Recently, I have been confronted with a rather unnerving situation where the sense of ethics of certain entities (could be people and/or organisations) was, to say the least, questionable. As I wondered how to react to this situation and how to convey my lack of trust in within the Open Badge Ecosystem, the idea of a Badge of Distrust came to mind.
As my take on badges is that they are trust statements (I’m working on refining that definition, but that will do for the time being), I realised that issuing distrust badges would put me at odds with my position where trust is understood as a positive value, as in I trust this person’s integrity. In my frame of reference, the utterance I trust that this person is a thief has a very different meaning coming from a gang leader looking at a prospective associate or from a law abiding citizen reporting a crime. For the gang leader it means “he/she is one of us” while for the law abiding citizen it means “he/she is one of them” — things can become quite convoluted when a thief steals from another thief…!
As one of the basic rules of the Open Badge Infrastructure is the recipient has the option to reject the badges they do not wish to collect, we could imagine a perfectly secure digital world where a thief would be very happy to collect a “master thief” badge from a gang leader as it could be beneficial to his/her idiosyncratic employability — crooks know how grow their own trust networks and prisons are their Open University! On the other hand, if a similar badge was issued by a law abiding citizen, it is very unlikely that he/she will ever collect it. A Badge of Distrust is not something that people are likely to collect — although, if someone like Tony Blair offered me such a badge, I would be delighted to promote it at the top of My Values badges. I would consider being distrusted by such an individual a badge of honour!
From the previous example we can foresee that accepting Badges of Trust from friends and Badges of Distrust from foes is a powerful means towards building and nurturing trust networks of all kinds — a property that should be fully explored in the development of the Open Badge Passport.
There remains the configuration of a Badge of Distrust sent to a foe. Why would a foe accept a badge of distrust? What would its value be if not collected? To explore that question further, we first need to reflect on why we would need to issue Badges of Distrust?
Why would we need Badges of Distrust?
The Open Badge Ecosystem is a conversational system, where things are not fixed once and for all. The value of credentials is not absolute, it varies across space and time, as well as with the position of the observer within the network. Looking at the dynamics of networks construction, their topology, how clusters are formed and relate to each other, will help us compute the level of confidence one might assume in making a decision based on the information provided by the network. Would the introduction of a distrust component, a Badge of Distrust (BoD), improve the quality of the decision making process? Are there potential risks associated with BoDs?
Trust is rarely an all-or-nothing affair. One entity might not be trusted for a specific thing while fully trustworthy for others. For example, the painter who is supposed to work on the windows of our house (who is definitely not the person at the source of the unnerving situation mentioned above) can be trusted for doing a great paint job while not really being trusted for delivering his job on schedule (two months delay and still not finished!). There are also different domains of (dis)trust: it is not the same thing not to finish within schedule and to be at odds with ethics. How could that be expressed within the Open Badge Ecosystem?
If the Regional Chamber of Trades and Crafts delivered to its members a series of badges such as qualified painter, on-time delivery, on budget, that clients could endorse, many endorsements of the qualified painter and on budget badges with very few for the on-time delivery badge would be an indicator of a trustworthy professional, that should probably not be hired if one is on a tight schedule! In the absence of such badges to (not) endorse, there is nothing I can do within the Open Badge Ecosystem to express that my painter cannot be trusted for on-time delivery. Yet it is a piece of information that would be useful to future clients — and to help the painter to reflect on his practice.
Now, consider a group of ten professionals declaring trust for each other by endorsing each others’ badges. Imagine that one of them has his behave ethically badge endorsed by only two colleagues, while the good presenter badge is endorsed by nine. One might be led to conclude that ethics is not his/her forte. If wondering whether the discrepancy in the number of endorsements is an accident (didn’t think to endorse it, my mistake, sorry!) or a conscious act (no way!) one means to confirm or infirm the initial impression would be to invite those who have endorsed the other badges to endorse the behave ethically one as well. None or too few responses could be interpreted as a confirmation of the initial impression. The negative message received from the lack of responses would be reinforced if, among the non-respondents, several had endorsed the behave ethically of other members of their communities.
