Facebook’s claims about the Ad Observer are wrong
blog.mozilla.org
blog.mozilla.org
I'm sure Adidas would be very interested to know with 99% confidence that Nike's ads are currently focused on people in X, Y, and Z cities in the southwestern US. And as we've seen with Cambridge Analytica, its somewhat naive to trust researchers vacuuming data to do so in good faith.
Not even to say how other adtech companies might be interested in data about how Facebook is doing ad targeting.
Trust is difficult, but it's mandatory for a functioning society. We have tools to appreciate trustworthiness, transparency is one of these tools. Not being for profit is another good indicator.
B. Users are ad targets. They do not pay for the privilege of being targeted. It's "free".
Obviously anyone falling into category A could have some legal claims against Facebook that those in category B could not. If Facebook wrongs those in category A, then advertisers can take action against Facebook. Whereas if Facebook wrongs those in category B, users cannot do much. At best, they can try to get someone else, e.g., a regulatory body, to act on their behalf. Maybe Facebbok pays a fine if some regulatory agency decides to act. Most times, nothing happens and Facebook pays nothing.
How would those in category A ever discover they were being wronged. Seems like Ad Observer-style data gathering might be one way.
Plus all those legal contracts are usually littered with words like “reasonable” which is a great term for lawyers billable hours.
I’m Danish and our laws regarding advertising may seem a little totalitarian to Americans, but here you have to be rather public about your advertising and who it targets. You obviously can’t tell who exactly sees a poster on a public wall, but you can get a general picture of how advertising is utilised.
The one place you can’t though is on SoMe because there is no record of your personal wall. Facebook at least makes overall spending public now, unlike LinkedIn and others, but no one can tell how the advertising that the money buys is targeted.
This is an issue because hearing something enough times changes your perspective on reality. But more than that, we have to take the SoMes word for it when they say they aren’t targeting children with tobacco, sugar or toy commercials, which is illegal here.
So it’s interesting that Facebook does these things to tools that might track what they won’t willingly provide.
Some of this is benign enough, like all the inexplicable videos of rollercoaster rides or the Crafty Panda nonsense. Some of it seems somewhat nefarious, like the very family-friendly, feel-good, decidedly non-political suggestions for things like Cute Cat Content at the EXTREMELY political Epoch Times.
I mean, come on. All systems can be gamed. Facebook will still, for a fee, put any content you want in front of any user you want in a completely untrackable way. And that's why they freaked out about this extension.
I gotta say, I don't believe you. I've never once seen a Suggested link that wasn't to some kind of obvious click farm. Never once. If you want to make a technical argument that Facebook drives revenue via other means than direct payments, then that seems reasonably likely (if deliberately misleading).
If you want to argue that these folks aren't spending money that ends up in Facebook's pocket in order for Facebook to put their content in front of my eyeballs... then I'm just going to laugh at you, sorry.
I mean, just on first principles your argument is ridiculous. If Facebook didn't see benefit from putting this garbage in front of me that I never click on, then why would it be putting garbage in front of me that I never click on? Clearly Facebook wants me to click on it!
Great. I have had nothing but the opposite experience. What now? If I can deconstruct your argument that easily, it must not be a very strong argument to start with.
Facebook themselves have no clue what they present to users: they don't like it when other entities start to form a general idea.
I've probably logged on to facebook fewer than 7 times, so I have very little experience with it, and so I wouldn't know.
There is literally no way to give FB money for formats outside those (with the single exception of when new formats are introduced they often run experiments with large advertisers).
If you are so sure it is paid for, it should be easy to show the steps you need to take to buy it. Where?
Maybe Facebook are being played, but not by anyone in my list of friends so they are complicit or we are expected to believe that they can't filter content by your social graph.
Potentially that's true: it does not paint a very good picture of The Social Network's ability to target ads.
