opposing) arguments
  • Majority (global or among my community) opinion
• Easy to update:
  • Personal profile (show me more/less like this)
  • Filter unwanted (gather desired) information
  • Trust (dis-trust) of sources (=> reputations)
• Easy and safe to share & collaborate:
  • Privacy preferences for sub-profiles
  • Anonymous/Pseudonymous persona/nym creation

Alas, while NewsPeek is connected to live newsfeeds, it is a demo and the actual implementation of these capabilities falls far short of what we need.

My friend Mitch Ratcliffe sent me an email today asking if I could help create some technology for the Dean campaign. There’s a lot to do - a lot of great opportunities for coders looking to create effective community action.

It’s all volunteer for now. For my part, I’m working on an identity framework for connecting social/community networks (simpler than Liberty/SourceID) borne from the Planetwork Collaboratory that my current organization CPV and indeed all social/progressive communities/organizations need. I’m dedicating my efforts on this, at least until I find paying work.

Tasked with designing a simple identity framework for the Augmented Social Network, a good place to start is a shared basic vocabulary.

ENTITY

any object (person, thing, profile, reputation, etc.)

IDENTITY

a unique and persistent (within a community) name for an entity

PROFILE OBJECT

• (optionally signed) information about an identity (tech speak: P := [[Abstract_URI, Key, Value] Sig]) note that Key and Value can be defined within a (community) namespace
• profile objects are entities and can be named (tech speak: profile objects are first class objects)
• profile objects can have access rights attached to them

PROFILE

a collection of profile objects created by, or about, an identity

REPUTATION

a usually domain (key) specific (and nearly always fuzzy) calculation of an identity’s reputation or standing (usually within a community)

CERTIFICATION

a signed authentication by a trusted (reputable) third party

Thus, these all may make up part of my profile:

Reputations are often calculated within a domain such that transitive trust issues can be more easily resolved, e.g., a process calculating Owen’s reputation as a networker may tap into Fen’s profile during its lifetime.

First step toward direct Emacs editing of blog entries

OK - so I finally installed MovableType (v 2.64) and I’m going to see if this format works for me.