Monday, August 1, 2011

True Names

anti-aliasing n. -soc. psych. Curiosity about the real flesh-and-blood people behind internet usernames, whose vivid individuality suggests that when our parents were tracing their fingers along our nameless faces looking for some hint of who we were to become, they really should have gone with Mr. Cookieface, Unicornpuncher, Dutchess Von Whatever, or Wookiegasm.

(The Dictionary of Obscure Sorrows)

Two weeks or so into the controversy, Google Vice President Vic Gundotra spoke to Robert Scoble to clarify the company’s position, and Scoble dutifully relayed the conversation to the world. As the person in charge of policy, Gundotra is making some tough choices – Scoble explained – and is prepared to be judged on the outcomes. He is trying to set a positive tone in the new social network. But most importantly:
He says it isn't about real names. He says he isn't using his legal name here. He says, instead, it is about having common names and removing people who spell their names in weird ways, like using upside-down characters, or who are using obviously fake names, like "god" or worse.

Like “god”, or worse. I’d be content to talk about those four words alone this week. The company that told us to go out and put all of the people in our lives in a series of circles doesn’t want us to be messing with esoteric symbols; the corporation whose motto is ‘don’t be evil’ forbids us to interfere with the divine.

But amusing as that is, it’s an exercise in misdirection. Google’s 'real names' policy is not about the scourge of upside-down characters, strange spelling or even god impersonators: it’s about ensuring that the social web continues to fulfil its primary function as the largest and most efficient market research platform ever built.


It started around the 8th of July, when the earliest reports begun to circulate that Google had been suspending accounts on its neonate social network, Google+, due to naming irregularities. The cull included businesses posing as individuals but also individuals who signed up with obvious pseudonyms, such as Second Life avatars. The backlash concerning the latter group was fierce. If you’re interested in the details, the best source is a polemical and very lengthy post by Kee Hinckley. Suffice it to say here that Google’s critics argued in the main for the value of and the need for pseudonymity (as opposed to anonymity) – stuff that by now ought to be utterly obvious to everyone except there will always be people who are prepared to mistake privilege for virtue. Here’s one, a chap by the almost-too-plausible name of Joe Carter:
It probably won’t surprise you to hear that I don’t have a lot of respect for the opinions of people who won’t sign their name—their real name—to what they write. Sure, I may engage you, agree/disagree with you, thank you for your comment, etc. But I can’t honestly say that I respect such opinions or give them much thought—and why should you expect me too? I put my name to everything I write on the Web (however imprudent that may be). Why should I take seriously the thoughts of someone who isn’t willing to undersign the statements they make?
Are you gay, disabled, kinky or an anarchist? You need to find yourself a nice little community of like-minded or like-bodied people with whom to discuss your marginal concerns. For everything else, you must sign your real name and constrain your personality and opinions to suit – in other words, be the kind of person who can speak their mind without the slightest fear of repercussion or unintended consequence.

In other-other words: keep the most distasteful bits of who you are the hell out of my feed.


True names is the title of a short but exceptionally clunky novella by Vernor Vinge often credited with articulating the first fully developed vision of cyberspace. In fact Vinge’s Other Plane postdates John Varley’s memory cube by roughly twice as many years as it predates William Gibson’s matrix, but practically its only feature of any enduring interest is that very clunkiness, its utterly literal and schematic prefiguration of the struggle to come between the huddled online masses yearning for free expression and the forces of state and corporate control insisting that every act, every utterance in cyberspace be traceable to a real person in the real world and stamped as to the real time of its occurrence.

In True Names, every user is by definition a hacker, and so to discover the real name of another user means to own them, for they’ll do anything if you promise not to reveal their identity to the police. However when an artificial intelligence by the name of Mailman makes a play to assume control of cyberspace, the hackers have no choice but to turn to the Feds (formerly their ‘Great Enemy’ – really) to help them vanquish this new and even greater threat to their freedom.

