Hacker News new | comments | show | ask | jobs | submit login
Problems coming for online publishing (harvard.edu)
168 points by jedwhite 32 days ago | hide | past | web | favorite | 64 comments



This is a rather asinine line of argumentation.

(a) The New York Time's newsroom and editorial staff are rather well-separated from their business department which makes decision about ad tech on their website. If you don't believe me, see this paragraph from a recent article at The Atlantic, which operates by the same principle:

As you read this article on The Atlantic, roughly three dozen ad trackers are watching you, adding your interest in this story to profiles they maintain on your online behavior. (If you want to know more about who’s watching you, download Ghostery, a browser extension that tracks and can block these “third-party” trackers.)[0]

That's a reporter arguing against the financial interests of his employer. And they are not going to fire him.

(b) There's one big difference between Facebook and generic advertising technology: the New York Times doesn't require you to sign up. Every time you change devices or delete cookies, the ad targeting starts from a blank slate. And while some websites try to get you to sign up, it's trivially easy to switch to a new account every once in a while.

(c) Ad tracking knows which websites I visit. Facebook knows which nude parties I attended in 2005

(d) The New York Times has actually turned to subscription revenue as its main source of income. While the split used to be 50/50, it is not much closer to 75/25 (subscriptions/ads). The "if you're not paying, you're the product..." cliché is getting tired anyway. But its premise simply doesn't apply to the New York Times.

[0]: https://www.theatlantic.com/technology/archive/2018/03/data-...


> Every time you change devices or delete cookies, the ad targeting starts from a blank slate.

This hasn’t been true for quite some time. These banks on you are really elaborate. They know where you live, they know where you work, they know where you drink your coffee and they know your online behavior.

Those factors mean that you’re never truely anonymous anymore. When you buy a new device it gets tiered to your access points, then your online behavior gets profiled and matched to people who frequent your access points.

Last summer I was on a business trip to another region of my country. Just before I went I had to turn in my work pc for repairs, which meant I had to use the borrow device from our IT department. Despite never having used this device before, and never having logged in from that particular hotel I still ended up getting adds for a very specific online course that I had taken earlier that month as well as a pair of shoes I had just bought.

I guess the irony in that is that even though they we’re able to pinpoint who I was, through what seem like impossible means, they still didn’t know I had actually already bought those products they were advertising.


> (b) There's one big difference between Facebook and generic advertising technology: the New York Times doesn't require you to sign up.

> (c) Ad tracking knows which websites I visit. Facebook knows which nude parties I attended in 2005

Facebook tracks you off the site and has been known to build shadow profiles of non-users.


It's plausible that a world of ad brokers (with many smaller ad buys) is better than a world of few but very large ad engagements, vis a vis the Chinese firewall. In that world, a single advertiser would have enough (purchasing) power over the newspaper, to the point where they could veto an article critical of that advertiser.


Having worked at nyt I worked on nyt_cookie.so an Apache module for parsing and forwarding some of the early tracking systems . So back on the topic over hacks . The source for it was in a cvs repo hosted on this guys second office laptop , running FreeBSD 3.2-STABLE . So one it was 2009 and 3.2 was incredibly old ; as was the laptop . One day the laptops disk dies and no one had backups of the repo . Nyt_cookie.so was kept on the prod web servers for another 4 or 5 years before it was abandoned. We used to binpatch the binary to fix it up .


Might I suggest reading the article and not just the title?

>By the way, I want to make clear that Zeynep, Brian and Natasha are all innocents here, thanks to the “Chinese wall” between the editorial and publishing functions of the Times. In fact same irony applies to countless other correct and important reporting on the Facebook/Cambridge Analytica mess by other writers and pubs.


The mentions your first point, fairly early on.


Yes, you are right (and I was wrong). I admit I stopped reading when they suggested the media was being "ironic" (i. e. hypocritical) on the matter.


If it makes you feel better, I had stopped reading your comment when I found that point, and posted mine. My original version was that "the article addressed your points".

I subsequently reread your comment and realized you made several other points. Oops. I then edited mine.


Arguing against one's interests should not necessarily be taken at face value. It's a viable persuasion strategy if you are in for a long con.