More likely, Facebook have no real content for me because noone in my list of contacts posts on Facebook anymore and Facebook push this shitty content so thier pages are not bare. Perhaps one in ten is a real ad. It's still 100% low grade ads, just 90% is Facebook trying, and failing, to drive traffic for itself.
I imagine they logged me looking at the titles long enough to write them down as active user engagement.
The first "suggested for you" piece is this link[1], about the Australian cycling team's bronze medal in the team pursuit. I'm a cyclist and Australian, and can confirm this link is interesting to me. It's from SBS, which is one of Australia's publicly funded broadcasters.
It's not a paid ad.
> More likely, Facebook have no real content for me because noone in my list of contacts posts on Facebook anymore and Facebook push this shitty content so thier pages are not bare.
Exactly. If you are only seeing junk content from your friends, then the suggested content is likely to be junk stuff that is globally popular too. That doesn't somehow make it some secret paid ad, it's just junk. Same way the old Buzzfeed listicles were junk, but popular.
> If you are so sure it's _not_ commercial content you have lost the ability to distinguish real content from adverts. That's a skill 7 years olds need to develop.
You say you and your friends don't use FB much. Whereas I do, and have spend 5 or so years going very deep on their advertising work.
Perhaps I do actually know what I'm talking about here? There is literally no way to pay for "suggested content".
[1] https://www.sbs.com.au/cyclingcentral/article/2021/08/05/aus...
One of the product teams at Facebook mistakenly believes you'd be interested in using marketplace or whatever and has enough clout internally to get their stuff added to your feed.
What we need, however, is it to be able to pull data from each campaign, see what democracies they target and how many times they appear on peoples wall. It doesn’t have to go into “moksly saw this 15 times” more in the range of “in the group of x, y individuals were shown this add z times at these hours” and so on.
This would make it more akin to the advertisement data we have on almost all other mediums. I personally think we should require something similar from Google, but they manage to sneak below political and public notice enough for that to have widespread support.
Apparently not just Americans... There are around 10k Americans living in Denmark [1], and around 1.3M Danes living in the US [2]. While I am sure that everyone you talk to in Denmark will tell you they love it there, that's due to the survival bias. The immigration streams tell the true story, which is that the Danes are voting with their feet and overcoming incredible financial and cultural challenges to move their families to the US.
With that in mind, no, I am not really keen on learning about all the wonderful hurdles your government has come up with to destroy every last bit of a chance that an entrepreneur has to advertise their business. Surely, the intentions were all good, but you know what they say about the path to hell...
[1] https://www.thelocal.dk/20190402/where-do-denmark-based-amer... [2] https://en.wikipedia.org/wiki/Danish_Americans
And the history in the article goes back to the 1600s.
I think you need to reassess your sources and position.
Btw, nothing against the Danes. There's no country in this world where the same is not true. You hear about the Swiss tax laws, or the German unions, or the Italian lifestyle, or the Swedish health care. And yet, all those people are playing the Green Card lottery (just like I did and everyone else I grew up with).
There are an estimated 30 to 35,000 Danes living in the US [1] (not Americans with Danish ancestry, which seems irrelevant). By your own logic re: voting with your feet, it seems the conclusion should be the opposite.
[1]: Embassy of Denmark in DC, cited on https://wise.com/us/blog/how-to-adjust-to-life-in-the-us-as-...
Your logic is reversed. Assuming that the US and Denmark have equally favorable living conditions, and the US population is 55 times larger, there would be a 55 times higher chance of a US citizen deciding to jump the fence than the other way around. The fact that this is not happening, shows that the US is perceived as a more desirable place to live.
There are so many potential reasons; it's silly to draw conclusions about what country is "better" (on some underspecified scale) based on just migration stats.
1) The median Scandinavian citizen is living a more comfortable life than the median US citizen
2) As a Scandinavian citizen, you're more likely to want to live in the US than as a US citizen to want to move to a Scandinavian country.
