4 ms·
> "The Claimant [Spiegel] argues that the Defendant’s [Eyeo's] business model endangers the Claimant’s existence," reads the judgment, which isn't final because
by koolba 10y ago
> "The Claimant [Spiegel] argues that the Defendant’s [Eyeo's] business model endangers the Claimant’s existence," reads the judgment, which isn't final because it can be appealed by Spiegel. Because users aren't willing to pay for editorial content on the Web, "it is not economically viable for the Claimant to switch to this business model."
In what situation is endangering someone's business model grounds for their illegality?
Also, I don't get why larger content owners don't self publish ads. It should be possible to have an appliance running on your own infrastructure or cloud, that gets proxyed requests from your own domain. That way the content it indistinguishable from the ads (or at least heavily merged). If you randomize the placement of ads / content images and both come from opaque URLs (ex: http://media.example.com/:some-uuid http://media.example.com/:some-uuid) it wouldn't be apparent which should be filtered.
Biggest issue I can see if that ad buyers would have to either trust the publisher or control the appliance. I suppose the latter is possible as if the request is proxyed you can align the counts of "Requests I've sent you" with "Requests you've received" to reconcile if there's cheating / lying happening.
This may be too complicated for a smallish site, but well within the technical capabilities of Der Spiegel, NY Times, etc.
- guntars 10y agoBecause it's not just about showing an ad to the user, it's about knowing which user to show that ad. The ad networks track you across the web, and in the process learn a great deal about your demographics and interests. This alows businesses to target you specifically. Spiegel can only track you across their own site which might not tell them much about you, making the ad space less valuable.
- koolba 10y ago> Because it's not just about showing an ad to the user, it's about knowing which user to show that ad. The ad networks track you across the web, and in the process learn a great deal about your demographics and interests. This alows businesses to target you specifically. Sure if you're not clearing your cookies, obfuscating your User-Agent, and rotating your IP address regularly. All that aside, the real baseline of comparison isn't a targeted ad, it's no ad at all. It doesn't matter how great third party ad networks are at tracking users, if their third partyness is what makes them so easy to block then they're going to go the way of the dodo. > Spiegel can only track you across their own site which might not tell them much about you, making the ad space less valuable. I wonder if that even matters. They already know what article I'm reading and possibly what other articles I've read (again assuming logged in / tracking cookies / etc). Should be more than enough. I really think a steady income stream of un-blockable but not quite as effective ads will win out over blockable but hyper targeted ones.
- TeMPOraL 10y ago> I wonder if that even matters. They already know what article I'm reading and possibly what other articles I've read (again assuming logged in / tracking cookies / etc). Should be more than enough. I agree. I mean, I would react much better to ads relevant to what I'm actually reading just now than to random crap based on stuff pulled out of some tracking database.
- irq-1 10y agoYou're right that it doesn't work for the advertisers. It also isn't a technical solution: ad-blockers don't just use domain names; they also block based on tags, ids and css classes, etc..
- intrasight 10y agoI'm sure the advertisers would rather have ads than not, so I'd argue just the opposite - this will work for advertisers. And you can't block arbitrary <p> and <img> tags, so I don't understand your second assertion.
- TeMPOraL 10y agoAd blocking in general is an AI-complete problem, but there are still lots of low-hanging fruits left to be picked - like blocking tags based on class names ("ad", "sponsor.*"), on content (basically spam filtering), etc., all backed by crowdsourced list of per-domain entries.
- majewsky 10y ago> blocking tags based on class names I would be surprised if uBO didn't do that. (Can't check right now because I use uMatrix; I care more about the tracking part than about the ad part.)
- intrasight 10y agoBy "arbitrary" I really mean "tags without classes". A plain <p> or <img> tag. But what I left unsaid is that I don't really mind such ads - they are like ads in a magazine. I use UBlock Origin to disable tracking not to disable ads.
- criddell 10y ago> This alows businesses to target you specifically. You know, I keep hearing this as justification for tracking. It's been going on for about 20 years now and I still see terrible ads. I don't think the ads I see online are any better targeted than ones I see in People magazine at my doctor's office.
- summarite 10y agoThis would probably result in less relevant ads and more power of ad providers, ie less neutral reporting. Think about how some companies or political parties 'punish' critical media by pulling ads and advertising with the competition instead. Imagine you little newspaper have a harsh truth to deliver about Nestle or P&G - and you know if you do so a large chunk of your revenue will disappear. On the other hand this would allow the site to be more picky and set prices - which could really increase overhead. I love the idea they use eg for wetransfer (a very minimalist file sharing site) - one huge, beautiful ad in the background . I'm sure those aren't cheap, but they are certainly visible and usually quite interesting. Everyone wins. But i doubt that would scale and seems unlikely to suit content-filled sites like news sites
- stevesearer 10y agoI self-publish and self-host my site's advertising and in my case the ads are actually more relevant. The different is that I target the ads based on the content rather than the reader. In that way I try to treat it like a magazine where the content itself narrows down who will be reading without the need for dertermining them using adtech solutions. No idea if the model would work on a general news site, but it does work and isn't terrible. I can't speak to your point about a brand pressuring me for being critical though since that isn't really the nature of my site and publication.
- rebuilder 10y agoAs far as advertiser pressure goes, is it much different now? Advertisers can, and do, pull their ads from specific sites at will. For example, letting a big company know their ads are appearing on a Russian-sponsored site full of chaosmongering "news" will usually get them to pull their ads right quick.
- jasonthevillain 10y ago> Also, I don't get why larger content owners don't self publish ads. You're right, it's about auditing. They need to be able to prove the quantity of ads were delivered as sold, which needs to go through a third party. Actually, Vox, NBC, Quartz, The Outline, and NYT have started building their own ad creatives in-house, which is an excellent idea because it gives the publisher control of what code runs. This gets straight to what the readers want solved. They're all technically capable of building their own delivery system too, but I'm not sure what the benefits would be?
- pbhjpbhj 10y agoCould they use sampling [+ trust!] instead for audit purposes? Or indeed serve from their domain with a third-party server that sits inside (eg by VPN) the orgs network.
- jasonthevillain 10y agoThat's interesting. You could serve the delivery system from the same domain with a proxy or CNAME or something, but that would actually be less secure. Moving them onto the main domain means they'd have access to its cookies/localStorage/cross-origin policy. Today, they don't. There may be a middle ground--using a separate, unique domain (i.e., nytimes-adserver.com). I'm trying to think through the side effects, what breaks, etc.
- clarkmoody 10y ago> In what situation is endangering someone's business model grounds for their illegality? Only in the case of government protectionism. And even then, there is no legitimate situation where it should be illegal to challenge an existing business model. I'm glad the Internet exploded before the government cronies really understood it.