Ah yes, the old "maintain a good reputation by behaving honestly" scheme.


This is the whole basis of a confidence trick. It's harder to believe someone's duping you after they earn your trust. https://en.wikipedia.org/wiki/Confidence_trick


Regarding (b): That's odd, maybe this differs depending on where the web surfer comes from? The New York Times definitely does require me to sign up.


It's pretty annoying when you pay (a lot) for local newspapers and still have to see advertisement.


> While the split used to be 50/50, it is not much closer to 75/25 (subscriptions/ads).

Thank Trump for that. This didn't happen naturally for them and was only made possible by the election of Trump. Its unlikely that this is going to sustainable for them beyond trump presidency.


I've updated that piece several times to clarify things, so you might check it again.

What it's a call for is hacking our way to publishing norms that don't involve tracking. And that's what we're starting to do at Linux Journal, where I'm editor-in-chief.

For the first time anywhere, we'll be agreeing to readers' terms, rather than the reverse. In legal terms, the reader will be the first party and Linux Journal will be the second party. I explain how we're approaching this in "Help Us Cure Online Publishing of Its Addiction to Personal Data": http://www.linuxjournal.com/content/help-us-cure-online-publ...

The barn we’re raising is for all of publishing, but we’re short-handed. So, if you're interested in helping, let me know (doc at linuxjournal dot com). Or show up on 2 April at the Computer History Museum: https://www.eventbrite.com/e/vrm-day-2018a-tickets-438706894... . Thanks!


Happens that I just posted an essay on that topic: http://yuhongbao.blogspot.ca/2018/03/google-doubleclick-mozi...


If you're going to mention this post in multiple threads, please do a regular submission instead.


Yea, I don't normally do this. I did it because docsearls was there.


Compare ad tech that news websites use to what Cambridge Analytica did.

1. Ad tech:

As I browse, my interests are accumulated into basic interest groups, demographic data, and maybe recently visited websites that use tracking.

The result: when I visit a news website, I might see an ad for tennis racquets (because of interest), or a local bank (because of location), or DigitalOcean hosting, because I recently visited that site.

2. Facebook's platform, used by Cambridge Analytica:

I install a Facebook psychological quiz app (supposedly from the prestigious Cambridge university), and give it permission to access my data.

Through that, Republican campaigns also access my grandmother's account, with all of her photos, likes, and comments. They use that data to learn her deepest fears.

They hire designers and writers to create fake news websites, and get their propaganda stories to appear in her Facebook news feed, where they appear to be legitimate right next to the news about the local sports team.

The result: when I visit for Christmas, I have to try to calm her down, that Hillary Clinton isn't actually running a late term abortion clinic out of a pizza parlor. And she won't believe me.


You are normalizing data collection by third party companies. The article is about selling our data to third party by both Facebook and other websites.

Truth is we don't know what any of those companies are doing besides showing ads. Maybe today they are not doing it, but tomorrow they can use it for other purposes. They know what people are browsing and what their interests are for a location.

Also they can pinpoint users by ip, canvas fingerprinting, fonts installed etc.. https://firstpartysimulator.net/tracker


Can we please keep the reddit style hysterics on reddit? I pretty much left reddit to escape the fake political nonsense. And with all due respect, you are doing exactly what you claim cambridge analytica did. You are pushing extreme fear and agenda. But for what end?

> As I browse, my interests are accumulated into basic interest groups, demographic data, and maybe recently visited websites that use tracking.

You are forgetting installing cookies and fingerprinting ( amongst others ) to track across platforms. You are forgetting using that data to target customers with tailored fearmongering news articles to better clickbait.

> Through that, Republican campaigns also access my grandmother's account, with all of her photos, likes, and comments. They use that data to learn her deepest fears.

I hate to break this to you but everyone does this. Obama did this. The democrats did. Hillary's campaign did this.

I love how you excuse news websites when they cause hysteria and panic over nonsense like campus rape crisis and pay gap in male and female sports. Not to mention the hysteria over cambridge analytica. Seems like the news has done far more harm to you than facebook did to your grandmother, assuming your story is legit and not made up.