If you found this exchange weird, I hope you didn't find me disrespectful or anything of that sort, but more like challenging some long-firm beliefs that have been repeated so many times that it's difficult to hear any other point of view. And if you still don't agree with me, that's totally fine - I might very well be wrong. The whole point of being on HN is to hear POVs that are contrarian and hopefully thoughtful.
Not to mention that a generally poor population are stuck in the US, while for the upper income brackets actually have it OK in the US. While in Scandinavia, almost everyone has the means to travel the world. Even store clerks, or otherwise "minimum income workers".
Continue to think the US is a dreamland while the median Dane have a much higher quality of life.
If you read the children comments, you will see a more nuanced take on those stats.
> Continue to think the US is a dreamland while the median Dane have a much higher quality of life.
I don't doubt that; the US is not optimized for the median citizen. But neither is Hacker News - I suppose a median community would be some kind of an offline newspaper with readers' comments. If that excites you, more power to you.
No one surveyed by Gallup has left because of political reasons and between 70-80% of them would prefer to return home if their work allowed it.
In retrospect, after realizing what their request was and losing some charity towards the argument, it also makes me feel weird that the first bulk of the argument is dedicated to they reviewed their own code and it totally checked out. Seems slimy, unless this blog post is by someone who is naive about tech issues
> the first bulk of the argument is dedicated to they reviewed their own code and it totally checked out
Rather, it's that Mozilla reviewed the code for the Firefox extension that was being hosted on addons.mozilla.org and found that it wasn't collecting any personal data other than what ads are being shown.
Are you referring to Mozilla reviewing the Ad Observer code? If I’m not mistaken that was built by the university researchers, not Mozilla. Also, the code is open source. As Mozilla says, if you’re uncomfortable taking their word for it, they link directly to the open source Ad Observer code so people can see exactly what it is and is not doing.
I think it’s understandable that you might be uneasy, but we can all see exactly what it’s collecting.
A company who’s ethics tell itself this is totally OK flies into a panic when researchers collect a relatively tiny amount of data on how facebook’s massive ocean of information might be manipulated by malicious outside actors to steer wider societal behaviors.
If they fly into a panic about this, it means they understand very well how dangerous or impactful even a small amount of information collection can be.
They get very upset and put together entire projects condemning companies like Apple for daring to limit their access to our devices.
I’d like to know why they believe we (society) should trust them at all if they’re not at all willing to be transparent in return.
Their data collecting has always been troubling, but I think their reaction to outside studying is very concerning and combined with their past behaviors should raise our concern levels by a lot.
I agree that's a valid interpretation of gaslighting.
Great article and definitely worth a read. https://www.nytimes.com/2021/07/14/technology/facebook-data....
Facebook can be a better place. They changed the feed algo right after the 2020 U.S. election, and for a month even the insiders called it the "nice feed" (or something like that). But that didn't last long because calm users don't rage-click.
When these problems get out of control, you get the standard line of "we are giving people new ways to connect". The concern that some people "connect" to gather at the Capitol to hang the Vice President is rather secondary.
I understand that this is not something you can do in a short amount of time; at the same time many Mozilla users have been objecting about this issue for years now.
Edit: they pre-install the Facebook Container extension while the Google Container extension is not even in the list of suggested privacy-focused extensions. If money can't possibly come from other sources, maybe they could at least do some work on looking a bit more impartial.
Edit 2: Firefox was found "phoning home" to Google after every new install - at the end of the installation process it would automatically load a page that was using Google Analytics, before the user had the time to enable any content blocker. (This was circulated on Twitter and they later removed the Analytics code).
While I wish donations could sustain a competitive browser, I'll settle for a reasonably independent one. Even if the lion's share of financing is from Google throwing them some bones in exchange for the top search provider slot.
Other browsers have yielded much larger parts of their stacks to Google.
[1] last time I've used Firefox, there was a built-in, official Facebook Container, while the Google Container extension was a non-built-in, unofficial fork, and it wasn't among the list of recommended privacy extensions.
