![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Lifelogs and their implications
May. 16th, 2007 09:49 amI had a few thoughts about this posting by Charlie Stross:
I'm dubious about some of the general projections he makes about lifelogging -- because I suspect that long before it became ubiquitous, there would be such a massive number of reasons for editing one's lifelog that the information would be unreliable from a pure data-gathering/chronicling point of view.
Quite aside from the fact that most people will turn their recordings on and off, or at least restrict inputs at all sorts of times -- when going to the WC, say -- people will want to "shape" their own lifelogs. Edit the embarrassing bits out. Make "art" out of raw data by adjusting material in any number of ways. Present an "enhanced" image to others, especially if they're keeping logs which are going to be "published".
Plus there will be people who aim at maliciously hacking other people's lifelogs, which further reduces reliability.
By the time they get to be generally adopted, let alone universally adopted, they'd stop being things on which you could rely. Sufficiently close analysis might indicate where something had been edited, but couldn't replace what had been cut or overwritten. And without reliability, the surveillance benefits would be pretty minimal.
Similarly, although there would be more data available for making "history" -- the journalling equivalent of photo albums, probably, about 90% of the time -- it would be so heavily "shaped" that the results of relying on it would be uneven and unpredictable.
I'm dubious about some of the general projections he makes about lifelogging -- because I suspect that long before it became ubiquitous, there would be such a massive number of reasons for editing one's lifelog that the information would be unreliable from a pure data-gathering/chronicling point of view.
Quite aside from the fact that most people will turn their recordings on and off, or at least restrict inputs at all sorts of times -- when going to the WC, say -- people will want to "shape" their own lifelogs. Edit the embarrassing bits out. Make "art" out of raw data by adjusting material in any number of ways. Present an "enhanced" image to others, especially if they're keeping logs which are going to be "published".
Plus there will be people who aim at maliciously hacking other people's lifelogs, which further reduces reliability.
By the time they get to be generally adopted, let alone universally adopted, they'd stop being things on which you could rely. Sufficiently close analysis might indicate where something had been edited, but couldn't replace what had been cut or overwritten. And without reliability, the surveillance benefits would be pretty minimal.
Similarly, although there would be more data available for making "history" -- the journalling equivalent of photo albums, probably, about 90% of the time -- it would be so heavily "shaped" that the results of relying on it would be uneven and unpredictable.
