Post-Verbal Language

James Beshara has a really interesting read on how communication will change and evolve in a post-verbal world, namely one where human/brain interfaces like Neuralink can more directly transmit thought between people than the medium of language allows today.

After reading the essay I wonder if people's thoughts or the neural pathways they activate, if they could be directly transmitted into another brain, would actually make any sense to someone else with a unique internal set of pathways and framework for parsing and understanding the world. The essay assumes we'd understand and have more empathy with each other, but that seems like a leap. It seems likely the neural link would need it own set of abstractions, perhaps even unique per person, similar to how Google Translate AI invented its own meta-language.

Today idea-viruses that cause outrage (outrageous?) in today's discourse  have been weaponized by algorithms optimizing for engagement, and directly brain-transmitted memes seem especially risky for appealing to our base natures or causing amygdala hijack. But perhaps a feature of these neural interface devices could counteract that, with a command like "tell me this piece of news but suppress my confirmation bias and tribal emotional reactions while I'm taking it in."

Panels Finished

My last panel was yesterday at 5 PM so now I’m taking as much time as I can to catch the panels that are left and learn a few things. Speaking was more stressful than I expected but most of the feedback was good. I still need to check out the web feedback though, as people are probably more candid online than when they’re shaking your hand.

Usable Security

Usable Security is a new blog about—you guessed it—the intersection of usability and security. This comes up every few weeks since I improved the error messages on the WordPress login (and bbPress) to specify which part of the login was mistaken, the username or the password. Security folks see this as a problem because you’re revealing more information but I see making the error message more generic as premature security optimization. Plenty of systems where login names are public or easily discoverable, such as Yahoo, Gmail, Hotmail, most email systems, and so forth, seem to be doing just fine.