← Back to Posts

Not the government, not Facebook: Charles Duan on who should write the standard

Sean Roberts•December 5, 2020•4 min read

Saying that Facebook should work like email is the easy half of the argument. The hard half is the protocol in between, and who gets to write it.

Charles Duan of R Street started with a warning about email itself. It is simultaneously very simple and very complicated. The simple part is one computer saying to another, I've got a message, you should pick it up, in terms both sides understand, so each knows the message arrived and where it goes next. The complicated part is everything piled on since: HTML messages, attachments, proving that a message is not spam, a stack of languages and protocols most users never see. Port the idea to social media and the same thing happens. The concept is a front end asking Facebook for your friends and your latest posts. The reality is authentication, protecting the servers from being flooded by every new client, and a hundred cases nobody has thought of yet.

One of the important things that we've learned in developing interoperability in lots of other contexts is that you don't want to just rely on the government to come up with a standard for these sorts of things, or just hope that Facebook comes up with the best standard. You generally want to have consortia of people, technical experts who really know how this technology works and understand all of the ins and outs.

His examples were the ones that built the internet. Email came out of the Internet Engineering Task Force, which is, in his words, just a bunch of volunteers who are experts in coming up with these protocols. The web has the W3C. Both work incrementally, and neither waits for a statute.

The line I wrote down was about scope. The goal, he said, is not to say what the standards are but to say how we should come to them. How is the process run, who gets a seat, and how do you make sure it is not stacked in favor of the companies that would rather the whole thing never happened? Those are difficult questions, but they have been answered before, and the answers are reasonably good.

Anyone who has spent time inside an open source foundation will recognize this. The arguments are rarely about the code. They are about who gets a vote, who can block a change, and what happens when the biggest contributor disagrees with everyone else. A process that gets those questions right can survive a lot of bad first drafts. One that gets them wrong produces a standard nobody implements.

I pushed on enforcement. The IETF and the W3C grew up without much government involvement. Would this need an FTC or FCC hammer behind it, to make sure the platforms actually followed the result? Charles answered with history first. There was an earlier attempt at a standard for exchanging social data, and it never took off, largely because the company behind it never built a social network that took off either. A standard with no platform behind it goes nowhere. And most of the standards we rely on were written before any single company was running the show, which is not the situation now. Convincing a dominant company to open itself to competitors will be harder than it was for email.

He still did not think it called for a strict mandate. Companies get real benefits from interoperability. When they open their APIs, third-party apps appear, and people find uses for the platform nobody planned, some worth more than the original. Some incentive is probably needed, he said, but likely much less than you would expect. The incentives already push in both directions, and the job for anyone who wants a more competitive internet is to use the ones pointing the right way.

This matters now because the idea has reached Congress. The House antitrust subcommittee's report in October recommended interoperability and data portability for the dominant platforms. It leaves open who would write the specifications, which is exactly the gap Charles was describing.

So here is the question I would put to anyone drafting a bill next year. Are you writing a standard, or writing the rules for how one gets written? On Charles's reading of history, only the second has ever worked.

From Lincoln Shorts, the interview series I hosted for Lincoln Network, recorded 5 December 2020.

Share this post