« Some (more) thoughts on repositories | Main | OpenID market research »

November 10, 2008

define:digital identity

I recently had need of a one line definition for 'digital identity' (as part of writing some blurb for a forthcoming (invitation-only I'm afraid) workshop with our newly funded 'identity' projects).

My usual course of action in such situations is to type "define:whatever" into Google and/or to go to the appropriate Wikipedia page (though more often than not both approaches lead to the same definition in any case).

In this case however, I felt a little let down.  Wikipedia currently defines 'digital identity' as follows:

Digital identity refers to the aspect of digital technology that is concerned with the mediation of people's experience of their own identity and the identity of other people and things. Digital identity also has another common usage as the digital representation of a set of claims made by one digital subject about itself or another digital subject.

Perhaps it's just me, but I find that opening sentence somewhat less clear and helpful than it might be.  "Concerned with the mediation..." - what's that all about?

Well... to cut a long story short, I spent some time looking around at alternative definitions, including those used in some of the proposals we received in response to this year's grant call, and came up with the following:

Digital identity is the online representation of an individual within a community, as adopted by that individual and/or projected by others.  An individual may have multiple digital identities in multiple communities.

I appreciate that this isn't technically one sentence but it is short and sweet - and reasonably easily understood.  I'd welcome comments about it.

As Pete pointed out in a comment on one of my previous posts, Steve Warburton of the Rhizome project has quite a nice set of slides exploring the issues of digital identity in the context of learning, teaching and research:

I particularly like the opening quote from Cole Camplese:

"As I try (and leave) more and more environments I am depositing small identity artifacts that I can no longer track and I am feeling like I am fracturing my identity more and more along the way.”

My suspicion is that most of us feel a bit like that! And, as Steve says on slide 8:

  • digital identities are performed across a variety of electronic spaces
  • we are in effect, curators of the self
  • leveraging a number of differing services
  • comprised of structured (transactional) and unstructured data
  • resulting in the creation of distributed, proliferating digital selves

To try and illustrate this I've added a short case-study about my own 'fractured' digital identity to the wiki that the projects are using to gather scenarios in advance of the workshop. The remainder of this blog entry contains a slightly updated version of my case-study.  It is by no means complete.  I wanted to try writing it all down partly because in a recent radio interview for Emerging Mondays I was asked a direct question about what I thought my own digital identity was and I didn't really have a sensible answer (not that this case-study is necessarily a sensible answer either).

The text below follows the formatting used in the wiki.


What was the setting in which this case study occurred?

