Dying for mail

[Dave Birch] I found the South-by-Southwest (SXSW) interactive sessions that I went to, without exception, first class. It may be because of the spectrum of people that they attract, or it may just be something in the Austin air, but I got caught up in a number of exceptionally stimulating discussions, all of which gave me new things to think about. Here's an example: I signed up for a session on Digital Wills run by Corvida Raven from she-geeks. She ran an outstanding session, and I can't resist blogging around it despite the morbid tone of some of the discussions that resulted from it! First of all, let me say that this is an aspect of the online world that I have been interested in for some time. I wrote a piece about it for The Guardian way back in 2004, reflecting on the fact that I had been making a will and had gone and got a booklet about it (I think from the bank, but I can't remember) and I was remarking that it didn't seem to cover my data.

It wasn't mentioned in the booklet of sample will elements I was using. That covered topics such as houses and kids, but it should have had additional specimen clauses along these lines: "I leave the 100GB external Firewire drive containing all of my emails and the back-ups of all of my personal documents, my iPhoto library and my iTunes to my wife. This volume was encrypted by Mac OS X using AES-128 and the password is the name of the band we saw together on our first date followed by the age of our first female cat when she died."

This may seem silly, but could become a serious problem in the future. My wife will need my username and password for Barclays, BT, British Airways and our family blog - and there was nothing about that in the booklet, either.

[From Second sight, Dave Birch | Technology | The Guardian]

This isn't a sophisticated enough solution, of course. What we really need, as a society, is proper security and privacy technology and we are an awfully long way from seeing this introduced at all, let alone introduced into probate law or custom and practice. Nothing much has changed since my article, as Cory Doctorow reinforced last year.

What I found surprising all through this process was the lack of any kind of standard process for managing key escrow as part of estate planning.

[From Tales from the encrypt: the secrets of data protection | Technology | guardian.co.uk]

There are clearly some business opportunities here, and not only for lawyers! Some organisations have already decided to take the digital afterlife seriously.

Facebook may not have been the first to create a specialized policy for deceased users, but it was one of the highest profile because of the way it handled the issue. Instead of merely agreeing to let a family member take control of the account, the company instead decided to take things a step further and let people turn someone's account into a memorial.

[From Death and social media: what happens to your life online?]

This is nice, but it seems to be still fairly rare. Take e-mail as a fairly standard requirement. If you die, Yahoo will delete your e-mail. But I may not want my e-mail to be deleted. Can I ask Yahoo not to delete my e-mail? No. But hold on, how do they know I am dead? If I just give my Yahoo password to my wife, then presumably she can carry on using it or archive the messages or even delete them. But what I if leave her the password and tell her not to delete them but just to save them for posterity and not read them? This is all getting a bit complicated.

identity, management, social networking

Naturally, I think a better model for digital identity and a working digital identity infrastructure provides a way to deal with these issues. By partitioning your online life across a number of virtual identities and providing a means for others to take over these virtual identities, then the virtual identities are just another kind of "property" and can be dealt with by the law. Then, if there are some virtual identities that I want to live on, some to go to limbo and some to die, I can sort of see how this might be achieved by giving control of the various virtual identities to different people. I suggested this during an excellent session of the digital afterlife run by the guys from the Digital Beyond blog, but then there was another discussion about another aspect of the problem: the relationship between identity and authenticity, which I think is opening up yet more thinking.

One final point. If you have given control of the password (or whatever) for a digital identity (in my model) to someone else, then that digital identity isn't really dead, is it? What's more, the virtual identities that are bound to that digital identity are surely indistinguishable from being alive -- so in that case, even the very simple problem of trying to find out whether someone is alive or dead will become very difficult. So I'm amending my prediction that IS_A_PERSON will be the most valuable attribute of a virtual identity to add the prediction that IS_ALIVE will be just as important: after all, no-one wants to send spam to dead people, do they?

These opinions are my own (I think) and are presented solely in my capacity as an interested member of the general public [posted with ecto]