This scenario anticipated some of the directions of cyberpunk, albeit without any of the poetic force, but Vinge was also far more comfortable than Gibson, Sterling, Stephenson and the rest with some of the more startling implications of the new socio-technological paradigm. The epilogue of True Names contains in fact one of the earliest genuine forays into transhumanism, in the form of the main female character’s preparations for an afterlife in the Other Plane.
My kernel is out there in the System. Every time I'm there, I transfer a little more of myself. The kernel is growing into a true Erythrina, who is also truly me. When this body dies […] I will still be, and you can still talk to me.
Erythrina’s true name is Debbie Charteris, but the closing line of the novella, spoken by her male counterpart, leaves no doubt as to which name best captures her essence.
Beyond those years or decades... were millennia. And Ery.

There would be a lot more to say about this brand of immortality-seekers that is of interest to the topic of this blog, but for now I just want to place a particular idea in time: thirty years ago cyberspace as we have come to understand it was already being imagined; it was invested with nothing less than the capacity to provide a support system for the human soul; and loss of control over one’s real name was seen even back then, and in a very non-metaphorical way, as the fundamental existential threat.


Which makes it all the more extraordinary that Google should have failed to anticipate what the reaction to their crude and clumsily implemented policy would be. Faced with the onslaught, the company’s apologists deflected the attention of the critics towards the extent in which a pseudonym is in fact allowed – namely if it’s the name you’re commonly known as. Say, Mark Twain. Or Lady Gaga. Or Vic Gundotra, instead of whatever his legal name is (what’s the bet it’s Victor Gundotra?). But if the condition for the use of a pseudonym is that it is publicly associated to your real name in real life, then it fails the test of a useful online pseudonym.

But like I said, this is not what it’s about. Facebook's policy surrounding account names, subject to the service’s grandiosely named ‘authenticity standards’, are even more stringent than Google’s:
  • Your full first and last name must be listed. Initials cannot stand in place of your full name. Nicknames can be listed if they are a variation of your first or last name, but only in the format "First Name ‘Nickname’ Lastname."
  • Your name must be listed using characters from one language only.
  • Your name cannot consist of any titles, symbols, numbers, unusual capitalization, or punctuation.
In spite of earlier pronouncements in praise of 'the freedom to be who you want to be', Google has chosen for its own social network an approach that is only fractionally softer than its rival’s because unique, traceable identities are very valuable to its actual customers – advertisers. And the endgame, remember, is not to sell advertising as such: it is to construct the ultimate social graph, a graph capable of matching your online behaviour to your consumer preferences in order to expose you to the perfect ads, thus maximising the revenue of both the platform and the advertisers. In time, the product sold by both Google and Facebook will be not so much the advertising space as the graph itself.

To prevent spurious aggregation of the demographic data, accounts on both networks are strictly personal. Says Facebook:
Please keep in mind that Facebook accounts are for individual personal use. Accounts representing groups, families or couples are not allowed.
But the principle of truthful representation is even more important. The requirement that ‘[y]our profile should represent you’ – a pillar of the ‘community standard’ on which Google’s policy is based – is not there to prevent us from lying to each other, but from lying to Google. Were I to impersonate a historical figure, or a person of another gender, or engage in any sort of conscious role-play altogether, I might declare to like products and services other than the ones that I actually consume, thus introducing false data into the set and polluting the graph.


Now I’m not saying that Facebook or Google will actually ever master the social graph, at least not any more than I’m saying that Vernor Vinge or Marvin Minsky are going to live forever inside their computers. But it’s a powerful fantasy, one that is forming very real perceptions about the values of companies’ stocks and – more depressingly – the social exchange itself. We may perceive its impact as promoting exclusion from these services, which is bad enough – for belonging there will soon likely have to be considered optional only in a very narrow sense, like it’s optional to have a telephone. I have talked about this. But in another, less immediately obvious sense, these restrictions promote a further narrowing of the meanings of the word identity.

Think about Web 1.0, with its sometimes ugly, sometimes startling or pretty homepages that took forever to load; with its the lack of templates not just for communicating your interests, but for thinking of yourself inside of a giant global house. Cast your mind back to before the Web itself, to email and bulletin boards and Usenet with their ubiquitous pseudonyms, as well as the sense of possibility: for personal reinvention, for activism, for accessing kinds of knowledge that you never even suspected could possibly have mattered to you – but did. And now think about the prescription to be yourself on the internet, and how it forecloses the multiple identities you might wish to play with, each orthogonal with your various interests, curiosities, public or secret kinks, and the languages you speak. Think of the requirement to be one person, separate from friends and family, mates and lovers, comrades and foes; to be of your own gender, and be interested in men/women/both; to be of your own race, of your own class. And watch your social imagination be eroded, inch by bloody inch.