For example, they have a built-in, official Facebook Container, while the the Google Container[1] is a third party unofficial extension which is not even among the "Recommended" extensions (even though other similar privacy-focused extensions are in that list).
In the recent Firefox versions, they are called only as containers.
Or maybe Firefox containers have changed now? I haven't used FF in a couple of years, how does it work now?
Only difference which comes with these extensions is, that they work automatically for selected domains, instead of you manually opening the site in container. Also, you need to be careful to not open other pages on container tab, which is possible. Maybe I was thinking this to be too obvious for most users, while it is not.
But yeah, there are plenty of domains if someone is really using all of them. However, it is pretty hard to miss Google login window for example.
Of course, one could speculate that one login into your Google account from your browser is enough to fingerprint you everywhere, even with all anti-fingerprint things, but that is another story.
Honestly though Mozilla should just cut the bloat that makes up the majority of their budget, bank the Google bucks while they're still getting them, and turn that into a sustainable endowment. Unfortunately the C-suites of large, failing organizations can justify larger salaries for themselves than small, successful ones I guess.
Mozilla was going to walk away from Google's millions back in 2007. Here, the Mozilla CEO said: "We've spent a lot of time and energy making sure that Google understands that it cannot turn us into an arm of Google" [0]
That was their signal to make money without relying on Google's millions. That also provoked Google to create Chrome which, er... means that Mozilla turned Google into a direct competitor all by themselves.
Now 14 years later, Mozilla is still criticising Google, YouTube whilst receiving hundreds of millions from them. Due to that u-turn, I cannot take them seriously on anything about privacy or these sort of investigations at this point.
[0] https://web.archive.org/web/20120105090543/https://www.compu...
Monetizing browser is super hard outside of ads, and all they have done is to set their default search engine as Google.
Nothing of what you write actually refutes the argument in the blog post.
Imagine if Ford bankrolled a consumer association which then proceeded to publicly criticize aspects of General Motors cars, while at the same time wanting to appear super partes. Wouldn't people be allowed to criticise that?
> Nothing of what you write actually refutes the argument in the blog post.
I'm not refuting the merit I'm refuting the fact that Mozilla is writing that post as opposed to e.g. the EFF
> more likely an ad hominem
what are you talking about? Mozilla is an organisation not an individual LOL
>what are you talking about? Mozilla is an organisation not an individual LOL
That does not change the fact that you are attacking the entity that is making the argument not the argument, which is a classic ad hominem. The nature of the entity making the argument is completely irrelevant.
And by the way, yes I am attacking the entity not the argument, you are right about that.
A counterargument is ad hominem if it rests on what it says about the arguer, not the content of the argument. It's not always a fallacy, but it is a danger sign for bad argument.
it's a case of conflict of interest, only the entity counts not the argument.
and mozilla is wrong of course
Their obliviousness to the horror that is their platform is a choice.
This is something I have considered doing before, but without a network effect it would be ineffectual.
What is not clear is how 0.1% of users publishing their Facebook passwords on Twitter or Reddit would bring about the destruction of Facebook. It looks like[1] Facebook grows by a little over 1% every quarter, so I don't understand why Facebook would collapse if a tiny fraction of their userbase started posting spam. In fact I imagine their existing spam-fighting tools would not have much difficulty handling this kind of thing.
Facebook itself estimates that 5% of its monthly active userbase is made up of spam accounts.
[1] https://www.statista.com/statistics/264810/number-of-monthly...
Their terms of service are not immutable laws. If the status quo stops working for them, they can restore old data and alter the bargain further.
For even more fun you could organize a day of protest, where everyone turned this program on, on Facebook. Though of course you'd need a backup page, Twitter, and/or Discord. Hopefully Facebook would ban the protest group and that could drive news articles covering it and get more attention and members.
Not sure how many accounts you'd need to participate - but it could go viral if other people saw posters (or maybe DMs to all your friends?) and joined the protest.
Even if it were illegal, which, I don't think it is, that might only be a reason against going it. Many protests involve illegal actions.