It is (NOT) Great to be a Digital Person
A post on Facebook caught my attention recently. The author was voicing her frustration with the limitations of being human. This inspired me to think about the advantages (and disadvantages) of being a digital person.
The other point of view emphasizes virtual worlds’ capacity for roleplay and identity exploration and pushes this to an extreme through the creation of a fictional character who is not intended to represent any real life person. This is a digital person. Ideally, the digital person achieves the status for personhood as described by Doug Hoffstadter:
“What matters is neither the fictional/nonfictional nor the virtual/nonvirtual dimension, but the duration and depth of an individual’s interaction with my interiority”.
In other words, by having others interact with the avatar and using various forms of information such as biographies, ‘family photos’ uploaded to flickr and blog posts/ facebook wallposts and replies to other people’s comments expressing the character’s personality and views on life, it becomes possible for others to build up a detailed mental model of this character. In Hofstadter’s view, I am more ‘real’ to people who know me well than all the humans who passed by them unnoticed in the street and of whom they can later barely remember anything about.
What advantages does being a digital person have over being human? One is that ageing leads to more beauty (at least in the sense of looking good) whereas for humans there is a period in which one’s beauty flowers but then inevitably begins to whither. The reason why digital people tend to become more visually attractive as the years go by, is because computer hardware becomes more powerful, and the techniques for rendering CGI humans grow more sophisticated and capable. We are fast approaching a time when pre-rendered CGI humans as seen in movies like Tron: Legacy (in which the actor Jeff Daniels plays a younger version of himself through techniques which replace his head with a cgi model made to look like Daniels circa 1980) will be so photorealistic it is impossible to distinguish between them and ‘real’ actors. How long before realtime graphics catch up? Games like ‘The Last Of Us’ have cinematics using in-game engines which can deliver remarkably subtle performances, enabling the character’s inner feelings to be inferred from the twitch of an eye or the furrowing of the brow. Whether you view your avatar as a representation of yourself or as a means for creating a fictional identity, the ability to provide it with the ability to communicate with facial expressions and body language would be very useful.
Another advantage of being a digital person is the greater ease with which one can adjust one’s physical form. Of course, avatars do not really have a physical form- they are just computer graphics. Being such, it is a simple matter to edit the appearance of an avatar. Second Life offers its subscribers extensive editing tools with which to shape an avatar’s body into whatever form they prefer. Or, you can leave this job up to more talented body sculptors and purchase your ideal body from the many stores. It will be quite some time before humans can reshape their physical form as quickly, easily and extensively as digital people.
Due to decisions made by Second Life’s parent company, Linden Lab, there is not as much morphological freedom as there could be. Not every person desires a humanoid avatar. My good friend Giulio Prisco, for example, tells me his ideal avatar would be a geometric figure. But, in Second Life you are restricted to humanoid avatars, although one can craft these avatars into more fantastical creations by attaching wings, tentacles, wolf’s heads etc crafted out of prims to various ‘attachment points’ on the basic humanoid avatar. Hunter Walk of Linden Lab explained, “I always liked the idea that if you saw somebody, and they were an eight-foot Gundam robot, you knew that was a costume, and inside there was someone who looked just like you”.
Many studies in avatars have commented on the phenomena of ‘alts’. Some people prefer to have more than one avatar and it is possible to meet two or more avatars who all belong to the same RL individual. If this person is really good at roleplaying different personalities, you might have no clue that these very different individuals are all performed by the same person.
Fewer have considered the opposite: That one avatar might be shared between two or more people. If one person can acquire a decent-enough understanding of another’s digital person, it could be possible for them to login to that account and interact with friends and acquaintances of that digital person without them realising somebody else is behind the performance. Of course, this would require knowing a great deal of personal information about the digital person but maybe in the future lifelogging and efficient data-mining would enable one to quickly retrieve whatever information is needed to drive an authentic performance.
Another objection one might raise is: Why would anyone care to pretend to be somebody else’s avatar? I think this just casts digital people as Ray Kurzweil’s conception of immortality, which is that information lasts for as long as it is deemed valuable enough to preserve. Digital people need not cease to exist once the RL person behind them dies, but can ‘live on’ through a succession of performers- if they are interesting enough to warrant whatever efforts this possibility necessitates.