I recommend you go watch the 13th on netflix to see how news and media can be dangerous. The problem now is that news has access to tons of data about us ( just like facebook ). The biggest difference is that facebook was pretty much "neutral" whereas the news/media are highly biased. Facebook allowed both trump and hillary political ads.


You are right, both sides are to blame, but this don’t make any of the them better.

In Brazil we have the same mentality of “sides” and I believe that this is destroying the good sense of some people.

It’s not because everyone does it that it should be the norm, or that is not evil.

Our true challenge is how we can stop this type of propaganda without hurting any freedom?

In Brazil All the big publishers are in someway connected to the people in power, so our non-fake-news were never really impartial or independent. Some even say that the media should be controlled by the government (which is to me a 1984 scenario, but not so much different from what it’s today).

I believe that the only way to end this fake-news madness will be a truly anonymous p2p communication, but this would generate another kind of problem.

(Sorry for my english, I’m still learning how to write properly and an argument is kind hard to do, I may sound like provocative but that is because of my lack of vocabulary not my intentions)


> You are right, both sides are to blame, but this don’t make any of the them better.

(Both sides refer to the broadcast/online media & commercial/political ad campaigns)

Blame can only be placed on either party if there exists the prior assumption that people should believe all things which are presented to them as fact. If this assumption does not exist and without facilities responsible for fact-checking then it's only the individual that is responsible for dividing fact from fiction.


> You are right, both sides are to blame, but this don’t make any of the them better.

He wasn't saying this, he was calling out someone who, in a very dishonest manner, attributed a commonly perceived as bad trait to only one third party of his choosing. This, of course, while ignoring the fact that those he was trying to defend did the same and so rendered his argument useless and turned his deliberately included emotional cues into something short of deception.


I am rather skeptical of the effectiveness of online advertising, despite all the tracking and profiling. I see companies like Google and Facebook post bumper profits, yet I see users becoming smarter at avoiding ads. Something doesn't quite add up, and because the validity of clicks is shrouded in mystery, it's not difficult for the platform owners to play dirty. I have had very little success with online ads for my app (https://usebx.com/app), yet what worked really well for acquiring users was releasing a JS library for free (https://osrec.github.io/currencyFormatter.js/)


What doesn't add up is that you and your circle are not typical users. Try talking to advertisers doing direct response on Facebook — they keep a very close eye on ROI. Brand advertising is less trackable, but advertising does and has always worked. If it didn't, businesses that didn't spend on advertising would outcompete those that do.


Then why are companies like P&G cutting digital ad spend significantly? And why is this actually having a positive impact on their bottom line?


Because advertising for big companies is all about brand awareness and has little to do with sales per se. You can't follow a P&G add and end-up in a page where you can buy the damn product, 9 out of 10 times you'll end-up in an info page.

It's the little guys that get the most out of digital advertising, especially e-commerce sites where they can link ads to products.


I would have agreed with you 5 years ago, but currently, I don't see the same in my business, nor in the businesses of my close friends or relatives. It could of course just be that we're in a niche where ads don't work too well!


I think that the pie of attention is still increasing which offsets the crisis, but the percentage of people running adblockers is rapidly approaching “typical user” in many countries. See the most recent PageFair report, percentage of users running adblockers increased 30% YoY:

https://pagefair.com/blog/2017/adblockreport/


you don't see users getting smarter about ads. trust me.

you see your close circle of acquaintances getting smarter about ads.

the general population is going downhill, by all metrics we have access to. In one case, people were trying to access a (fake) bank site, which had ads. it was an ad scammer that bought mistyped domains and cloned the homepages, in an automated way. and very badly. yet it had millions of impressions, most with several PVs from same user showing that they did click around.


Almost everyone tech savvy that I know runs an ad blocker and I've installed one for all my family members who aren't. At family holiday gatherings they've even started asking me to "do that thing you do" to clean up their browsers. Regular people are getting very very tired of so many obnoxious ads, and even though they don't think about it much, no one is like "hey, I have WAY to much privacy! I need to give away my data!!"


You're still the exception. I don't use an ad blocker, of my two immediate IT staff coworkers, only one does use one.

