they always told me I had a face for radio
07 June 2019
On a podcast with Katherine Druckman and Doc Searls at Linux Journal.
Some of the stuff we talked about...
advertising and signaling
Doc: The latest Firefox does block third party tracking. And that's a huge move. But there's a drift of the other browser makers in their own different ways are doing things. So and that's part of a larger trend. So maybe you could just sort of unpack that for us.
Me: Apple Safari got out ahead of the rest of the pack and a lot of interesting ways. And people certainly joke about the $999 monitor stand. But it's pretty clear that the people who buy that kind of high end equipment are also the people who are least trackable by conventional adtech. So if you want to reach the kind of people who can either afford a tricked-out Macintosh, or have a job where their employer will buy them one, then you really have to think about how do we place advertising in such a way that it doesn't depend on the kinds of creepy tracking that the Apple Safari developers have put so much effort into avoiding.
Doc: Katherine, maybe I'm wrong about this. But didn't we finally put the nail in the lid of the coffin of Google Analytics for our own website, when we found that as Don was just saying our readers block tracking for the most part?
Katherine: It was it was so far from being accurate that it was not useful anymore. Who doesn't block tracking? I'm kind of throwing this out there, but I want to say a good 60% of our traffic was blocking it.
Me: I've seen some numbers from web developer sites, and from blogs that focus on web development, and those are often showing a 30 to 40 percent block rate. But it's really interesting that Linux Journal readers are coming in at 60.
And so there are, right now, marketing organizations that are going out and trying to reach the kind of people who buy virtual private servers or software as a service products or developer tools. And if you do conventional data driven marketing, when you're going after that kind of audience, then you're really going to get a lot of fraud bots, and your marketing operation is going to be making decisions based on what bots like to read, not so much what those what those those high tech, or highly protected, users are interested in.
Doc: Have you seen a sign of that knowledge being generalized, beyond, you know, people like us talking about it, where it's becoming obvious to some people in the marketing side that the most valuable people are going to be the ones that are most protected?
Me: I'm cautiously optimistic because of the change over from targeting millennials to targeting Generation Z. I don't know if you've seen it, but marketing thought leaders are changing up all their slides. And they no longer say the millennials are different. Now millennials are boring, and Generation Z is all different. But it's really recycling a lot of the same millennials material. So there's a nice niche opening up for a marketing thought leader to scrap the generation-driven slides that everybody has already seen, and become the marketing thought leader of the tracking protected segment.
Doc: Today you get injected with all these third party cookies that get arranged like a DNA string that gets presented at every site you go to subsequently. Do you see any hope for either ending that or modifying it with the kind of things that are going on now or blocking the third parties? I guess the question is, are we stuck?
Me: I don't think we're stuck. I think
that a lot of the talking points that we're getting
from adtech and martech today are very
similar to what email spammers were coming up with in
the early days of spam filters. The early spam
filters, of course, were done by technical early
adopters, the kind of people who read Linux Journal
and know how to write .procmailrc files. And when
those people started rolling out their original,
simple spam filters to the the less Internet skilled
users, the spammers started started saying, hey,
wait a minute, users like getting messages about
opportunities for great savings on HERBAL VIAGRA, or
whatever the latest spam campaign was.
The message from the email spam scene was really that privacy nerds are less in tune with the preferences of regular users than we the spammers are, so you should pay more attention to what spammers want, and less attention to what spam filter developers think is the right thing to do. And we really saw that not come across very well as as email spam moved from a niche issue for people who had had their email address out there for a long time to being a mainstream day to day annoyance. The general population of users turned out to be more like the privacy nerds than like the the way that the spammers predicted they would be.
Doc: I'm thinking that you could go into how people actually have a pretty good sense of behavioral economics, that they're good behavioral economists to some degree if you want to have any money left. For us, because obviously, we're we're all walking through a minefield that all of us understand in somewhat different ways. But I think one of your points is that people do become pretty adept even if they don't fully understand what's going on behind the surface.
Me: And you can't make the optimal decision for most of the decisions that you have to make in your economic life. I don't have the time to buy the optimal pair of socks. So what is a set of tools that I can use to get an adequate pair of socks in the amount of time that I have to make that decision? And of course, advertising isn't the entire story behind building brand reputation. But it sets up part of the information that people can use to evaluate a product or, or figure out the reputation of a brand. Brands are really interesting. Brands are a cognitive hack that uses our brains hard wired circuitry for evaluating each other's reputation.
What would be a really interesting piece of research would be comparing TV ad spend on cars to later Consumer Reports ratings. Are TV ad budgets a reliable leading indicator of how well that car actually does in the independent test? It's kind of like when Rory Sutherland at Ogilvy compared advertising to someone betting on their own horse at the track. If you go to the racetrack and see that the horse's owner is betting heavily on that horse, they've probably got more information than you do.
Open source and incentives
Katherine: I feel like I see growing resentment, and not just individual open source developers, but small companies and whatnot, just that are becoming more painfully aware of the inequality of, you know, fortunes being built off of what they perceived to be their own backs. And, and the consequences are well, unforeseen.
Me: It's not just a matter of resentment, there's also an element of risk there. One of the side effects of having good dependency management tools is that real world IT projects are building deeper and deeper dependency trees. So the success of your web site might depend on some software component three levels deep, whose maintainer is going to burn out, right as you deploy your site, but you don't know it. So there's this risk for anybody, depending on open source, that somebody's choice to stick with it just doesn't pencil out. And the people who you need to stick with it in order to have your thing be successful might be several hops away. There's still no good way of getting that information, propagating the developer pressureMaybe a market mechanism (PDF) could help? from the developer who's experiencing it to the leader of the project that depends on their work.
Bonus links
Mark Ritson: Binet and Field’s research may not be perfect but that doesn’t make it wrong
Why we're relicensing CockroachDB
Digiday Research: Most publishers don’t benefit from behavioral ad targeting
When it comes to privacy, default settings matter!
Is High Quality Software Worth the Cost?
AT&T Explains Its Surveillance Advertising Hellscape
Privacy Preserving Ad Click Attribution For the Web
The Best Ideas Are The Ones That Make The Least Sense, by Rory Sutherland, Entrepreneur