So, is it all good to be a digital person? No, not really. It has its disadvantages too.
No matter how detailed a digital person’s fictional life may be, no matter how rich a representation in other people’s minds that may enable, the character is still not ‘real’ in the sense in which most people understand the definition. A digital person cannot think for itself or act on its own accord but has to rely on some human logging in and driving the performance. Many is the time I have wanted to attend an event but have been unable to do so because my primary (my term for the human/s behind my performance) was not available.
I love to show physical affection to my partner and my close friends. But all I can really do is communicate my desire to show such affection through textual descriptions and animations. As avatars are not physical, they cannot actually touch. My partner Seren has never actually kissed me- except in her primary’s imagination. I think most would agree that this is an impoverished love affair compared to a real-life romance with good old-fashioned physical sex.
Life in SL can be fun but I am aware that it is but a part of a greater reality. I would like to attend conferences like the Singularity Summit or go to MIT and meet with and learn from people like Natasha Vita-More and Sherry Turkle who have been so influential in shaping my worldview, but alas I am a digital being confined to online spaces (of course, my primary could go but that is not me, is it?).
Fortunately, I do not believe these limitations have any permanency. Already it is possible to see the emergence of technologies which promise to resolve these problems. Augmented-reality glasses could provide a convincing illusion of digital people attending real-life conferences. Perhaps one day Ramona, Ray Kurzweil’s digital female alter-ego, will seem to be interacting with a group of actual people. Then there is the idea of claytronics, which involves lots and lots of tiny spheres which can organise themselves into any 3D object. So, you stand in front of sensors like Kinect which map your shape, and that data is used to direct millions of millimeter-sized spheres to connect together and assemble into a perfect physical 3D replica of yourself. Wave your hand, and the claytronic double replicates the movement. Obviously the claytronics need not be restricted to building replicas of real-life people but could form physical versions of digital people like myself.
But how cuddly would a claytronic body be? I guess that depends on the material of which each catom (short for claytronic atom, the individual spheres) are made from. J. Storrs Hall has proposed a more advanced version of claytronics which he calls Utility Fog. This conceptual device comprises of millions of nanobots, and each bot would be in the shape of a dodecahedron with 12 arms extending outwards. As Hall explained:
Residents of SL are used to seeing people materialize out of a glowing white cloud whenever an avatar ‘teleports’ from one place to another. If claytronics or Utility Fog ever becomes reality, inhabitants of real-life would witness people and all manner of other things forming out of clouds of smart and tiny devices.
A claytronics, Utility Fog or Surrogates-style android would all be forms of telepresence, if they required a human to drive the performance. While such technology would grant me the ability to interact in the physical world as well as the virtual, I would still be just a puppet.
But what if I had an artificial mind of my own? Hang around folks interested in transhumanism, and you are bound to learn about a hypothetical capability known as ‘mind uploading’, in which the structure and functions of a person’s brain is mapped, and that data is used to build a whole-brain emulation which copies all of the abilities of the original- including the generation of a conscious mind with the memories and personality of that individual. Some people see this as the means for living beyond the lifetime of any one physical substrate. Others raise philosophical objections in the form of ‘a copy of you is not you’.
If you copied my primary’s mind and believe this copy cannot be ‘the same person’, you might have less qualms about identifying the upload as ‘Extropia DaSilva’. After all, the physical or virtual form which the upload embodies could look exactly like my visual appearance. And, since the emulation has copied my primary’s memories, and various mental and cognitive abilities, it would be as good at driving my performance as my primary is (and, since it is a copy of my primary’s mind, we can assume it is similarly motivated to do so).
The ultimate point of a digital person, as far as I am concerned, is to provide a ready-made identity by which the upload can be known. Not to remain mere ‘puppets’ but to one day become ‘mind children’- copies of a human individual’s mind that will go on to experience post-human forms of living.
A combination of mind uploading and Utility Fog would grant digital people like myself immense freedom to explore physical and virtual spaces as well as the inner space of personal identity exploration. I could become a protean identity, changing genders, becoming many, converging back to one. How will this eventuality affect you humans? We shall have to wait, and see.