However, I do do things that mean I see fewer ads, especially obnoxious ones. I spend a lot of my time in RSS feeds, not often venturing to the actual sites. When I do go to those sites, if ads are burdensome, I use Safari's Reader View and I have enabled the "always use" option for a few sites (e.g. The Verge). I'm also on sites like this and Reddit, which is quite ad-lite. I subscribe to one political news site which gives me A) full-text in RSS and B) reduced ads on the site (I occasionally go to the site because the RSS doesn't have images or embedded content).


> You're still the exception

Depends in the measure, I guess. Recent versions of Chrome have started blocking certain obtrusive ads by default. That's getting pretty mainstream.


Google is the source of most of those ads. think about it.


> Almost everyone tech savvy that I know runs an ad blocker and I've installed one for all my family members who aren't.

How do you provide support to them for websites that malfunction due to the adblockers? IME, they often don't recognize that the problem is the adblocker much less have the patience to call me for help - and if they did, I don't always have time to help them when they need it.


I install Firefox + uBlock Origin and tell them that if they have specific issues they can hop back to IE.

That being said, I'm trying to think, and I honestly don't know that I've ever seen any sites break due to ad blockers (it may be that I'm guilty of not recognizing the source of malfunctions myself). Regardless, the web is a big place and life is short. If a site can't get it right I just move on. Mental bandwidth is a limited resource and (IME) ad-blocked pages consume less overall, despite possible issues.

Oh, I have seen the ones that throw up anti-adblock walls. I hate those.


> Oh, I have seen the ones that throw up anti-adblock walls. I hate those.

The anti-adblock list in uBlock can help with that. Read the README.md though because it requires a user script too.


I know someone who has a company with over a billion unique visitors a year and they only bring in tens of millions a year in revenue.

I have a site with over a million uniques a year and we did almost thousands a year in revenue.

We are far from experts on selling ads but still it’s a tough way to monetize something small.


Your examples aren't contradictory. A site with a bn uniques can have let's say 10bn impressions. With a CTR of 1%, assuming by the the high number of users that it's news related and so doesn't get that many clicks, and a CPC of $0,15 it could bring a low 8 figure income.

It all depends on the kind of content. Target women and you get higher CTRs. As for the guy with the billion uniques, well he should start contacting advertisers himself. With that kind of traffic it's a crime not to.


To see what's most effective it might be better to watch how ad companies are selling their own services than to follow their advice. It seems like they're getting more through sales staff and content marketing and that online ads are only a small part of what gets them sales.


> yet what worked really well for acquiring users was releasing a JS library for free

I notice this with a lot of new OSS. E.x. Etcher is basically an ad for Resin (tired of flashing SD cards? Use resin and do it with a git push.)

This is the best form of advertising because I get a free useful tool out of it.


Yes absolutely. I think if you want people to give you their time and trust, you have to give them something of value in return. We're working on releasing more OSS as it really does help to bolster our business. In fact, we are even thinking of making Bx itself free, and simply taking a small cut when users receive payment for invoices raised on the platform.


If the backlash lasts long enough to really change the landscape of non explicit permission data gathering then an opportunity for hacker news one million jobs may arise.

How much is my data worth? I’ll sign a buncb of revenue sharing agreements with YouTube, Google, Mozilla, Facebook etc if the price is right. I know that HN has a lot of really heavy pro privacy advocates but I hope you can recognize a lot of people don’t share those concerns and would happyily welcome a little passive income stream in their lives.


> people don’t share those concerns and would happyily welcome a little passive income stream in their lives.

This is what I gathered from moviepass privacy thread on HN. Most people didn't care about sharing location if it gave them cheap movie tickets.


> How much is my data worth? I’ll sign a buncb of revenue sharing agreements with YouTube, Google, Mozilla, Facebook etc if the price is right.

That's not how it would work.

In that scenario, you'll have a choice of paying $N per month for the higher privacy version of Facebook, or you can use the advertising supported low-privacy option.

They are not going to pay you. Your other choice will be to not use their platform and they will be fine with that because everyone else will continue to choose ad supported because they do not want to pay $50 or $100 per year for Facebook when they can get it for what they perceive to be as free.


Great point. Makes you think there may be room for another ad middle man that offers more to consumers because pretty much all adtech right now is built on top of Google, FB, or Apple ad platforms.


