“I’m working on advertising at Google.”
“Why? I sincerely do not understand that anyone can do this. “
Someone recently asked me: “Why am I working on an advertisement?”
I wanted to write something more solid than just a comment… (Even though this is my professional topic, this is a personal post and I only speak in my own name.)
One answer is what I earn to give: half of what i earnI give to the most effective charities I can find and the more I earn the more I can give away… However, this is not a complete answer, since when people ask me about it, they usually assume that they view ads (or perhaps online ads) as bad, and the question is more like “Why did you decide to work on something bad?”
So: what is advertising good for? I mean, isn’t it annoying when sites show you ads instead of what you want to read? The question is, what is the alternative? I see two main funding models:
- Paid access (paywall). You pay with your money.
- Ads. You pay with your attention.
It is also possible to fund projects with donations or as a hobby, but making most of what is readable takes more money.
(I use the Internet-specific term “paywall” to refer to the general concept of “paying money for access”: buying books, paying admission tickets, subscribing to streaming services, etc.)
Both paid access and advertising have a number of advantages and disadvantages. Some of them differ depending on the medium: books are expensive enough to print and cannot be funded by advertising; analog radio so simplethat paid access would require draconian legal force. However, I believe that online advertising is better suited for two reasons:
- Minimal friction. You can follow links from site to site without obstacles. You don’t have to decide which sites to subscribe to. If someone sends you a link to an article, you can read it.
- Irregularity. Paid access, like other fixed costs, is regressive: a newspaper at a cost of $ 220 per year is much more expensive for a person earning 10 thousand dollars than 100 thousand dollars.
You can reduce friction by doing a merge: you subscribe to a streaming service, and then you can watch (or listen to, or read) anything in their collection. This approach has its advantages, but it does not work well for articles. Web browsing works best when people can read and share anything without a subscription (“sorry, this article is for Conglomerated Media Group subscribers only”). To substantially eliminate merger tensions, you will need to limit yourself to a small number of subscriptions, which will then give these organizations tremendous (and dangerous!) Power.
Micropayments can potentially resolve these frictions in a decentralized way, I hope this will happen. On the other hand, this is a really difficult problem: people were working on it more than 25 years ago (Digital’s Millicent). There were many offers and startups, but nothing came of it.
However, even if we could solve the problems with payments, we would still be stuck with the main problem, which is that some people have much more disposable income than others. Universal basic income would help, and I strongly support him, but I don’t think it will be politically feasible anytime soon.
So: advertising. Funding for the open web.
Or perhaps: advertising is better than paid access.
However, I don’t want to take advertising too superficially: there are many bad things in online advertising today. For example, advertisers don’t have enough incentive to limit bandwidth usage or publishers to avoid annoying ads. But the biggest issue people raise is the privacy impact of targeted ads.
Most foods work better for some people than others. If you try to sell fishing bikes, few people will be interested and you will mostly waste their attention. This means that ads cost a lot more if you can show the right ads to the right person.
One way to do this is to place your ads in locations that are likely to have disproportionately interested people. Model Railroad Advertising on Simulation Railroad Forums, Sponsored Products on Amazon, Booth at Trade Show. This works great if you want to blog about cool new credit cards, but what about all those sites that don’t have a strong commercial connection?
Most of the advertising on the Internet today is targeted at previous views. When I wrote all these posts about cars, I visited many car sites, and then I saw lot ads about cars on other sites… I didn’t buy a carbut the advertisers were right that I was much more likely to buy a car soon than the random person.
Historically, these ads have been created using third-party cookies. When I visited one of these car sites, they probably put some HTML on their page, like this:
My browser made a request for this image and got an invisible “tracking pixel” with something like:
The vendor probably kept a note like this:
Later, perhaps I visited a website about flowers and they sent me:
This time, my browser already had a cookie for adtech.example, so it was included in the request:
This allows the vendor to update their entry for me:
interests: cars, flowers
Some time later, I read something irrelevant on a site that contracts adtech.example to serve ads. My browser is sending an ad request and my cookie is enabled. The seller is running an auction, bidders are especially interested in paying to show me car ads (more profit than flowers), and I get ads about cars.
This model has several serious privacy flaws. Typically, a salesperson doesn’t just know you’re interested in cars, they get the full URL of the page you’re on. This allows them to compile a fairly complete picture of all the pages you have visited on the Internet. They can then link their database to those of other vendors and get even more coverage.
The situation began to change in 2017 when Safari announced “Intelligent Tracking Protection”. “This is the first of highly many rounds iterations, about years ago, he led Safari to complete blocking of third-party files cookie. Behind him followed Firefox, and Chrome announcedthat they will do it too.
Well, sort of. Chrome’s announcement was a little more streamlined:
After initial dialogue with the web community, we are confident that with constant iteration and feedback, privacy and open standards mechanisms such as the Privacy Sandbox can keep the ad-supported network up and running so that third-party cookies are out of date. As these approaches will meet the needs of users, publishers and advertisers, and we will develop tools to address workarounds, we plan to phase out support for third-party cookies in Chrome. We intend to do this within two years.
The idea is to create browser APIs that allow such well-targeted ads without sending your browsing history to advertisers, and then get rid of third-party cookies.
One of these suggested APIs is TURTLEDOVE… It allows the advertiser to tell your browser, “Remember that I know this user is interested in cars,” and then “show this ad to users who I said were interested in cars.” Because the browser stores this information and handles bids, reports and ad serving very carefully, it prevents advertisers from knowing which sites you visit or sites which ads you see.
I figured out how ads can use TURTLEDOVE by helping create open-source plain-JS implementation for testing and experimentation and suggesting ways to improve the API (# 119, # 146, # 149, # 158, # 161, # 164). I think this is largely why I have been blogging less lately: the implementation of these ideas draws energy from the same source.
Advertising is how we finance the Internet where you can navigate freely from site to site, and my main job is to help figure out how to move ads to less powerful, more private APIs. While I think the vast majority of my altruistic influence comes from donations, I don’t think my advertising work is harmful.