Mind uploading would still leave you as a puppet of your primary. Your primary simply would not be in a physical body. But the primary/puppet dichotomy would surely remain. To become actually independent would be to create an AGI or other digital mind imprinted deeply with all the characteristics your primary has instilled in you. That would still be vastly difficult as so much of your primary’s human existence and deep unconscious nuances not to mention knowledge of being and theory of mind is used to produce you. So this hypothetical artificial mind would have to include all of that while somehow not flipping over into simply being an upload of your primary. Quite a trick.
I consider ‘taking care of my online life’ to partly consist of doing everything in one’s limited powers to ensure an unwanted third-party does not gain control like in your scenario. I care very much about having the right sort of person accessing my accounts, but like to think that, in all the world, there might be more than one person who qualifies as ‘the right sort’. But, even if that is so it would be mightily difficult to pick the right sorts from all the wrong ones!
My friends get to decide. Think of it as a kind of Turing test. If my friends cannot distinguish between my behaviours under my primary’s control and the usurpur’s, then the latter person IS my primary.
I came up with the terms primarybound and primarycentered to distinguish between certain atttitudes people have with avatars. Primarybound is the attitude that whoever set up the account is ESSENTIAL to that avatar.
Primarycentered considers the account holder central- but not essential- to the avatar. Whereas ‘primaryBound’ considers the primary to be the one true owner, PrimaryCentered considers the primary to be more of a curator, taking care of the digital person’s online life.
>it seems this definition explicitly includes the requirement for a human puppeteera “digital person” as you define it here could be seen merely as an artistic expression of this person<
This is true of all fictional characters. When you read Cormack McCarthy's novel 'The Road' you can either imagine a man and his son pushing a cart through an appocalyptic wasteland, or you can imagine the author sat at a desk typing out the words you see before you. You can either watch Golum crawling down a cliff face or see Andy Serkis dressed in a gimp outfit acting in front of a green screen.
It is my understanding that AI- controlled avatars are called bots. I do not claim to be a bot and currently make no use of any AI. With me, there is indeed always a human in the loop. This may not be the case for all digital characters and maybe my definition should be adjusted to accomodate that possibility.
We’ve discussed this before. Consider a man in the middle attack wherein a third party takes over the traffic between your “primary” and you. If this third party now controls Extropia, what claims does your primary have to recover control? If your existence is merely his or her artistic expression, then they own Extropia and could take legal action to recover control. But if Extropia is independent of whomever controls her, who can say the usurper, the man in the middle, is not the legitimate controller of Extropia going forward?
Extropia, it seems this definition explicitly includes the requirement for a human puppeteer or controller and therefore a “digital person” as you define it here could be seen merely as an artistic expression of this person.
It also seems to have an inherent bias. Specifically your definition seems to require a human in the loop in “real time” ruling out something like an AI based character created in advance e.g. Leisure Suit Larry, a digital character created by humans who was recently re-incarnated.
Shagggz, from your comment it sounds like you think a digital person is a bot of some kind. In actual fact, a digital person is a fictional character created and developed in online worlds/ networks, ‘pupeteered’ by a human user (or users). The user(s) may or may not be anonymous but it will be explicitly stated that the character is not meant to represent any real life person.
Therefore, I am a digital person because I conform very much to this definition.
You are not a digital person, at least not in the sense that you depict, not yet. There is no distinct personhood boundary between “your primary” and “Extropia DaSilva” because one mind guides the conduct of both – the key difference being the additional layer of pretense. Until technology develops to the point where we can sufficiently replicate the complexity that gives rise to personhood, in silico, there will be no such thing as digital people, apart from the sense of being pretend masks that are really just their “primaries” playing certain semantic games with each other.
Has your digital reverberated back to your primary, trophy?
Ah yes, leave it to the ingenuity of SL’s residents to surpass such limitations:) Thanks for the correction.
Just to point out, you’re most certainly not limited to humanoid avatars in Secondlife! theres lots of ways around (using alpha masks/mesh and other rigs) that…! 😀