Canonical URLS, Network Effects, and the Digital Me

Jon Udell has a typically thoughtful post where he segues from the topic of canonical URLs to that of the network effects that flow from defragmenting fragmented online conversations. It is worth reading, and something about which I feel strongly.

To build on one of Jon’s examples, why must I post book reviews at Amazon’s site? Sure, it’s good for Amazon, but it isn’t all that great for me. Among other things, it further fragments the “digital me”, putting bits and pieces of my online persona in yet another place online, one not easily found by people curious (for whatever reason) about the various places where I have contributed.

Better, for example, would be for me to put reviews and such on a site of mine, but tag them as reviews. Amazon (and others) could grab them and aggregate them, adding value and, perhaps, compensating me accordingly (in much the same way it already does with its Amazon Associates program).

And while I busy tagging content I’m also implicitly tagging identity. Because managing identity is becoming a large and frustrating issue, as I am reminded every time I send an email. Who am I right now? Am I responding to this person from my university address? My personal address? My venture capital address? Who am I right now? Managing communications fragmentation is a bitch.

At root, how do I best I tie my myriad activities and online persona to a canonical digital me? Or, at least as important, how do I properly protect my privacy via an online separation, where necessary, between all those activities?


  1. Why this is a bad idea: your outbound link from Amazon could point to child pornography, a phishing scam, or other unwanted content at any point in the future. How would they know? No content site is ever going to host arbitrary outbound links…not unless they like the idea of being a redirection service for suspicious content. Even if they made some content-assurance system like an MD5 hash of reviewed content, the outbound link could simply redirect crawlers originating from Amazon’s IP range to the “approved” content and everyone else to….whatever.

  2. Yup, absolutely true, which is another reason why robust identity management would be nice. In the interim, of course, we are stuck with the nutty current world of scribbling comments all over the place as a proxy for actually being able to prove that I’m me.