What is lifelogging for?
When is anyone, the lifelogger or anyone else, going to access the pile of data? I noticed this in the article that XiXiDu linked:
The first person I met doing this was Ted Nelson in the mid-1980s who recorded every conversation he had, no matter where or what importance. To my knowledge his archives have never been revisted, even by him.
And I remember reading a long article about Ted Nelson and Xanadu which said the same.
I'm not signed up for cryonics, so it appears from my inaction that I don't take alpha simulations vitally seriously, and beta simulations would come way behind. If some future people want to LARP what they know of my existence, good luck to them, but I can't see it as something I have any interest in. No, the only reason I could have would be a practical purpose here and now: a resource for me to access, a prosthetic memory.
So indexing and searching are fundamental, yet they seem curiously neglected or even defective in the examples in that article. Recording everything against the day that the tools arrive is, to my mind, backwards. When will that day come? What can I actually use lifelogging for already, here and now?
In practice, I do the opposite. I do not keep a journal. On occasions when I've had a specific reason to write one, I have always thrown it out (shredded) after the reason has passed. My diary is for future appointments, not past memories. I keep no financial documents except according to current need: when my shoebox of bank and credit card statements fills up, I shred the oldest half.
At work I keep all emails except for mailing list stuff -- partly for legal reasons (I work at the same university where ClimateGate happened, although in an unrelated department), and sometimes I really do need to search through the archive for something. At home, no email survives in the files more than a few years.
I have hundreds of old photographs gathering dust that I am minded to either scan and dump, or just dump. Obsolete audio cassettes that I could transfer to the computer and dump, or just dump. (How often do I play them? Never.) What would you do?
Some uses:
There are probably more, these are the ones I could come up with right now.
Are you doing any of these right now, or are they hypothetical for when the technology gets there?
Hypothetical unless you count the IRC and IM logs, which I already employ for many of these purposes. To be exact, I'm saving funny conversation snippets to a separate file, recording past conversations, and reverse-engineering arguments. I've also done a bit of the saving good moments bit, though less than I could.
Are people in practice really so tolerant of those who want to record them constantly? Unless I'm absolutely forced to be in the same room with someone who does it (or, of course, if I specifically want something to be filmed), I would insist that one of us must leave, no matter what. I wouldn't even trust them that the damn thing is turned off when they say it is. (And if done secretly, I would consider it a voyeuristic offense against my person, effectively an act of war.)
I see taping lectures and other public events as an entirely normal thing. However, the idea that someone would want, or even tolerate, to be taped during private emotional moments and in situations where funny stories are told and passionate arguments made honestly baffles me. (With a few traditional exceptions like taping family events for sentimental purposes etc.)
You do it too, you record everything with your brain. Sure, right now it is hard to read-out and memories are still vague. But the time will come when we'll be able to download memories. And that memories are vague and sometimes counterfactual will be even worse because people will believe them based on the persons credence. Further you are effectively speaking out against transhumanism with this stand. You are going to hate all people with advanced memory enhancements? You are going to hate all people which possess brain implants?
Anyway, what are you going to do the day when reading-out memories will be easy?
Identifying natural images from human brain activity
A challenging goal in neuroscience is to be able to read out, or decode, mental content from brain activity. Recent functional magnetic resonance imaging (fMRI) studies have decoded orientation position and object category from activity in visual cortex. However, these studies typically used relatively simple stimuli (for example, gratings) or images drawn from fixed categories (for example, faces, houses), and decoding was based on previous measurements of brain activity evoked by those same stimuli or categories. To overcome these limitations, here we develop a decoding method based on quantitative receptive-field models that characterize the relationship between visual stimuli and fMRI activity in early visual areas. These models describe the tuning of individual voxels for space, orientation and spatial frequency, and are estimated directly from responses evoked by natural images. We show that these receptive-field models make it possible to identify, from a large set of completely novel natural images, which specific image was seen by an observer. Identification is not a mere consequence of the retinotopic organization of visual areas; simpler receptive-field models that describe only spatial tuning yield much poorer identification performance. Our results suggest that it may soon be possible to reconstruct a picture of a person's visual experience from measurements of brain activity alone.
However, the idea that someone would want, or even tolerate, to be taped during private emotional moments and in situations where funny stories are told and passionate arguments made honestly baffles me. (With a few traditional exceptions like taping family events for sentimental purposes etc.)
Status quo bias?
No. Merely the desire not to be forced to obsessively ponder my every word and act, for fear that it might be published on the internet tomorrow, or otherwise shown to some relevant authority figure who would be judgmental about it. I also want various mishaps and unpleasant events that happen to everyone from time to time to be resolved, overcome, and forgotten, not to be permanently recorded like sleeping demons.
Not everyone will find your funny stories funny, your honest opinions respectable, and your demeanor in various relaxed situations likable. (Or mine at least, and those of practically anyone I enjoy socializing with.) People are pretty damn judgmental, and unless your life is a complete bore and your opinions a paragon of exemplary conventionality, you will likely have some moments in your private life that you don't want at least some people to see. Also, some things in life should be complete and absolute bygones for the good of everyone involved.
