Alongside DefCon is the smaller conference BSides, where multiple talks about improving Human Interfaces to software were presented. Improving User Experience in interacting with complex systems has become, in some cases, a matter of life and death. We don’t need, nor are we perhaps ready for, software that does everything for us. But, we do need software that makes us more effective at what we are doing.
The theme of Augmentation, that the machines are a path to expanding human capability at the individual level as much of the societal level, found expression in other talks, villages, and workshops at DefCon.
No longer the realm of pulp Cyberpunk Novels, “BioHacking” is a compound word that had received growing significance at DefCon. A decade ago, many of the talks in this area were about theoretical futures and risks inherent to them. This year however, practical applications and problems governed the tone. Discussions consisted of how to know what code is running on devices that operate under people’s flesh and “Malware that Kills”. Who is going to monitor the security logs on your pacemaker?
Humanity is a dualistic beast and the risks of increasing the amount of technology we integrate with our biology is tempered with the promise of augmenting our capabilities by doing the same. The homebrew “Grinding” community of engineers and hobbyists experimenting with their own implanted devices continues to grow each year.
Out of all of this is something that a few of us, myself included, have been predicting and pushing for over the last decade, a growing crossover between the bioinformatics and information security fields as we realize we share a lot of similar problem spaces, both mathematically and operationally.
The need for ‘expert systems’ in both areas, combined with ‘organic’, emergent, self-correcting systems is both fields have in common. It’s no surprise that practitioners in either field are starting to dabble in the other.
So, despite a Hollywood-inspired tone of machines replacing human responsibilities, the actual lesson of the massive amount of presentations, work and discussions at DefCon this year spoke a different story.
Augmenting Human ability is the way forward still. Human beings have an unparalleled ability to make massive leaps and connections in context-switching, pattern matching, and creative reasoning that software still cannot replicate – and yet there remains a vast amount of work that computational devices are exponentially superior to ourselves in. Freeing up the attention span and automating as much replicable cognition as possible seems to be the most productive way forward. Have the humans train the machine and leave them free to work only on the exception-cases that remain outside the software’s “experience”.
We are building a global information infrastructure where the integrity of that software, and our ability to simulate and predict outcomes of changes to that infrastructure, is as vital as our ability to do the same with metallurgy was in the industrial age. The difference today is that software rarely has the same immutable rules of physics and materials science. To keep up effectively, we’re going to have to invest into our capability to automate complex human experience and reasoning, to match our pure data processing abilities.
Inevitably, the path forward is not systems that can entirely replace us, but offload repetition and multiply our effectiveness with the things at which humans excel.