4 ms·

And unsurprisingly, he just lies. Here's the crux of the bullshit: "It’s about measurement (aggregate counts of impressions and conversions) rather than target

And unsurprisingly, he just lies. Here's the crux of the bullshit: "It’s about measurement (aggregate counts of impressions and conversions) rather than targeting." No new technology is needed for measuring impressions and conversions. Your ad server can count impressions, and a simple referrer URL can track conversions. An Apache 1.0 server could handle this 30 years ago. What the advertisers want is all sorts of other info on you, whether you end up as a customer or not. And they and their willing accomplices, now including Mozilla, ardently frame this as a necessity to keep the ad-supported content industry alive. What they never mention is that industry was thriving with the bare minimum of data they had 30 years ago, and more to the point, thrived with NO DATA AT ALL for over a century before it moved onto computers. Do you think Macy's got any data on "conversions" when they listed their sale prices in the local paper? Yet somehow they were happy to keep paying for it, without any apocalyptic screeds about the imminent "death of newspapers" if we didn't start making customers record their unique customer ID before opening the paper or entering a retailer's store. The truth is quite contrary to the ad industry's narrative- it's clear the shift of power balance in favor of the advertisers has been associated with all the industries dependent upon it getting worse.

nine_k 2y ago
To be honest, Macy's a century ago of course wanted data on conversions from newspaper ads. They could ask some customers whether they had seen the ad prior to going to shop. They could print coupons that you need to physically bring in to get a deal. They could pay agencies like Gallup to run more comprehensive customer research. Shooting in the dark is a so-so strategy, and every business strives to do things in a more cost-efficient way. (Corollary: if you want businesses to stop building dossiers on you, you have to make it unprofitable.)
Vinnl 2y ago
Disclaimer: I didn't work on this and I'm not an expert on the topic, but... As I understand it, it's not so much measuring how often a particular ad was seen, but how often the people who converted have seen a particular ad before, without anything about the people who did. Conceptually, it's similar to cutting out a coupon from a newspaper to get a discount; the store then knows that that newspaper ad led to the sale. (Though of course, in the coupon case they can also see which customers saw those ads.)
- authorfly 2y ago
  It's not similar to cutting out a coupon. It's placing a token with a permanent bluetooth tracker on every person that happens to use the nearby convenience store (firefox) without their knowing and having your staff pick up the tokens of those that walk into the paint shop with bluetooth connections. The coupon analogy falls down anyway - Coupons are loss-leaders designed to impact and increase future spending trends usually, and typically provide a discount others cannot access, or a discount upon a known previous price point - in other words some type of (possibly spurious) discount. Nobody leaves without a tracker, the description is vague and the direction of travel for purchasing becomes more Ad-laden. Company conversion and profit rates going up means the consumers are paying more than the otherwise would, either in price (profit) or volume. Can't you see how this can only be worse and is nothing like a coupon? Advertising is not a net 0, people who see more advertising spend more of their money. This is correlated with quality of life and happiness negatively. Furthermore, if you have libertarian tendencies, advertising makes unknown companies like startups less appealing to consumers due to primal branding psychology, and stops them challenging incumbents. It's a net negative. It's worse for the consumers. What's the primal reason you are contributing to this thread? To defend someone and their policy - or to avoid some other fear?
  - Vinnl 2y ago
    My reason for contributing was that I think there's a misunderstanding if someone thinks it's targeting rather than measurement, which the comment I replied to stated. There's a separate discussion to be had about whether advertising in general is a good or bad thing, whether Mozilla should be actively fighting it, and if that would even be viable, but I'm not too interested in that at the moment.

nine_k 2y ago

To be honest, Macy's a century ago of course wanted data on conversions from newspaper ads. They could ask some customers whether they had seen the ad prior to going to shop. They could print coupons that you need to physically bring in to get a deal. They could pay agencies like Gallup to run more comprehensive customer research. Shooting in the dark is a so-so strategy, and every business strives to do things in a more cost-efficient way. (Corollary: if you want businesses to stop building dossiers on you, you have to make it unprofitable.)

Vinnl 2y ago

Disclaimer: I didn't work on this and I'm not an expert on the topic, but... As I understand it, it's not so much measuring how often a particular ad was seen, but how often the people who converted have seen a particular ad before, without anything about the people who did. Conceptually, it's similar to cutting out a coupon from a newspaper to get a discount; the store then knows that that newspaper ad led to the sale. (Though of course, in the coupon case they can also see which customers saw those ads.)

authorfly 2y ago
It's not similar to cutting out a coupon. It's placing a token with a permanent bluetooth tracker on every person that happens to use the nearby convenience store (firefox) without their knowing and having your staff pick up the tokens of those that walk into the paint shop with bluetooth connections. The coupon analogy falls down anyway - Coupons are loss-leaders designed to impact and increase future spending trends usually, and typically provide a discount others cannot access, or a discount upon a known previous price point - in other words some type of (possibly spurious) discount. Nobody leaves without a tracker, the description is vague and the direction of travel for purchasing becomes more Ad-laden. Company conversion and profit rates going up means the consumers are paying more than the otherwise would, either in price (profit) or volume. Can't you see how this can only be worse and is nothing like a coupon? Advertising is not a net 0, people who see more advertising spend more of their money. This is correlated with quality of life and happiness negatively. Furthermore, if you have libertarian tendencies, advertising makes unknown companies like startups less appealing to consumers due to primal branding psychology, and stops them challenging incumbents. It's a net negative. It's worse for the consumers. What's the primal reason you are contributing to this thread? To defend someone and their policy - or to avoid some other fear?
- Vinnl 2y ago
  My reason for contributing was that I think there's a misunderstanding if someone thinks it's targeting rather than measurement, which the comment I replied to stated. There's a separate discussion to be had about whether advertising in general is a good or bad thing, whether Mozilla should be actively fighting it, and if that would even be viable, but I'm not too interested in that at the moment.