  • Prior to the 'Web 2.0' age I was reasonably successful at focusing the bulk of my digital identity at a reasonably small number (3 or 4) Web 'home' pages. (There was other stuff of course - like every email I've ever sent to a public list - which was more distributed but I'll ignore that for now.) A Google search for 'Andy Powell' still returns two of these (both on UKOLN servers) as 3rd and 4th hit -  coming after the guitarist of Wishbone Ash (my long-term Google nemesis).
  • This was achieved in part simply because of the high ranking of UKOLN pages, but also thru a reasonably consistent approach to linking back to my UKOLN home page from email footers, other web pages, open source software README files, and so on.
  • Since the advent of Web 2.0, my digital identity has become dispersed across a large number of sites - Facebook, Flickr, Typepad, Blogger, Hotmail, GMail, andypowe11.net, Animoto, Slideshare, Del.icio.us, the Eduserv web site, LinkedIn, Plaxo, FriendFeed, YouTube, Yahoo Pipes, Ning, … I could probably go on and on.  (Note that not all of these offer public profiles and that for some of them I contribute under several accounts).
  • I take reasonable care to name myself consistently within these services as 'andypowe11' (note the use of digit '1's rather than letter 'l's) but that name is not always available (e.g. Google) or appropriate (e.g. Facebook and Eduserv). In any case, some of the material is hosted on joint work-related 'eduserv' or 'eduservfoundation' accounts within these services - with hindsight this was probably not a sensible approach to take but I'm now stuck with it.
  • The choice of name (using digit 1's reather than letter l's) has resulted in a reasonably unique name, but can also lead to some confusion, e.g. where people mis-read it as 'andypowell' (with letter l's) - it is certainly not an intuitive search term for people to use when looking for me.  It is also not used totally consistently, for the reasons outlined above.
  • I have 3 primary email addresses ([email protected], [email protected] and [email protected]).
  • My public relationships with other people are mostly embedded into the sites listed above (using their internal friending mechanisms) - particularly Facebook, Twitter, Second Life.
  • Facebook is probably the biggest of these. It contains both personal and work relationships. In the main, other services contain mainly work-related relationship details.  Note that the ultra-simplistic use of 'friend' as the only available relationship type in these services doesn't capture any of the more subtle aspects of my relationships with other poeple in any case.  Note also that one of the things that has changed significantly over the last few years is that I have a much stronger 'personal' presence on the Web.  Before about 5 years ago, my only visibility on the Web was professional.  There is an ongoing tension around exposing my professional life to my personal friends and my personal life to professional colleagues.  Whilst this isn't a major problem or headache for me, I am conscious that there is now much more cross-over than there ever used to be.
  • I also have an alter-ego, in the form of Art Fossett (my Second Life avatar), who appears both in-world and on the Web (via the ArtsPlace SL blog and a Flickr account for example) and in email ([email protected]).
  • I am the developer of Second Friends, a Facebook application that allows people to share parts of their Facebook, Second Life and Twitter accounts (in relatively limited ways). This is the major point at which I draw together my two primary digital identities - though in general I make no secret that Andy Powell and Art Fossett are one and the same.
  • As Andy Powell, I blog at eFoundations, which I author jointly with a colleague (Pete Johnston) at Eduserv - i.e. this blog isn't all my own work. I also maintain a personal blog (intermittently) at a7eleven.
  • I have at least 3 OpenIDs (as Andy Powell - and others as Art Fossett) of which my preferred one currently is http://claimid.com/andypowell. ClaimID have a nice 'verified' option, allowing readers to verify that I own the things I say I own (at least in the context of trusting what the ClaimID site says about me).
  • The content of many of these fragmented parts of my digital identity(ies) is quite fragile - in the sense that it resides at external Web 2.0 services over which I have little or no control and which are probably less persistent than I am.
  • All in all, it's a confusing picture.


What was the problem to be solved, or the intended effect?

  • I would like to consolidate my Web presence as far as possible, at (or around) andypowe11.net (and [email protected]).
  • Note that this will not be completely possible (or desirable) - my Eduserv work related material will always reside at (or around) the Eduserv Web site for example and will always be associated with my [email protected] email address (as its most visible Eduserv unique id).
  • I also have something of an identity crisis around Art Fossett - specifically concerning how closely the digital identities of Andy Powell and Art Fossett should be related.


What was done to fulfil the task?

  • Aggregating blog, Flickr, Twitter and other content at http://andypowe11.net/ is reasonably easy to do, and I have an ongoing (lightly resourced) activity to do this (based on a combination of Yahoo Pipes and PHP scripting primarily).
  • I still need to resolve the issue of whether material associated with Art Fossett should be included in this aggregation.
  • Note that I have to dis-aggregate my contributions to the eFoundations blog before aggregating them with my other stuff (so as not to confuse stuff that is written by my co-author). This is done with a Yahoo Pipe based on a 'dc:creator=PowellAndy' tag which is auto-inserted into the Typepad RSS feed from the blog.
  • I don't currently aggregate stuff from Slideshare and YouTube, though ultimately I would like to - again, much of this material has been made available under a single 'eduserv' account - so I will have to disaggregate it before adding it to my own stuff. Again, this will be done based on the same tag.
  • I would also like to transfer the Google-juice that is associated with my 'old' personal pages at UKOLN to my new Web presence. Technically, this can be done by asking my old employer to issue a 301 (Moved permanently) from the old URL to http://andypowe11.net/. Whether UKOLN would be willing to do this is another matter - my guess is that they might be willing to do so for my old personal page (http://www.ukoln.ac.uk/~lisap/) but not for my more formal old work page (http://www.ukoln.ac.uk/ukoln/staff/a.powell/).  I haven't got round to asking yet.
  • I want to increase the machine-readability of the information held at http://andypowe11.net/ using micro-formats (hCard) and related technologies.
  • I plan to move my OpenID to andypowe11.net using delegation to ClaimID.


  • No results to date - other than a minimal but growing presence at http://andypowe11.net/.

Lessons Learned

  • Think carefully before pushing content into external web 2.0 services using a shared account because disaggregating content back out may be difficult.
  • Think carefully about where you build up Google-juice because moving it around may be outside of your control.
  • Using external tools is fine, but whenever possible host the resulting content at a domain name under your control.  For example, using Wordpress or Blogger and hosting the resulting blog at efoundations.com or artfossett.net is much better than using Wordpress, Blogger or TypePad and hosting the resulting content under wordpress.com, blogspot.com or typepad.com.  Why?  Because the result should be more persistent (or if it isn't, it is at least your own fault).
  • Stuff on the Web is messy and it's probably going to get messier... get used to it.

Of course... I could have made all this up and it is actually someone else that creates all this stuff!


TrackBack URL for this entry:

Listed below are links to weblogs that reference define:digital identity:


"Perhaps it's just me, but I find that opening sentence somewhat less clear and helpful than it might be. "Concerned with the mediation..." - what's that all about?"

My dissertation research was actually on the construction of identity in CMC systems (and how IT designers are implicated in the identities users of their systems construct). If you want to understand the 'mediated' aspect of it all and why 'online representation of an individual' isn't quite sufficient, I'd recommend starting with Goffman's The Presentation of Self in Everyday Life, as well as a few of these:

Kendall, Lori (1998) "Meaning and Identity in "Cyberspace." The Performance of Gender, Class, and Race Online." In: Symbolic Interaction, Vol. 21, nr. 2, pp. 129-153.

Balsamo, Anne (1996) Technologies of the Gendered Body. Reading Cyborg Women. Durham and London: Duke University press, 1997.

Jones, Steven G. (ed.) (1997) Virtual Culture. Identity and Communication in Cybersociety. London: Sage Publications Ltd.

Fundamentally, all of these theorists are operating from the position that identity is not simply a matter of how you present yourself. Your identity is something constituted through your interactions with others, and since identity emerges through interaction, it is obviously always mediated in some sense, and the communication mechanisms/affordances available to you for your presentation of self will have a significant impact on the identities you construct.

thanks for the pointers. That's very helpful. As you can tell, I'm new to the theoretical side of this. I admit that I primarily understand digital identity in a practitioner's sense - i.e. in terms of what I choose to do, or not do, online with my own identity - but not in a theoretical sense.

I suppose I'm more likely to use the phrase, "my digital identity is ..." than I am, "digital identity is concerned with ...". Looking at it again now, I see that Wikipedia provides definitions for both these forms - mine allows for only the first.

In that sense, Wikipedia's is better.

I certainly understand the 'constructed' aspect of digital identity - which is what I was trying to capture in the "as adopted by that individual and/or projected by others" part of my definition. But, I see that being more explicit about this would be good.

Focusing on the second part of the Wikipedia definition only, we could say...

"An individual's digital identity is the online representation of a set of assertions about them, made by themselves and others, as constructed through their interaction within a community. An individual may have multiple digital identities in multiple communities."

...though I accept that I have now made the definition longer than the Wikipedia original.

Your definition reads much more clearly to me than Wikipedia's.

But one of my big problems with a lot of discussions of digital identity is actually the effort to define online and offline identities as totally discreet. I went into this in more detail in my dissertation, but the move to define online identity as distinct from offline identity is an ideologically freighted move. It also tends to break down pretty quickly if you look at it for more than five minutes. Assertions made about you offline can affect your online identities, and vice versa.

The Wikipedia entry on digital identity reflects an effort to reduce identity to an authentication problem that computers can deal with, and computers don't handle shades of grey (and the offline world) very well. My suspicion is that even within that limited realm, though, trying to develop identity (and identity authentication) systems that treat online identity as totally discreet from offline identity is a good way of opening up holes in your security system ripe for social engineering exploits.

At least a guitarist is a fairly acceptable google nemesis. For a long time I was competing with the 'Watermelon Queen' from North Carolina...and how can you compete with watermelons :-)

Oh, I dunno... I think that signing off all your emails as Nicole "Watermelons" Harris would have a certain je ne sais quoi about it!? :-)

Forgot something... my Facebook stripper name - Rocky Tailsnake. Does that count as a digital identity? lol

The comments to this entry are closed.



eFoundations is powered by TypePad