A lot of people would probably dislike the idea of being recorded (which I think ought to be respected), though I suspect folks would get used to it. People already save emotional e-mails and logs of very private instant message conversations without others objecting, though obviously those have more deniability value. It's much easier to write an e-mail and claim somebody else wrote it, than it is to forge an audiovisual recording of them saying something.
In general, my attitude towards people saving logs or e-mails is that if I trust them enough to tell them whatever it is that I'm telling them, then I also trust them not to make the recordings public in a way I'd disapprove of. My attitude towards somebody lifelogging me would likely be similar.
"Why would anyone want to do this?
I can think of several reasons. Initially, it'll be edge cases. Police officers on duty: it'd be great to record everything they see, as evidence. Folks with early stage neurodegenerative conditions like Alzheimers: with voice tagging and some sophisticated searching, it's a memory prosthesis.
...
We may even end up being required to do this, by our employers or insurers — in many towns in the UK, it is impossible for shops to get insurance, a condition of doing business, without demonstrating that they have CCTV cameras in place. Having such a lifelog would certainly make things easier for teachers and social workers at risk of being maliciously accused by a student or client."
--Charles Stross, http://www.antipope.org/charlie/blog-static/2007/05/shaping_the_future.html
For uses, you could also read Gordon Bell's Total Recall. A little imagination suffices to think of uses for lifelogging (getting mugged, getting into accidents, arguments, etc.). LW itself is responsible for an interesting use - sending data to another person so they can see whether you are slacking off/giving into akrasia.
I would point out that no one has ever 'revisited' cyronics patients as well.
Total Recall is an excellent text on the subject. Plus, it's pretty easy to read and Bell does a good job elaborating on the usefulness of lifelogging.
Check out the Vicon Revue. I'm into Lifelogging for years, I just don't record continuously yet. Note that I do not own the Vicon Revue but it is the only product I'm aware of that can do the job right now (in terms of size etc.), although without audio/video functionality. I'm simply using a compact camera and my iPhone so far. I even have a stationary camera in my room that takes a photo every minute. Last year I reached over a million files. Right now I got over 2TB worth of personal videos/photos/screenshots/logs etc., not sure how many files there are now though.
I often fantasize that lifelogging might be a cheap alternative to Cryonics, in terms of beta level simulations as described in some of Alastair Reynolds’ novels:
A poor cousin to an alpha-Level simulation, a beta-level is based on modelling the behavioural patterns of the person copied, attempting to predict their reactions to a given stimulus.
If you got my DNA so that you can clone me it might even be possible to imprint memories and behavioural patterns based on the lifelogs. I suppose that even without a DNA sample, given sufficiently powerful AI, such a beta-level simulation might be sufficiently close so that only a powerful posthuman being could notice any difference compared to the original. At least that's a nice idea :-)
Resurrection without a backup. As with ecosystem reconstruction, such "resurrections" are in fact clever simulations. If the available information is sufficiently detailed, the individual and even close friends and associations are unable to tell the difference. However, transapient informants say that any being of the same toposophic level as the "resurrector" can see marks that the new being is not at all like the old one. "Resurrections" of historical figures, or of persons who lived at the fringes of Terragen civilization and were not well recorded, are of very uneven quality. [Orion's Arm - Encyclopedia Galactica - Limits of Transapient Power]
A link on the subject (there were others, but they are behind a pay-wall now and I'm too lazy to look for a backup):
My problem is that I regard the Revue (and the photo thingy Gordon Bell uses) as being hilariously awful. It isn't worth considering for a second. That anyone thinks that they can sell it is symptomatic of how terrible the market for lifelogging devices is.
The Vicon Revue costs 500 pounds, excluding VAT; or >$800.
And that gets you something far more useless than, say, the uCorder I linked which costs an order of magnitude less. Not half as much, or a third as much, but an entire order. For that much, I could just buy 5 or 6 uCorders, or buy 1 and hire a local college student to hook it up to an external battery, and still have a ton of money left over.
Nitpick: I'm not sure if that definition of a beta-level is canonically accurate; all I remember from Revelation Space is that that's how one specific beta was constructed. Also, there are unspecified technical reasons for believing that betas are nonsentient. I've been using the word "reconstruction" myself.
To give an example of how awful the Revue is, it is exceeded on a price/picture ratio by just about anything - dozens of hunting accessories or even a camera for recording your garden.
Not that the garden cam is all that bad; here's a nifty video/presentation employing it or something similar.
(Links courtesy of foucist.)
Has there been any progress on battery life? I'm rather surprised I can't even find a reliable source of audio-only recording that is as convenient and useful. I had a system with a recorder I ordered off of thinkgeek, but it was rather crappy and lost data. I've tried with my android, but it eats up battery life. There don't even seem to be any good phone conversation recorders, at least for the Nexus.
There is a project on kickstarter that seems to be quite a bit better than the looxcie, called Zion Eyes, but they only have ~3 hours of battery with the note that: 'Aftermarket plug in battery packs are available to extend the battery life.'
