Format
Author / PGP Key: The full text of the author's PGP key.
Author / PGP Key URL: URL to a page containing the author's PGP key, delimited by the usual tags.
Content / Detached signature: A detached PGP signature for the content item's data.
Overview
This is for uniquely identifying blog entry authors
-
As with anything to do with identification, an author should have a canonical URI
-
Inclusion of aliases (e.g., nicknames, pseudonyms)?
-
This would help to automatically thread discussions. This also gives the log owner a range of control options: allow posts from a fixed group, exclude certain people, only allow signed posts
-
This will also enable experiments with broad reputation systems.
Other extensions to author
-
author metadata: institution, home URI, full name
-
is this its own module? see stuff like Foaf
-
Contributors - People besides the one Author of the entry.
These are currently lumped into BiblioGraphy.
Discussion:
[JoeMadia] Question: Should a single Identity map to a single single human person or should it also be allowed to represent groups or machines? I would prefer that identity be tied to a single person but it seems that some thought should be put into automated feeds (error logs, agent-like notifications, etc) as well.
-
[MarkPilgrim] On the author page it seems to have been decided that an entry should have exactly one author, and that an author can be a system. If you have more than one person involved, the primary author is identified as the author and others are identified as contributors. This implies (to me) that each person is uniquely identified, not the group.
[JoeMadia, RefactorOk] Data elements for a single person Identity: +1 on Canonical Uri and nickname/handle. I would like to add email (optional) and primary web page (optional). What about support for stronger authentication (Public key, pingback mechanism, etc) to help avoid Identity theft? Obviously this is not a problem today but it could be trouble in the future.
-
[MarkPilgrim] +1 on all that. Email address *must* be optional. Sucks to be BillKearney (and others at Syndic8) who need to track down people to contact them, but there's just too much abuse.
The simplest approach would be for the author to maintain a web page with identification (which may be merely a pseudonym) and a public key. This could be a weblog, a personal web page, or a page provided say by a weblog service. Any entry can be guaranteed to come from that "person". With that foundation, add features to distinguish between real persons, bots, PR agents, ....
-
[GregElin] - Agree with author maintaining a web page. One way of avoiding "identity" theft is to look at it issue as one of attribution. Could I, as an author, simply indicate on the system that I contorl, that "Yes, I did make that attribution." that way a reader cold follow a URI to the author (and/or check an objective source for the author's URI) and see the author does indeed take credit for the article.
-
[JimDriscoll, RefactorOk] One of the founding assumptions of PGP was that there would be a WebOfTrust - ie, that you would be able to tell how far to trust a signature by how far the people you know trust it - normally by people countersigning each others keys. Would it be feasible to, for example, use TrackBack to help affirm an identity signature - ie, to allow a remote blogger to assert their confidence that the key is held by the person it refers to, or that the person is who they claim to be?
(What is the prior art for simple distributed identity systems? Less grand than the "Liberty Alliance"(?) )
The network of relations among authors will be rich like that among weblog entries.
Projects:
-
FOAF
-
Foaf has an mbox_sha1sum element, basically a one-way hash of an email address. Impossible to reverse engineer (spam-proof) but uniquely identifies an email address.
-
"In FOAF we use digital signatures to attach email addresses to documents (strictly, PGP key IDs, but the email address is the common currency of identity in FOAF.) It's the bottom layer of a system of trust -- you still have to figure out yourself whether you trust the keyowner or not -- but at least it enables a link between a document's state at a particular point in time with an identifying key." EddDumbill
-
Augmented Social Networks builds on the identity issue. Jack Park
[JimDriscoll, DeleteOk] I think this should generally refer to OpenPGP (the protocol) not to PGP (the product). I like GnuPG myself.
[NickChalko RefactorOk] Why only sign the "content". I think many of the attributes are also important enough to sign. Title,Date,Author, basically everything except the sigs themself. EntrySigning
[MichaelManley RefactorOk] Can the managing editor or some other authority sign the feed as a whole? If so, that could open up possibilities for mirroring of feeds without fear of the feeds themselves being compromised. On the initial subscription to a feed, the aggregator would connect to the feed originator and get the public key of the keypair used to sign the feed. The aggregator could pick up the feed from any mirror (or other distribution mechanism) and be reasonably assured that the feed had not been tampered with since the original publication by verifying the signature. Mirroring feeds with authentication, alongside whatever caching mechanism the transport provides, could mitigate bandwidth concerns for popular feeds (thinking of feeds distributed via bittorrent, for example). Also, should pointers to public keys be made part of the AutoDiscovery mechanism?
See also Foaf, Security, EntrySigning, EntryAccountability, CommentAuthentication