Odds and ends.. Coffee by juan manjarrez from the Noun Project

After Neutrality

I have some scattered thoughts after the FCC did what the FCC was going to do: unshackle your ISP from the hardships of supplying internet access without pernicious bullshit.

Yes, this is a speculative exercise. In attempting to be realistic, I’ll stick with tactics ISPs have tried before, the publicly available evidence of their motivations, and the technical limits and consequences of those things.

Comcast wants to block BitTorrent because they think it’s for pirating movies.1 Beyond that, it’s unlikely for them to block sites their customers want outright.

The ISPs’ goal is to squeeze more money out of people. If I build a popular VR app and my customers are frustrated by the performance, they can sell me a peering agreement and sell my customers Xfinity 360—a technically dubious upsell that supports VR. If I have no customers because they can’t even download my app, there’s no money to squeeze.

The least vulnerable products to shenanigans like these will be the ones that run over existing channels and don’t depend on speed because they work offline: email newsletters, ebooks, podcasts, downloaded digital music. If you throttle a podcast download by 50%, would anyone even notice?

Emerging technologies like VR or 4K video that use a lot of bandwidth are most vulnerable. That’s in part because they depend on extremely low latency, and in part because they’re new and it’s hard to prove they shouldn’t work over the regular pipeline.

Likewise, if you work from home, it might be in the ISP’s interest to throttle VPN or video chat software to persuade you to upgrade to their Professional tier. What makes this different from past tiering schemes is its not service agnostic. They can explicitly target needs with elastic pricing: Netflix can stay fast because consumers won’t accept an upsell and Netflix already bought a peering agreement, but the various communication tools you depend on for work will be slow until you upgrade.

Any ad-dependent products face risks even if they’re not the main target. Most text-focused news sites are already slow and depend on a kluge of moving parts. Blocking or throttling one component of the system can grind them to a halt.

For developers, it’s a good time to learn about Progressive Web Apps. If the network might be hostile, the less data you need to send over the wire, the better.

Data caps are a possibility. So are apps and sites that are exempt from data caps. This has already been widely discussed: Netflix and HBO can afford it2 but it raises the barrier to entry for small businesses.

And as usual, there’s bad news for privacy. One of the few things we know is that ISPs want in on the programmatic advertising market. Now they can add scripts or cookies to any insecure page you visit. It’s not clear of me if they could do that with HTTPS pages. I’d need to spend more time digging into this before I can write about it coherently.

You’ll notice that none of this is apocalyptic. There’s no realistic reason to think the web is dead. That doesn’t make it acceptable. It’s still a massive giveaway to monopolists, and they wouldn’t have asked for it if they didn’t intend to take advantage of it.

The rest of us get stuck with the check.

See All Writing