Really does seem like an interesting use of smart contracts?

And it's not just social data -- there's probably real personal and societal benefit to collaboration around medical data too.


Ha, redmorph's website doesn't even load without enabling third-party scripts (it loads http://ip-api.io, I guess to geolocate visitors)


I think adtech companies are at greater risk than publishers. Publishers will move to a subscription model and most of them will be okay. A whole lot of adtech companies’ entire business model is sleazy at best and outright criminal at worst.


Yeah but it's also crazy how many pubs (the word is nice ;)) already had to close or at least merge which others. Google News made it quite obvious how redundant a lot of content is. Often articles are just exact copies of Reuters, AP, ... texts.

If subscription is the only solution, I expect more pubs to disappear. Being from Germany, if I had to subscribe to nytimes.com to read articles from there I'd stop reading them probably. Just pretending I would subscribe, I would also have to subscribe to 5 more US Newspapers, 2-3 UK ones and 10-15 German ones. NYT Basic is 1.75 € a week, Washington post 10 $ a month. That would be more than 140 € a month. No way :-) (Oh, I forgot all the commercial tech pubs I read)

I hope the publishers come up with a better model.


> Google News made it quite obvious how redundant a lot of content is. Often articles are just exact copies of Reuters, AP, ... texts.

Reuters and AP are newswire services which is meant to be used by multiple pubs because local pubs don't have reporters all around the world. The local papers' reporters are busy telling you what your local schoolboard is up to and whatnot.

But I agree more pubs will likely disappear or be merged into larger media conglomerates, and that leads into issues like Sinclair forcing all their pubs to push certain agendas.


I would like to mention that I find it extremely irritating that despite paying for a subscription, the NYT still collects data on me and serves me ads. That's the worst of both worlds.


Yep. I'm not anti-advertising, but I am anti-tracking. Serve me ads, but please don't personalize them.


No, change isn't coming. Sorry.

Trump's victory signaled a social media gold-rush for political campaigns. 2018 is a major election year in American politics. You can bet that campaigns are evaluating and signing onto social media related efforts. To meet this demand, as Cambridge Analytica buckles, others will take it place. The people who benefit by these services are those who make the laws. While legislators benefit by social media targeted services, there won't be a political will sufficient to pass a law that will slay the goose laying golden eggs.


Interestingly, the Wall Street Journal had an Editorial piece yesterday that Facebook should pay for news and media content to increase its trust. There is a lot of self-interest in much of the the commercial-scale hating on Facebook. This is the article although it may be paywalled (WSJ uses variable restriction based on user prediction ironically enough): https://www.wsj.com/articles/facebooks-public-reckoning-1521...


Media is definitely taking the chance to try and kick Facebook to death, and often for selfish reasons. That shouldn’t distract us from the very real fact that Facebook is a terrible value proposition for users of their service, untrustworthy, and even dangerous. Facebook deserves to be kicked to death, even if the media wants it for their own reasons.


The original title was "Facebook’s Cambridge Analytica problems are nothing compared to what’s coming for all of online publishing" for those wondering what more specifically this post is about. I shortened it when I posted it to get it under the 80 char limit, but the further mod edits probably lost some information of value :)


Craigslist replaced paper's classified ads. Pretty much every website is using online ads. I gotta wonder who is doing the coupons that a fair number of people use the paper to get?


Coupons are a fascinating business, with a lot of middlemen taking cuts. From the agency that creates the advertising campaign (and possibly books the media) to the media platforms that display the coupons (whether inserts in physical papers, direct mail flyers, digital coupons via coupons.com/Quotient and retailer sites and emails, etc). Plus the clearinghouses that process the physical coupons for retailers, and the processors that handle the financial aspect of cutting a check from the manufacturer to the clearinghouse and onto the retailer for the redeemed coupons.

There are a few big names in each of those sections; but "doing the coupons" is an entire process that sets off a chain of necessary suppliers, rather than just a single name like Google.


The title should end with "publishers", not "pubs".


There is an 80 character title field limit on the HN input field so the title had to be shortened.


The title should be reworded, I had to read the article's title to understand it.

Edit: and as I say this it was just modified.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: