A Personal Panopticon

One of the recurrent ideas that surfaces in science fiction and in the predictions of futurists is an always-on record of a person’s life. Typically, it’s presented as the stuff of a dystopian nightmare future, where the record is falsified and you end up in jail for a crime that your wayward clone committed. But what if such a record could be created and it were useful?

The technology is damn close to being practical. Simply carry slightly fewer MP3s on your iPod and you’ve got more than enough recording space for your entire day to be recorded in voice-quality mono audio.

Why would you want such a thing? That’s always the first question that arises. My own desire for this device dates back to my late teens, when I used to make really bad music with my friends. My constant frustration was that I didn’t have audio records of all the found-sound elements that I stumbled across in the course of a day. Snippets of conversation, ambient sounds, messages left on voice mail, obscenities shouted at me by strangers: these were all elements that I was determined to appropriate.

Later, as I realized the boundaries of my talents, (somewhere between a lump of coal and Ja Rule, which is to say, nonexistent) I started to realize that perhaps this audio record would be valuable on its own.

Having a complete audio record of my day, indexed by time, would allow me to refer back to a moment in conversation when quoting it later, or when I needed clarification on a specific point brought up while talking to someone. The timestamps alone, with perhaps quarter-hour index markers annotating a day’s recording, would offer chronological indexing of the audio diary in a way that parallels permalinks in a weblog. Links to specific points in time turn out to be a very powerful way to manage microcontent.

The next level up would be an awareness of presence being linked into that audio stream. I suppose the super-futuristic version would be some FOAF-enabled scanning device that recognizes each of my friends’ odors. It’d be worth it just for having alleviated that particular burden on my scarred olfactory memory.

More realistically, though, the immediate implementation could be made presence-aware by assigning stamps in your audio stream referencing the people invited to meetings on your calendar application, or by simply allowing you to pick people off of your contact list and associating them with the current point in the audio stream.

I’ve already recorded snippets of audio onto the PocketPC that plays my MP3s for me today, and that device carries a fully-synched, updated copy of my contacts on it. I’ve also recorded audio notes on my laptop using OneNote, and I could assign those notes to a contact’s information in Outlook. But I’d much prefer an iPod-type device that I could just tap on while it was recording, scrolling to any given name in my address book and just clicking on it to select my companions at any given moment.

Now we’ve got an audio record that’s searchable by time or by person. And it’s aware of who those people are, and what level of trust that I’ve assigned to them either via relationships defined in FOAF or by using categories I’ve assigned in my contact management application.

This is where the audio record starts to get valuable. I can set my playback preferences so that this tagged audio is automatically available to the people who were marked as present at any given time. When those people access the audio stream, permission would be granted to listen to the dialogue without any significant barriers to their access.

From there, the software could identify simple, useful relationships in order to control access permissions for other people. An audio range tagged as taking place during a business meeting that’s in my calendar would be accessible without passwords to anyone who’s marked as a coworker of mine, but I’d receive a notification that they had done so when they clicked on the audio file. Move down one level of trust, and employees of the company I was meeting with would be able to access the audio, but only after submitting a request for me to authorize them. All the way down at the lowest trust level, complete strangers, the entire clip would be off limits to them, including obscuring of the fact that an audio record exists at all.

How do all these people get to this data, though? I’m not going to pass around my super-iPod to everybody I meet. Well, these individually-tagged clips of audio constitute some very valuable microcontent, and we’ve finally got a set of decently usable tools for publicly (or semi-publicly) publishing microcontent. Those tools are what we use right now to publish weblogs.

The necessary evolution is for personal publishing tools to start to allow far more granular control over permissions for reading the content that they generate. This doesn’t just apply to the audio we’ve recorded with our always-on iPods, but to the text we’re publishing to our weblogs as well. It should also encompass the new text that we’ll feel comfortable adding to our weblogs once we know that we can control access to our sites with at least the level of control that LiveJournal users take for granted right now.

Weblog publishing tools will grow to allow us each to create a personal panopticon, at least to the degree that we’re comfortable doing so. We’ve already seen that security through obscurity is untenable, and I’ve discussed how privacy through obscurity is going to be just as impossible to maintain in the future. So given that the key to maintaining privacy is going to be by choosing which parts of one’s identity to expose, it’s time that our personal publishing tools start to allow us to manage all of our information, using simple logic to determine which people are permitted to access which subsets of that data. Now that even Microsoft has embraced reasonable default settings in its software, it might even be possible that we won’t see too many people accidentally making their darkest secrets world-readable through their weblog. We may still see the people who choose to do so on purpose.

Where does all this go, then? I’m hoping to eventually have a video camera with audio in roughly the form factor of today’s Exilim. Another evolution will be the integration of personal publishing APIs into devices themselves. Instead of worrying about storage capacity on an audio or video recording device, manufacturers should be treating the memory cards in a camera or audio player as a local cache, emptying the cache to the central publishing repository for that user every time the device comes in range of a network connection, preferably a wireless one.

My PDA or tablet would let me take notes, and those notes would be upstreamed to my blog, marked as private by default, and then flushed from the immediate local storage as soon as it got full. A digital camera that performed those same functions would solve many of the issues with synchronizing and downloading individual pictures right now. There has been talk of individuals having data storage that lives "in the cloud" for years now, but the focus has historically been on security and implementation. The more important issue is that this storage cloud have an easy enough interface that publishing to it is almost invisible and nearly effortless. That simplicity of publishing is what blogs are all about, and the tools will allow new types of integration.

Today, we’ve already got bloggerazzi that appear at every gathering of web geeks. It seems that we’ve already accepted a future where we’re all celebrities. If we’re going to accept the negative implications of that reality, then we’d better get working on creating some positive implications to go along with it. The personal panopticon is one of the positives.