There is a grotesque coda to this story. In the middle of the controversy, Norway happened, and some days later somebody had to ask: Why didn’t Google catch Anders Breivik? It was one of those weasely questions that journalists – or, in this case, academics – like to put in the mouths of the public sometimes, posed only so that it could be unpicked and debunked. We know that Breivik spent an alleged 200 hours on Google searching for terms relating to bombs. We know that he published a 1,500 word manifesto under his own name. But for god’s sakes, we also know that Google is not the internet, and it’s not a secret service, and it’s not empowered, nor is it subject to the necessary checks and balances. And we’re not fucking stupid, so we’re not even asking the question. Yet we must sit patiently while we are told the answer.

Not in the category of the horrific, but of the chilling, are the images of Breivik that have circulated. Two in particular: the one in which he is wearing the blue Lacoste sweater and the one currently on Wikipedia. How polished they are, like professional portraits. That is the perfect Facebook face. That is a man that is comfortable with who he is. And this is also what the massacre was about: obsession with fixed identities – racial, cultural and political – the disease that is eating Europe from within.

Perhaps it was Facebook that ought to have caught Anders Breivik. Perhaps that’s what was reckoned by the person who filmed Breivik’s Facebook page before it was taken down and then put it on YouTube with the accompaniment of an epic score. Perhaps there should be another graph – an anti-social graph, a murderers’ graph – that accounts for the darkness, the madness, the criminal intent.

Except it is a dangerous fantasy that these identities of ours are unique and transparent, and a perverse imperative that we should seek to hold them together at all times, and always be true to our true names.


James Butler said...

Posted without comment.

Giovanni Tiso said...

I had something about that in draft (mostly I wanted to include Figure 8, in all its blinding awesomeness), but then it just became too much. And Twitter really does deserve a separate treatment, what they are doing *is* different even though there will be people trying to extrapolate a social graph from it nonetheless - with the eventual aim of selling stuff to Samuel Pepys, no doubt.

James Butler said...

I'm also intrigued by the relevance of the humorous definition of "anti-aliasing". Real anti-aliasing is a technical workaround for Nyquist's theorem - a consequence of the information that is lost when transforming an analogue into a digital signal is that artefacts (untruths) are introduced, and the only way to get rid of them is to smooth them over, erase the detail, and pretend that we never knew that about the signal in the first place.

Giovanni Tiso said...

That's the thing about the Dictionary - it's really quite clever.

Ben Wilson said...

As a stipulation, a true connection between the person and their handle is scary in proportion to the reach of the organization collecting the data. Anyone with a small friends network they developed for some minor purpose is hardly sinister if they wish to know the actual identity of their members. But something the size of Google Wave....

As a practice, I've personally found using my real name quite a good far. This is precisely because it does limit the extent to which I can engage in bad faith discussion.

However, as you say, flights of fancy, alter egos, secret lives, etc are not only fun but important institutions to be able to access for everyone who wishes to.

Google, leave anonymity alone.

Greg Dawson said...

(mostly I wanted to include Figure 8, in all its blinding awesomeness)

Figure 8 made my day - thanks :)

George D said...

There is also the fact that many people, millions of Indonesians for example are mononymous, and first name last name has no place in their reality.

I'll come back here, but the comment about cyberspace rung entirely true. Cyber has shrunk, realspace has intruded.

stephen said...

About whether or how Google or Facebook could/should/might have caught Breivik: I just want to note that years ago, at the University of Waikato, the most frequent reader of the Usenet group alt.politics.white-power, and most frequent poster, was me. To this day, I read some vile and disgusting things on a semi-regular basis precisely because I have a special interest in opposing them. I'm also interested in chemistry and radical politics generally... I shudder to think how a monitoring or profiling strategy would react to my internet usage habits.

To some extent I despair over the possibility of online pseudonymity. On the one hand, if the authorities ever really care, it is already possible to identify people by their online behaviour with a high degree of accuracy, and on the other, it is evident that the authorities don't care that much about accuracy. (Viz the REAA and the cases they mount and often succeed in in the US).

I can easily imagine trials in New Zealand where my expert testimony would be impenetrable to a judge or jury or at least obfuscated by a determined prosecution.

I am pessimistic about genuine trustworthy promises of anonymity on the internet. They exist to no greater, arguably lesser extent that they do in real life. On the internet, we know you are not a dog. We know who you are, Mr Tiso, no matter what you call yourself.

Giovanni Tiso said...

It's the most terrifying thing, isn't it? An apparatus of total surveillance run by idiots. That was Brazil's inspired twist on Orwell.

However, this:

We know who you are, Mr Tiso, no matter what you call yourself.

needs to be unpicked somewhat. There is the inscription in the Net of various forms of control, including state control (the problem with The Onion's take on this, as usual, was that it simply made too much sense) and then there is the intersecting, horizontal plane of people spying on one another, keeping tabs on one another, including the net cast by the marketers, as evidenced in the link upthread to the gender recognition software applied to Twitter. And then there is self-surveillance, the process of interiorisation of all of these instruments of observation and control. But none of these systems are perfect. Some of them aren’t even manned, as Breivik’s case makes clear - they are potential manifestations of the cybernetic will to absolute knowledge that the transhumanists both desire and dread.

In reality, ‘you’ actually don’t know who I am. And not just because I write under this clever pseudonym that happens to be identical to my legal name, but also because all of the different Giovanni Tiso avatars on the Net whose authentication details I am periodically asked to recall and confirm – lest my identity is stolen – are the product of different discursive environments the interplay of which is very complex indeed. That researchers are excited that their algorithms can correctly guess the sex of a speaker with 75% accuracy really shows up the gap between their fantasy of total insight and the reality of their not even being able to tell if the person behind the screen is a man or a woman (a 50-50 guess) a full quarter of the time.

stephen said...

Yes, statistical algorithms are weak. But when we want to unmask someone, we don't use statistics alone, or even at all. Eg, see how Tom McMasters was uncovered, not by agents of a state, or any sort of professional service, but simply by ordinary people pooling their individually limit3ed information.

Given merely the textual evidence of this blog post, we can know nothing for certain about who you are, but assuredly if "Giovanni Tiso" were an alias for a 20-something Syrian lesbian who merely pretends to be Tiso in order to draw attention to the plight of expatriate Italian translators, this could and would be uncovered, at least to the extent that your post became of significant interest to the right people.

I guess I'm thinking about looking through the other end of the telescope. Given a sea of people, we cannot sieve out a Breivik. But given a Breivik, we can find most anything connected with him.

wv: "ramst", an obsolete second person verb form.

Giovanni Tiso said...

"Given merely the textual evidence of this blog post, we can know nothing for certain about who you are, but assuredly if "Giovanni Tiso" were an alias for a 20-something Syrian lesbian who merely pretends to be Tiso in order to draw attention to the plight of expatriate Italian translators, this could and would be uncovered, at least to the extent that your post became of significant interest to the right people."

True, but consider also for how long that rather loathsome charade was successfully carried out in spite of the intense scrutiny. I would also counter that it is one thing to catch the Great Liars who lay claims to very precisely situated and contentious offline identities, quite another to fix the subject positions of more ordinary con artists and experimenters.

One of the strongest voices I have ever encountered online was a male writer posing as a female on several Italian literary newsgroups. The name was an obvious pseudonym (it translated more or less as “Mary Stanza”). The gender was a put on (as many of us suspected). But aside from that there were no forged biographical details, no especially elaborate pretence. The maudit persona allowed the author to engage in some serious trolling, but s/he also wrote very compellingly most of the time. And besides giving you an opportunity to observe how writing always involves creating and maintaining a separate identity, the experiment left you uncertain as to what unmasking the author would even exactly mean.

James Butler said...

It's interesting how the authors of the Twitter gender paper "verified" the genders of their subjects only by checking against existing online profiles - ie. what they were really testing is whether they could guess the gender a given person self-identifies as on the internet. I don't know if it's dealt with in the paper, but the reporting I've seen seems to assume that they can find your "real" gender - whatever that means - while obviously this method can tell you nothing of the sort (not to mention that they've thus restricted their sample to people who are comfortable with assigning themselves to a binary gender).

Obviously none of this is necessary to Google - they only need the largest set of associations to a given "person", any other measure is irrelevant (in the same way that Google Translate doesn't bother with frivolities like semantics, only the statistical mappings between groups of words). So I'm not sure this:

Were I to impersonate a historical figure, or a person of another gender, or engage in any sort of conscious role-play altogether, I might declare to like products and services other than the ones that I actually consume, thus introducing false data into the set and polluting the graph.

is the problem (would you really express false preferences, or just those true ones you mightn't otherwise be comfortable revealing?), so much as that you would now have two smaller, disconnected preference graphs, with no way for Google to know to show to one ads targeted at the other.

Giovanni Tiso said...

Fair enough. I didn't mean that the polluting of the graph would literally be an problem, and the point that a pseudonymous you may in fact reveal truer preferences is well taken. In fact market researchers have ways to discount random answers and outright lies. I suspect that the aversion is more ideological. The social graph, much like transhumanism, is predicated on the idea that identity is autonomously and truthfully constructed by necessity, in real life as much as on the Net. You are either a pattern of firing synapses or a pattern of consumer behaviours, but either way the possibility that the pattern may not be faithful to reality is best programmed out of the system.

George D said...

I think the commercial imperative is made manifest in rather obvious ways, and this does help us in thinking usefully. However, opens up another line of enquiry by identifying this as a use of power. ""Real names" policies aren’t empowering; they’re an authoritarian assertion of power over vulnerable people."

She covers at least some of the territory, so I'll only add the suggestion that the act of identification is itself an assertion of power (micropower) and a claim to a right to exist in a certain form. Obviously, if you exist in the form of a white, college educated male or female, this is self-evident, and need not be stated or acknowledged. For those in less advantageous positions, or the subject of their own power struggles such positioning is to reposition yourself. It's easy to see Google etc as merely profit making machines, but they're also in the business of replicating and enforcing social orders (largely, I would argue,the result of the cultural norms of those they constitute)

Giovanni Tiso said...

"It's easy to see Google etc as merely profit making machines, but they're also in the business of replicating and enforcing social order"

Especially those aspects of the social order that underpin (and are in turn the product of) the political-economic system in which Google makes those profits, namely neoliberal capitalism.

"Real names" policies aren’t empowering; they’re an authoritarian assertion of power over vulnerable people."

While I believe this to be true, what I wanted to argue this week is that they constrain social imagination across the board, and not solely to the detriment of the most vulnerable.

George D said...

Especially those aspects of the social order that underpin (and are in turn the product of) the political-economic system in which Google makes those profits, namely neoliberal capitalism.

Especially, but not exclusively. My experience is that companies create and replicate aesthetics with an aim of producing a profit, but these are frequently at least partly at odds with doing so. Most companies will forgo some profit to make them feel more comfortable with their operations.

I'm quite prepared to entertain a functionalist argument that presents capital as a primary motivator, but this is coincidental (largely, but not exclusively) with other markers of social power-accumulation.

Giovanni Tiso said...

"Most companies will forgo some profit to make them feel more comfortable with their operations."

This article by James Gleick for the New York Review of Books looks at "Don't be evil" amongst other things, and in spite of the very unsatisfying ending is a useful refresher on the company's history. It's not clear to me that Google has done any forgoing of profits whatsoever at any time in this history.

My suspicion is that Google's extraordinary success is due to the remarkable convergence of its commercial objectives with the 'markers of social-power accumulation' you speak about. It is not so much a company that espouses digital ideology as one that embodies it - far more so than Microsoft or even Apple.

James Butler said...

LOL, apropos this my wife just tried to sign up to Facebook - the better to keep tabs on her brother in the US - and was denied the name "Tobias Fünke". She is Not Amused, as I am "friends" with people she has no wish to be noticed by.