Slashdot Mirror
Digital Ads Are Starting To Feel Psychic (theoutline.com)
It seems like everyone these days has had a paranoiac moment where a website advertises something to you that you recently purchased or was gifted without a digital trail. According to a new website called New Organs, which collects first-hand accounts of these moments, "the feeling of being listened to is among the most common experiences, along with seeing the same ads on different websites, and being tracked via geo-location," reports The Outline. The website was created by Tega Brain and Sam Lavigne, two Brooklyn-based artists whose work explores the intersections of technology and society. From the report: "We are stuck in this 20th century idea of spying, of wiretapping and hidden microphones," said Brain. "But really there is this whole new sensory apparatus, a complicated entanglement of online trackers and algorithms that are watching over us." It is this new sensory apparatus that Brain and Lavigne metaphorically refer to as "new organs," as if the online surveillance framework used by social media platforms like Facebook has somehow transfigured into a semi-living organism. "These new organs don't actually need to listen to your voice to know that you like Japanese knives," Lavigne told me. "They actually have ways of coming to know things about you that we don't fully understand yet." In other words, these new methods of data collection have become so uncannily accurate in their knowledge of you as to occasionally feel indistinguishable from actual ears listening in on and understanding intimate conversations.
There are a few things that we do already know about these new "organs" of data processing, as defined by Brain and Lavigne. We know, for instance, that they have an insatiable appetite for personal data. They gather this by first tracking online activity, which is enough to tell them what people like, what they search for, what they listen to, what they read, where they're walking for dinner, and also, worryingly, who their friends are and what they like, read, purchase -- data that is gathered without their awareness. But, then, the organs also gather information purchased from commercial data brokers about people's offline lives, like how many credit cards they own, what their income is, and what they purchase when they go grocery shopping. And all of this information is triangulated with friends' data, because if they know what those dear to you are buying -- a Japanese knife, for instance -- there is a good chance that that person will be interested in that very same thing. The new organs process this enormous amount of information to break you down into categories, which are sometimes innocuous like, "Listens to Spotify" or "Trendy Moms," but can also be more sensitive, identifying ethnicity and religious affiliation, or invasively personal, like "Lives away from family." More than this, the new organs are being integrated with increasingly sophisticated algorithms, so they can generate predictive portraits of you, which they then sell to advertisers who can target products that you don't even know you want yet.
There are a few things that we do already know about these new "organs" of data processing, as defined by Brain and Lavigne. We know, for instance, that they have an insatiable appetite for personal data. They gather this by first tracking online activity, which is enough to tell them what people like, what they search for, what they listen to, what they read, where they're walking for dinner, and also, worryingly, who their friends are and what they like, read, purchase -- data that is gathered without their awareness. But, then, the organs also gather information purchased from commercial data brokers about people's offline lives, like how many credit cards they own, what their income is, and what they purchase when they go grocery shopping. And all of this information is triangulated with friends' data, because if they know what those dear to you are buying -- a Japanese knife, for instance -- there is a good chance that that person will be interested in that very same thing. The new organs process this enormous amount of information to break you down into categories, which are sometimes innocuous like, "Listens to Spotify" or "Trendy Moms," but can also be more sensitive, identifying ethnicity and religious affiliation, or invasively personal, like "Lives away from family." More than this, the new organs are being integrated with increasingly sophisticated algorithms, so they can generate predictive portraits of you, which they then sell to advertisers who can target products that you don't even know you want yet.
Advertising stopped working on me about 25 years ago. The internet CURED me of being able to watch advertising. My computers have Every form of ad blocker known to man on them. And any adds that DO get through, go on the "Do not buy" list. because they "Annoyed" me. GOOD LUCK ADVERTISERS ;-P
It seems like everyone these days has had a paranoiac moment where a website advertises something to you that you recently purchased or was gifted without a digital trail
That doesn't feel like someone reading my mind at all. To me it feels more like someone peering in my windows and following constantly.
I'm pretty sure most people find it just as creepy as I do, even non-technical people I know have mentioned this un-prompted and also that they found it creepy.
Companies have to be really careful using techniques like this, because they often fail in horrible ways that paints the company with a brush they would not want if they knew.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
I keep hearing Amazon and Google and all of these companies bragging about how incredible their AI are and all of these places telling us how unbelievably accurate their ad tracking and advertising methods are for targeting an individual.
Then I go online (to Amazon or anywhere else) and the ads I get are for the thing I *just* fucking bought yesterday.
Hey, dipshit... how many 65" widescreen HDTVs do you think I'm in the market for this week? The time to try and sell me one was *before* I bought one yesterday. Not after.
Sorry, but AI has yet to be more than a bunch of "if then else" loops. I don't give a fuck. Bots aren't taking over jack shit in this century when they can't even figure out when I'm less likely to buy a $5k tv.
Google Analytics is what makes things look "psychic".
I was wondering why my iPad, which doesn't have any blockers, always shows ads for the last things I shopped for from companies I shopped at.
Common denominator that made the ads look "psychic"?
Google Analytics.
Ghostery and uBlock are pretty good.
These days, 1% is good
In a previous life (Xanaro), we were doing bound-in ads in a print pub, and knew we would have succeeded sy a 3% response rate.
These days, advertisers struggle for 1%, which means they're doing something rather badly
davecb@spamcop.net
Er, his 'software' is an app that downloads a list from multiple sources, sorts and merges and spits out a host file.
Or so I'm told. By a friend.
It also has the capacity to add your 'favourites' to the beginning so that they are found more quickly.
All the increases in performance that he claims are based on making a lookup from a host file vs making a DNS request, or by comparing resource usage from a host-only solution vs a mix of browser extensions and/or AV.
His claims hinge on some very specific comparisons with shifting goalposts. A lookup from a host file is milliseconds faster than querying DNS, but the savings you make are swamped by the minutes you need to spend tending the host file each time you browse - opening his app, downloading, sorting, merging, writing etc. Using hosts alone uses considerably less resources than using browser extensions and or antivirus products, but a host-only solution isn't recommended by anyone - even he won't make that claim any more.
He posts some comparisons of browsers using (IIRC AdBlock) vs not. It's a bit dated, but it's of the order of low triple digits of MB. When most machines, these days, are running with GBs of RAM, a couple of hundred MBs is single digit percentage of system resources at maximum.
The entire point of computing resources is to use CPU, RAM and disk to perform tasks that would otherwise have to be performed by me. Using RAM to run a browser app to make administering a blacklist is exactly what I want to spend resources on. Saving those resources by increasing manual intervention is losing sight of why those resources are useful.
If you are on a system that's constrained in some fashion, then perhaps a lower level and lower resource solution like using the host file makes sense. It's hard to to find real world cases. If the resources are constrained enough, why is it being used for general browsing, and if it is being used for general browsing, why not look at blocking at the router/firewall than on the device? For most people it's a false economy. As a solution, blocking based on a host file was more useful when people had a single internet connected device that was a lot less powerful than today.
Using a host file is a blacklist. From a security perspective, a black list is of the form 'allow all and block [list]'. This is inefficient and ineffective. You are better from an efficiency and effectiveness to whitelist - 'allow none except [list]'.
You mention uBlock, have you taken a look at uMatrix? By the same author, it has a fairly intuitive interface that lets you block/allow by site and by resource requested (script, css, image, etc.). You get to choose what, apart from the domain and child domains that you visit are allowed to be called. That can either be a temporary permission, or saved as a rule. No association with the maker - I've used and like uBlock but have mostly moved to uMatrix.