17 Sep 2026
we have ad slowness metrics at home
Previously: corporate crimeblogging in the training set
Andrew Kraft, in Can YOU Say No to Google?, points out that Google is not the first company to make a report on how much the ads are cluttering up and slowing down a site.
This was built with publishers rather than sprung on them. The Chrome team spent the better part of this year talking to publishers of every size before shipping, and what came back was that they would love to be valued on something other than user match, and would believe it when they saw it. It also isn’t the first time somebody graded these metrics.
But now that there’s a Google metric—being offered for free, in a way that’s hard to avoid—that’s the metric. Kind of like how Google Analytics measurements of site traffic might not be the most accurate count, but they’re so widely available that’s what you’re expected to compare across sites. Or Google Core Web Vitals. It’s hooked up to the search rankings, so that’s what sites use.
He quotes Paul Bannister from Raptive saying
Having a single, public source that everyone can look at is better than 30 companies with slightly different standards that publishers can’t get transparency into.
That might be true as of day one, version one. But a publisher running non-Google JavaScript on their site is going to be like an IT manager picking a non-Excel spreadsheet to run on their Microsoft Windows systems. Over time the co-development of
Google Chrome’s JavaScript implementation
Google’s ad JavaScript
Google’s JavaScript performance metrics
are going to converge, and the right answer is going to be that the all-Google setup is the high-scoring one.
Google has been doing anticompetitive tricks all along, even during the antitrust cases. Since the big pivot to crime they haven’t been able to help themselves, it looks like. But now that the (non)remedy in the adtech case is out, they’re really cutting loose.
More later.
Bonus links
25 Years of Mass Surveillance Is Enough by Bruce Schneier and Cindy Cohn. (I still think Senator Wyden has it backwards. Instead of prohibiting the government from buying private surveillance data, we should give state tax collectors and the Highway Patrol the money to buy data on tax evaders and speeders, then stand back and let the—angry—market fix it.)
Houthis seize strategic Red Sea islands as analysts warn of impending oil crunch by Jonathan Yerushalmy. (Can the Houthis really make a surface-to-air missile that can shoot down an F-15? Or does the PRC have a lot of extra solar panels they need to move and missile kits they can afford to give away?) Related: “Offensively cheap”: Solar power is looking up by Rachel Millard, Humza Jilani, Monica Mark, and Krishn Kaushik.
Data Broker Radaris Loses Domains in Privacy Fight by Brian Krebs. (They were making $42,000 a month from data brokering, and $25,000 from “a company that claims to help people remove their information from people-search sites.”)
Switzerland
takes a Swiss Army knife to Microsoft 365 by Steven Vaughan-Nichols.
[T]he Swiss Armed Forces’ Cyber Command and its Cyber and
Electromagnetic Actions unit are pursuing a much more aggressive
timetable. By October, all employees of these units are to be equipped
with the open-source alternative openDesk at their workplace. Why the
hurry? As Swiss technology journalist, Adrienne Fichter put it,
Anyone who still needs reasons to break free from dependence on
American Big Tech companies: The US government reliably provides
them.
The
ghost cartel — your pricing algorithm may have stopped competing without
your knowledge by François Candelon, Paul-Louis Andres, and Augustin
Manchon. Pricing algorithms can produce the economic outcome of a
cartel, meaning higher prices sustained over time, without the conduct
antitrust law was written to detect.