The problem with this kind of inference (not endorsed, therefore not to be trusted) is that it is all wrong. Having no information about something cannot be taken as the basis for a negative inference! If we have no information relative to tomorrow’s degree of sunshine, it does not mean that we should expect rain. While a 20% rain forecast is equivalent to 80% dry, 10 people believing that one person behaves ethically is not equivalent to the rest of the person’s network minus 10 people believe that this person does not behave ethically. It is the kind of reasoning used by totalitarian regimes and warmongers — if you are not with us, you are against us! Moreover, as there is no reason to force everyone to have a behave ethically badge, not endorsing it is not even an option.
If we believe that a piece of information relative to distrust is critical to the efficiency of a trust network, then we probably have no option but to introduce something akin to Badges of Distrust.
How could Badges of Distrust Work?
As discussed above, Badges of Distrust will most likely not be collected by their earners, although they might be sent to them to give them a chance for reflection and a rebuttal. And if collected, it is very unlikely that the recipient nor the issuer, will make them public.
One possible solution would to keep a copy of a BoD in the restricted area of the Open Badge Passport of the issuer, an area only accessible to services and people holding the right credentials — e.g. having the right badges!
If the Badge Alliance offered a trust mapping service, this service could have access to BoDs to inform the map. A good trust mapping service should be able to differentiate between authentic BoDs and those that would come from trolls or other cyber-warmongers.
For example, if Amnesty International created a generic War Criminal Badge one might be tempted to create an instance for a certain T. B., keep it in their Open Badge Passport, then provide access to Amnesty International and other social activist networks. To avoid abuse, there could be conditions associated with instantiating such a badge, for example limiting it to Amnesty International members (i.e. holding an AI membership badge). As creator, Amnesty International would also also have the ability to revoke unfairly issued badges.
A BoD issuing sequence could go like this:
- Amnesty International creates a new Badge Class: War Criminal. Those badges can only be instantiated by one of AI’s officers.
- An instance of a War Criminal badge is created by an officer for a certain T. B. which is kept in the restricted area of Amnesty International Passport.
- Amnesty International members are invited to endorse this badge then share their endorsement within their trust networks as part of a campaign for the indictment of T.B.
- Members of the trust networks of AI members are invited to endorse the endorsement badge and invite the members of their own trust networks to endorse their own endorsement
The choice to “endorse the endorsements” scenario instead of endorsing the original BoD could make sense in a context where Amnesty International only accepts endorsements from their members and not from second degree connections. It is also a means to show how chains of trust could emerge out of chains of endorsements. In this example, a network of trust is expressing strong distrust towards a a war criminal.
Of course, once introduced, less reputable organisations and individuals will be tempted to use BoDs for less worthy goals such as flaming and bullying. While it is not a risk to be taken lightly, those things do not need BoDs to happen. What we should rather explore is how the information collected from trust networks could help smother any such attempt and confine trolls and other cyber-warmongers within their own trust networks — after all, creeps have the right to create their own communities of practice, in cyberhell!
At its current stage of development, the Open Badge Infrastructure, and practice has solely focused on positive credentials. The only negative action possible with a badge is its revocation. How can this be reconciled with what happens in the real world where positive and negative statements are being issued, e.g. recognise that one person a has high sense of moral integrity while another has a poor sense of ethics? At a more trivial level, how could I convey the information that my window painter works well but not on schedule?
While Badges of Distrust (BoDs) have been explored as a possible response to the need of expressing negative credentials, their introduction might have a perverse effect such as creating the conditions for slander, defamation, bullying, etc. Simultaneously, BoDs could trigger a reflection in the receiving party that could lead to further positive credentials this time — and/or the emitting party could revise an unjustified judgement after rebuttal.
Is it a path worth exploring? Is it safe or are we going to open a can of worms? Should those initiating their development deserve a BoD?
PS: I would like to express my gratitude to those (unnamed) who have (unknowingly) contributed to feeding this post and stimulated improvements in the design of the Open Badge Passport.