My friend Aaron wrote me in response to my piece Tower of Babel. It was thoughtful and he gave me permission to copy it here.

Regarding filter bubbles: you addressed information consumption but not information production.

Have you ever been to an out-of-control town-hall meeting? You know: one of those nightmare evenings on a hot-button issue that's poorly moderated, overly emotional, under-factual, with low information quality, self-promotional speakers, hardened political positions and an absolute din of side conversations? Such things happen in the real world, too, and that's what the current online state of affairs reminds me of.

Part of the reason for this is that Facebook, Twitter and online-comment forums have driven barriers to entry to near zero for OPINION production and broadcasting. And in the attention economy, a frictionless virality is the Holy Grail for any profit-maximizing firm. Nothing drives virality like outrage. And nothing drives opinion production by Homo sapiens sapiens like the opportunity to be liked a whole bunch instantly. Facebook has pulled all the design stops to encourage attention whores to feed the social graph and has used its market power to usurp news traffic while avoiding a concomitant journalistic responsibility. Twitter => 140 characters => anyone can tweet => often. And of course attention-economy profits are further increased for firms that can also minimize their moderation and editorial costs via automated algorithms and user-operated filters.

Is the end result at all surprising?

Add to these things the explosion of images, meme gifs and video -- further developments that change the nature of what is communicated and the relative densities of information vs. emotion; we have basically experienced the TV-ification of a once text-based Internet. You will recall that much of the BETTER-WORLD HOPE of the Internet 20 years ago was projected from the idea that the web was a super-fancy set of linked-up electronic books and that email was the means for any scholar to write to any other for the advancement of knowledge [imagine any suitable emoji here].

Meanwhile, journalism's FACT-finding remains expensive while its business model has been eviscerated. And: few facts go viral.

The results for our political discourse: 1) the raw information published today is lower average quality; 2) the mechanisms for information refinement are dramatically weaker; 3) the incentive to be heard drives a Darwinian process towards particular signals that cut through whatever present background exists.

I think adding these production-side aspects to the consumption-side points that you made provides a more complete picture. Also important (to my eyes anyway) is that we seem to be entering a period of some paradigmatic uncertainty in the United States and globally; such uncertainties would stress political discourse no matter the technological backdrop. It might only be possible to understand some developments with hindsight.

A question for readers: what systems, in any realm, at any point in time, rewarded quality over provocation, for a mass audience? I can think of systems that have done so at very small scales, but when the problem to be tackled has the advantage of scale, the solution may need to work at that same level.