Terrell Russell: This Old Network

With great pride I would like to announce that my dissertation was accepted by the Graduate School at UNC-Chapel Hill yesterday.

Congratulations. Your submission has cleared all of the necessary checks and will soon be delivered to UMI.

Hooray!

I’ve posted the entire dissertation online, with supporting materials created along the way.

It is published under Creative Commons with a BY-NC-SA 3.0 license.

Title: Contextual Authority Tagging: Expertise Location via Social Labeling
Author(s): Terrell Russell

Publishing Settings & Copyright
Traditional Publishing
Do not delay release to ProQuest
Allow search engine access.
Do not allow third party sales.
Do not file for copyright – I am requesting that ProQuest/UMI not file for copyright on my behalf.

Degree/Department Information
Year degree awarded: 2011
Degree Awarded: Doctor of Philosophy
Year Manuscript Completed: 2011
Department: Information & Library Science
Advisor/Chair: Deborah Barreau, Gary Marchionini
Committee Members: Barbara M Wildemuth, Sri Kalyanaraman, Phillip J Windley

Subject Categories
Information Science [0723] – primary
Organization Theory [0635]
Social Psychology [0451]

Keywords
cognitive authority
expertise
identity
knowledge management
reputation
tagging

Abstract
This study investigates the possibility of a group of people making explicit their tacit knowledge about one another’s areas of expertise. Through a design consisting of a modified Delphi Study, group members are asked to label both their own and each others’ areas of expertise over the course of five rounds. Statistical analysis and qualitative evaluation of 10 participating organizations suggest they were successful and that, with simple keywords, group members can convey the salient areas of expertise of their colleagues to a degree that is deemed “similar” and of “high quality” by both third parties and those being evaluated. More work needs to be done to make this information directly actionable, but the foundational aspects have been identified.

In a world with a democratization of voices from all around and increasing demands on our time and attention, this study suggests that simple, aggregated third-party expertise evaluations can augment our ongoing struggle for quality information source selection. These evaluations can serve as loose credentials when more expensive or heavyweight reputation cues may not be viable.

Language
en ( English )

Tags: dissertation - phd

Mark Pesce has done it again. He went ahead and wrote (spoke, actually, at LCA2011) what was in my head, better than I had (not) written it. He talks about how we’re social creatures, how we mimic innately (mimesis), and how to truly fight the man, we’ve got to own our information, our networks, and our dignity.

From Smoke Signals:

Your social graph is the most important thing you have that can be represented in bits. With it, I can manipulate you. I can change your tastes, your attitudes, even your politics. We now know this is possible – and probably even easy. But to do this, I need your social graph. I need you to surrender it to me before I can use it to fuck you over.

His four design principles to make sure we own the future:

1. Distribute Everything
2. Transport Independence
3. Secure Everything
4. Open Everything

It seems so straightforward.

So why, again, have we capitulated to Facebook?

Tags: distributed - facebook - lca2011 - network - pesce - socialgraph

This was originally posted at the TextDrive blog on May 7, 2005. Copying here for posterity.

Please consider your fellow servermates and avoid the use of weak passwords.

What Not To Do

Strong passwords are great. Cryptographically secure passwords are even cooler and highly encouraged. That said, under no circumstances should anyone be using something like “jason/jason” or “damelon/damelon” as their login/password combination.

Dictionary attacks have been monitored on these servers from the very early days and are considered “constant” today. Expect that if you are using a weak password for your account to be compromised by these attacks. This escalates the possibility that other users will be affected by your oversight. This is a very bad thing.

Choosing Good Passwords

Information about how to choose good passwords can be found in many places. A good summary can be found at the Australian Computer Emergency Response Team’s site here.

Choice Selections

“It has often been said that ‘good fences make good neighbors.’ On a Unix system, many users also say that “I don’t care who reads my files, so I don’t need a good password.’ Regrettably, leaving an account vulnerable to attack is not the same thing as leaving files unprotected. In the latter case, all that is at risk is the data contained in the unprotected files, while in the former, the whole system is at risk.”—Klein, 1991

“I remember seeing a great phrase on the Mexican Hackers Emergency Response Team page, which went something like ‘Passwords are like underwear: don’t share them, hide them under your keyboard, or hang them from your monitor. Above all, change them frequently’”—SecurityFocus

Thanks, Terrell

References

[1] AusCERT. Choosing Good Passwords. (2001) http://www.auscert.org.au/render.html?it=2260

[2] Klein, Daniel V. (1991) Foiling the Cracker; A Survey of, and Improvements to Unix Password Security. Proceedings of the 14th DoE Computer Security Group. May 1991. http://www.klein.com/dvk/publications/

[3] SecurityFocus. (2001) Password Crackers – Ensuring the Security of Your Password. http://www.securityfocus.com/infocus/1192

[4] Smith, Richard E. (2002) The Strong Password Dilemma. http://www.smat.us/sanity/pwdilemma.html

Tags: passwords - textdrive

I just read the most plausible of law review papers suggesting the potential for protection of a private space within social network sites (SNS). Fellow UNC grad student Woodrow Hartzog proposes the use of Promissory Estoppel as a means to protect self-disclosure in online communities. It would create a type of contract or agreement between users of a site whereby a protection would exist for information disclosed in that community or site. If someone else shares the disclosed, private information, with a few caveats, they can be held accountable.

Abstract:
The unprecedented sharing of private information on the Internet is leading some to herald the demise of privacy. It is far too facile, however, to conclude that because people are sharing private data online, they should expect no privacy. The need for confidential disclosure is no more prevalent than when sensitive information such as dating profiles, candid thoughts and past substance abuse is revealed in online communities. What happens when information leaks outside these communities? Traditional remedies will likely fail to protect people when members of an online community violate the confidentiality of other members. In this article, I contend that the law can ensure confidentiality for members of online communities through promissory estoppel. This is the first article proposing the application of promissory estoppel via a website’s terms of use as a method for protecting disclosure in online communities. Under the third-party beneficiary doctrine or the concept of dual agency, these agreements could create a safe place to disclose information due to mutual availability of promissory estoppel.

Hartzog goes on to quote Professor Daniel Solove in a passage on practical implications:

The use of promissory estoppel to protect self-disclosure in online communities is consistent with many legal and public policy considerations besides privacy. Additionally, it could help create a stronger normative culture of confidentiality to protect the well-being of online community denizens. Professor Daniel Solove has asserted that “[p]rivacy, in the form of protection against disclosure, regulates the way people relate to others in society…[I]t promotes one’s ability to engage in social affairs, form friendships and human relationships, communicate with others and associate with groups of people sharing similar value.” … His conclusion underscores the need to create a safe place for disclosure online.

The four part analysis of whether a promissory estoppel should be applied is proposed as:

1) Was there a clear and definite promise?
2) Did the promisor intend to induce reliance on the part of the promisee, and did such reliance occur to the promisee’s detriment?
3) Must the promise be enforced to prevent an injustice?
4) What are the damages?

Hartzog ends his paper:

The proposed theory of recovery advances privacy as control over personal information, one of the foundations of information privacy law. It focuses on reliance instead of a commercial-based bargain theory. It also encourages speech by offering a safe place for sensitive self-disclosure and an easier process by which potential disseminators of information disclosed within a community can determine the appropriate level of discretion to apply to accessed information.

Ideally, if utilized over a significant period of time, the promissory estoppel remedy could create a stronger normative culture of confidentiality through improved channels of internalization of duties of discretion. Additionally, the solution is likely compliant with the First Amendment, as analyzed under the Cohen standard. Finally, although the available damages under promissory estoppel are less than that in tort, the theory could potentially have an effect on other torts, such as the tort for breach of confidentiality.

It is difficult to predict the full impact adoption of the promissory estoppel remedy would have for online communities, but the provision of a safe place for users to disclose personal information online would likely promote both speech and the personal well being of online community denizens.

Tags: disclosure - estoppel - hartzog - law - privacy

I keep meaning to write down when these things happen… The march towards consolidation seems so obvious to me, and yet people are still confused when I suggest they can keep things separate.

Deep Throat
A few years ago now, in 2005, the world finally learned the identity of Deep Throat. He had remained pseudonymous for over 30 years. Mark Felt came forward himself when he allowed the release of his name in a Vanity Fair article by his attorney. The disclosure was on his terms. He decided to end the secrecy before he died.

This is something that I claim would be impossible in today’s interconnected and recorded world. Are there stories today that are being published where the sources are on “deep background” and the public is clamoring to know the source’s identity?

The Fake Steve Jobs
The Fake Steve Jobs had a good thing going with his blog The Diary of Fake Steve. He was continually witty and received rave reviews for his poking fun at the mystery and aura that is Apple and Steve Jobs, proper. Of course, over time, his identity was revealed by the New York Times to be Dan Lyons. And like Felt, there was a book deal shortly thereafter. The ruse lasted 14 months — much longer than expected.

“I’m stunned that it’s taken this long … I’ve been sort of waiting for this call for months.” — Lyons

He has since taken up the writing as Fake Steve again – and it’s still just as funny – but without the cloud of intrigue as to who would be so bold…

_why the lucky stiff
Yesterday, _why, a fairly well known programmer in the web2.0 space apparently deleted his online presence. This is news, regardless, but what’s more interesting is that “_why” is a pseudonym and so far, we don’t know for whom. He has deleted his accounts, his blogs, his code and for now, the community of programmers and hackers have yet to unearth his identity. The thread at ycombinator seems to be getting close – I suspect it is only a matter of hours before we get some confirmation.

John Resig posted a remarkable eulogy (his word) to _why:

At this moment, _why’s online presence appears to be no more. All of his sites and code are gone. This includes, and is not limited to:

* http://twitter.com/_why
* http://github.com/why
* http://whytheluckystiff.net/
* http://poignantguide.net/
* http://hackety.org/
* http://shoooes.net/
* http://hacketyhack.net/
* http://tryruby.hobix.com/

Two conjectures are common at the moment: His account(s) were hacked and sites taken down or he simply decided to delete his online presence. I personally believe that he did this deliberately and with some amount of forethought.

Still Hidden?
What examples do we have where we still don’t know who is behind a widely-known* piece or body of work? Does it still happen? The timeframe for the ability to remain unknown is correlated with visibility, no doubt.

I’d love to keep a list somewhere…

Tags: consolidation - identity - peudonyms - why - why the lucky stiff