Has anyone messed with the fitbit? I know it isn't for video recording, but it sounds pretty good on paper. I've found that the withings scale is a damn good modern way to track weight without having to do your own spreadsheeting.
If there is a more recent conversation about this, I will happily go there. This is just the first thing to turn up material from this year on the subject via the google.
foucist has suggested that one quick route to a lifelogger is to modify the uCorder to have longer battery-life, since it is almost perfect in every other respect. The uCorder recharges through its USB port, so one could hook it up to a USB battery.
USB batteries are popular for things like iPods, so we benefit from economies of scale; it seems possible to boost recording time to upwards of 10 hours with a USB battery like this $42 one.
Unfortunately, foucist also found that the uCorder manual says it cannot record while charging, and presumably wouldn't automatically draw on the battery anyway. So this wouldn't work (without a lot of manual intervention, and the uCorder takes hours to recharge...)
But the basic idea is sound - we just need to find a nice recorder which will run off a battery pack.
What about the PocketPro II? It draws 240 mA, so a 1 Ah external battery gets you 4 extra hours.
Wouldn't it be possible simply to carry 2 uCorders, and swap then every 2 hours?
Putting the one that is not in use onto the USB battery.
Sure, but then we are well past trivial inconvenience and into real inconvenience.
I mean, if you don't mind the constant switching, you could just buy six of these at around ~$120 and switch every 2 hours!
(Significantly cheaper too: 120 vs 180+USB battery.)
I've been doing audio-only with a $40 dictator from Wal-mart that fits in my pocket. It averages 150-200 MB a day. I generate hashes of each file and timestamp them so they're more likely to be useful if I ever need them for proof of something.
The thing that prompted me to start doing this was frequent arguments with close ones that often got down to "you said this", "no I didn't" type of stuff. It's oddly very assuring to have this recording. (FTR, I used it for that purpose more or less once. Although I find it useful for recording therapy sessions too.)
Combine this with speech-to-text transcription software and you get a searchable archive of your recorded interactions!
ETA: In theory. In practice, dictation software algorithms are probably not up to the task of turning noisy speech from different people into text with any reasonable accuracy.
It's too bad the Looxcie isn't very good, but it's good to see the idea that lifelogging has benefits is starting to go a bit mainstream - the NYT article doesn't include any of the usual inane criticism.
New Scientist published a life blogging article with the Vicon Revue today: http://www.newscientist.com/article/mg21328532.700-total-recall-diary-of-a-lifelogger.html?full=true
Engadget published a Vicon Revue review as well: http://www.engadget.com/2010/06/25/two-days-at-e3-with-a-vicon-revue-life-blogging-camera-video/
Note I am not a Vicon employee, just a potential customer. Has anyone here purchased a Vicon Revue? Current price is 299 English pounds.
Can't read the NS review, but I certainly agree with the Engadget conclusion:
I'm sure with a little searching some far cheaper alternatives could be found... even though life-blogging today is decidedly less trendy than it was six years ago.
foucist recently pointed out the Epic action video camera which is nice & small, but its Achilles's heel seems to be battery life - 2 AAA batteries won't take you very far.
I also recently ran into a description of the 'Cylon Body Worn System' which from the listed specs and mentions elsewhere, seems like the perfect lifelogging system - first sold in 2007 - but not available publicly. A more recent model does 720p HD, and the docs claim that its battery lasts 6 hours 40 minutes.
In other news, Flash continues its precipitous price drop. Now $40 will get you ~32GB of SD memory, twice when I wrote this post back in October.
Further foucist hits:
(We think that there must be a cheaper version of the second one somewhere, minus the silly baseball hat, but haven't found it.)
The old idea of lifelogging seems to be a reality now. It has the potential to be quite useful, and not just in distant contrived scenarios like cryonics or being recreated by an AI.
One of the classic objections was that we couldn’t afford to store the many gigabytes - possibly hundreds of gigabytes a year! - such a practice would generate, but right now you can buy 1 terabyte for <$50. And there’s no end in sight to whatever Moore’s law has been governing hard-drives over the past decade or two.
But how is one to record it? That seems to be the rub. All the storage space we could want, all sorts of new formats like WebM or Dirac or x264 to store the videos in - but what camera generates the data in the first place?
We don’t care about sleep time, so we don’t need any more than 16 hours or so of recording a day. We can probably get away with 12. Even 8 might be enough (to record yourself on the job - or off). An encoded compressed video might be 1 megabyte a minute or 60 megabytes an hour, but let’s be generous and assume 15x worse than that, or about 1 gigabyte an hour. So perhaps 16 gigabytes.
16 gigabytes of Flash costs $40 or less. So that’s not an issue either.
And presumably optics and microprocessors are very cheap given the incredible popularity of web cameras, digital cameras, digital camcorders and whatnot over the last decade.
But for all that, I can’t seem to find a mini-camcorder which will record even 8 hours and be a useful lifelogger!
Am I wrong? Are there existing products? It seems to me that it ought to be perfectly possible to take something like the uCorder, slap in $110 of batteries, and get it up to 8 or 12 hours’ life. But I have yet to find such a thing.