Monday, 3 August 2015

Technical Anonymity

This has been floating around in my head for ages.

Basically, the suggestion of many people is that the internet is anonymous and this causes a lot of problems. Well, there's a few problems with that. Firstly, a lot of people have a very cursory experience of the internet largely restricted to social networking sites or non-account sites. As the term "social networking" probably suggests, the anonymous don't get too much from this. Although, of course, the whole catfish sort of thing and other variants of fake profiles are real Things. Secondly, for those people who aren't like this (of which there are also a great many, if not more) there's this idea of technical anonymity.

To understand what I mean by technical anonymity you must first grasp something really important about names. Names aren't information, they're data. .

In an academic context (i.e. Infosys 110), data and information are distinguished. A fact is a single idea and facts are data. Information, though, draws relationship between facts. Essentially, information requires at least two facts. (You can build on this foundation to derive knowledge and wisdom but that's just trivia here.)

So, back to names. If I told you my name, you'd know my name (i.e. Harry). You don't, however, know anything about me. My telling you my name says nothing about the person to which it belong. What you've got (i.e. Harry) is a simple, trivial little fact that doesn't help you do anything. However, if you know me and someone mentioned my name to you, you'd automatically recall a bunch of different facts associated with that name and you'd lump them together and reach some sort of conclusion such as "Oh, yeah, I know him: totally unreliable". In other words, the name by itself says absolutely nothing interesting.

People potentially get confused about this point because, in everyday life, we generally deal with names of people we know. Every time we hear/encounter these names we automatically recall all the different things we know. For instance, "Oh, yeah, Tom: tall, scruffy, wears glasses" or, "Ah, Gloria: tall, scruffy, wears glasses". The name is the fact which we use to organise these tidbits and create some sort of meaningful information out of. There's one big, initial, implication of this.

Anonymity Doesn't Depend on People Not Knowing Your Name

This follows pretty immediately. If your name is just trivial data to people who don't know you, what's the difference between being called, say, TaichiFan0111 and being just plain old Paul? Well, you say, the difference is that people who know plain old Paul don't know that Paul is TachiFan0111. That's actually pretty important, yet, at the same time, we know that the thing that matters is not Paul versus TaichiFan0111 but rather that people have a whole bunch of other things associated with Paul but not with TaichiFan0111. That is:

Trait(Knows) Paul(Doesn't Know) PaulTaichi0111

To be honest, there's more that I would like to do with this table but I am not that skilled with html and, currently, somewhat too time poor to research things thoroughly. For instance, I'd prefer to put the table in the centre of the document and, also, to have dividing lines but, and this is crucial, only within the body (i.e. between the columns "(Knows) Paul" and "(Doesn't Know) Paul") and only covering the size of the bottom two rows.

So what is Technical Anonymity?

Imagine that Paul uses the TaichiFan0111 (henceforth, TaichiFan) name/handle for a while on a particular website that has a number of fairly regular users. Furthermore, assume that these users have functioning memories and will remember the activities of TaichiFan over time. Sound reasonable? Honestly,. I am not sure to what extent that this is true of corrupted forums like Reddit or huge sites like, say, Twitter, but I have seen this in action in reasonably sizeable forums so that's something. Why does this matter?

Well, remember how names work? When you hear a name you associate a bunch of things with that name. Names are literally classifying tools, right? In that table example we see two aspects of Paul. One one hand we have something physical (i.e. Tall) and on the other we have some emotive/mental trait (Funny). That's a really important distinction to draw in considering technical anonymity. On a text based medium, such as this, it's pretty impossible to learn any physical facts about me (unless I choose to post a photo or something) but it's pretty easy to develop ideas about those other traits, isn't it?

As a quick illustration, look at the commonly used labels at the bottom of the homepage. We see education is way up there, so's New Zealand and so is "auckland university". You are already drawing some conclusions about me from this. I won't bother guessing at those conclusions but you understand, don't you? In some important way, you know me just from reading these blogs. You only know some stuff and only what I choose to reveal to you (whether explicitly or implicitly) but there's enough in that isn't there? In the same way, if I were going by TaichiFan then you'd still know those same things, it's just that they'd be classified slightly differently. (And, to be honest, TaichiFan0111 actually tells you stuff as well, probably. As should my choice of TaichFan as a hypothetical name.)

So, then, technical anonymity is the idea that even though you don't know who someone is, you can still know who they are.

Conversely, you can know who someone is, but not know who they are.


Someone who has been using a long term name (e.g. TaichiFan) has built up something in that, if you will, brand/identity and if they wanted to start trolling, for instance, they must sacrifice that in a way that is akin to (but not the same as) choosing not to use one's IRL identity or adopt another name (e.g. HMSYamato) to conceal their identity. That's actually a quite important thing to bear in mind when considering the dynamics of internet conversation.


I think this is a pretty interesting idea. It's not really trivial in that it does say quite a lot about how the internet alters our identities and our understandings of things like privacy. Naturally, this is going to be really big for people who will spend their really early formative years online. And it should be big for people considering these sorts of things (sociologists? managers? politicians? marketers? etc. etc.) as well.

In some ways, then, I guess, the internet means that how we understand anonymity should change, in at least some ways. That's probably as close as we're going to get to a proper big ending point.

No comments:

Post a Comment