Hacker News Reader

I can imagine websites refusing to link and instead displays the URL so the user can enter it themselves.

A lot of new companies will pop up offering their LINKTAX compliant packages that offer automatic downgrading of off-site links to user-selectable link fields, some even with custom functionality to automatically place the link in the clipboard.

Tech savvy users will use adblockers to disable this auto-delinking script.

All uploads will go through the few behemoths of media that can implement the upload filtering, like YouTube.

User uploaded profile images will be substituted for generated images by the website.

Uploaded text content such as comments will be disabled for all but the websites that live off them. Get used to seeing "Your comment is not currently visible, but awaiting moderation".

What a miserable, sad web...

It's not the link that's being taxed, but the short excerpt that's being displayed. If a website writes its own summary instead of taking a snippet directly from the article, there shouldn't be any fee.

Those auto-summary bots on reddit are suddenly a hot commodity.

Isn't that what the meta description is for?

Recent advances in DL for NLP will easily be able to help with it. Just rephrasing, nothing fancy. Though, I am not sure whether it will be considered non excerpt or whether EU will introduce new laws to work this around.

>Unless they're a top 2 media conglomerate of an European country, and would like for all their weaker competitors to be killed off.

You'll be accused to be a conspiracy theorist. Is that legal in the EU still?

You mean is it legal to "think this is a good idea"?

Yeah, it's absurd.

> Unless they're a top 2 media conglomerate of an European country, and would like for all their weaker competitors to be killed off.

Maybe that's the idea, but I think it's shooting yourself in the foot, even then.

It's even worse. Everyone who you link to has the right to claim damages(in case they are press). So you will have to be careful who you link to since a single link could already fall into the new legislation depending on the member state.

Does anyone know how this "press" is defined in the proposal?

More and more legal entities (governments, companies,...) seem to be moving to actively blurring lines so as to induce fines and damages. Suppose instead the proposal mandated any content hoster (press or not) to include a <linktax> tag so that hosting software can be configured to suppress the inclusion of links to pages that contain the tag.

Without such an explicit tag, the situation will end up similar to ad blockers, with random distributed lists of sources that would probably qualify as "press"

In the Commission proposal: "‘press publication’ means a fixation of a collection of literary works of a journalistic nature, which may also comprise other works or subject-matter and constitutes an individual item within a periodical or regularly-updated publication under a single title, such as a newspaper or a general or special interest magazine, having the purpose of providing information related to news or other topics and published in any media under the initiative, editorial responsibility and control of a service provider."

Thank you very much for looking it up!

The 'definition' is exactly as vague as I feared it to be:

Consider a blog, one could easily argue that it is a (samizdat) press publication, the blog has a name or title, is regularly updated, provides information to news or other topics and is published as a medium (everything on the web is published in the medium of HTML, PDF, ...) under the initiative, editorial responsibility and control of the "service provider" of whomever maintains the blog.

The same can be argued for any website, forum, government webpage (so we can't even safely link to government pages without exposing oneself to the risk of a fine!)

If the goal was not to expose normal daily citizen activity to the risk of fines, then they should simply require "press publications" to self-declare themselves to be such by means of a <linktax> tag, or perhaps even better, by having those providers who WISH to be freely linked to provide a <linkpermission> tag containing a cryptographic signature endorsing free circulation, such that upload filters can verify without users having to rely on decentralized lists of educated guesses of who would or would not sue for linking freely...

I don't quite get this, the problem is not with linking, but the link previews.

It seems pretty clear that the intent is for this to cover the current linking/snippeting practices. If it wasn't covered, there is no chance of this law providing publishers a new revenue stream.

And what are those current practices? On Google News it appears to literally be just the title and maybe a tiny thumbnail image. On search it's those plus less than a sentence of text. It's pretty hard to see the intent here as not covering even just a link that uses the original title.

Links that can't even use the original title would be totally nonviable, they might as well not exist. And again, this proposal would disallow even an opt-in arrangement.

Maybe the problem is that most of the articles nowadays could be shortened to just a few sentences as the rest of the body is just a filler and people just don't want to waste time going to the linked sites to ingest it?

What about HN, which is a link + title?

This would absolutely apply to sites like HN or Reddit. Of course sites outsite EU jurisdiction aren't required to follow EU law, but may eventually find themselves ordered blocked by European ISPs if they don't.

It's absolutely insane that freebooting [1] on Facebook was allowed to persist for so long. It's easy to see the right to monetary remuneration as a method of addressing a massive loophole exploited by Facebook and certain users for years.

[1] https://www.youtube.com/watch?v=L6A1Lt0kvMA

There are a few less obvious reasons why you wouldn't want to allow people to link to your content:

- Privacy: You want to be in control of how the information on your site is procured. This is common on government sites that generally allow public access, but require all sorts of forms and certifications to gain that access.

- Competition: If you are competing with Google or another to create the user's initial landing page, you would reasonably want to prevent them to access your own landing page content.

- Trust/Reputation: Google tries to link to as many pages as it can, but it can't link to them all. If your goal is to provide a trusted comprehensive service, you may want to deter "easy" but "non-comprehensive" access to your content. For example, it's conceivable that a medical journal would prevent access because Google only indexes 40% of it's content, and end users may develop a false (and potentially deadly) understanding that Google is comprehensive.

Absolutely, but not getting linked to has always been an option.

Publishers want Google to link to their content. At the same time they want Google to pay for the honor and prevent competitors from letting Google link to their content for free.

The only reason this ridiculously brazen set of demands doesn't get laughed out of the room is that these publishers have tremendous political clout.

The big issue is that Google over time has moved to scrape the web.

One big thing Google has done recently is they extract what you search for from a website, and show it inline.

E.g. when I search for a recipe, I get the result in the Google search, and don’t need to visit their website. https://i.k8r.eu/aILGEg

Google extracts reviews from Yelp!, and I can read them without visiting Yelp!.

They’re doing this for everything, and AMP furthers this.

Google is trying to keep the user as long as possible on Google, and show snippets instead of letting users leave.

As result many of these sites lose out on ad impressions, and money.

Google has no way to disable this feature short of entire unlisting your site.

I can see the problem with that. But as I understand it the proposed "link tax" is not limited to this sort of thing.

Yeah, it’s why I hate the law as it is, but I can obviously also understand why Google seems like a threat to the actors behind this.

> The big issue is that Google over time has moved to scrape the web.

This is true, and a lot of Google's practices are monopolistic/hegemonic.

However banning/taxing links is not the answer.

But what are you gonna do about it then? Ban snippets? Tax snippets? Wait for Google to be the web? Specially regulate Google?

There’s no good solutions.

> But what are you gonna do about it then? Ban snippets? Tax snippets? Wait for Google to be the web? Specially regulate Google?

I think I would ban search engines from doing other things, ban or regulate targetted advertising, require social networks to be interoperable, and ban social networks from owning other social networks (so Facebook would have to sell off Instagram).

That’s an idea I can agree with, and that Vestager visibly sympathizes with, but that’d mean breaking apart Google, and as result would anger a lot of people.

Google's stated goal is to provide people access to information. But in reality, they provide access to what's easily accessible to them. Any robots.txt tool is far too blunt of an instrument. It does not respect other highly valued goals of privacy, competitive fairness, and comprehensiveness, among others.

To put it more starkly, to assist with privacy, there should be a robots.txt tag/line that allows scraping, but prohibits indexing anything that is tagged as "personally identifiable information". It should be easy to Google what the newest bankruptcy law is, but should be a bit harder to figure out if a specific person filed for bankruptcy.

Similarly, with comprehensiveness, robots.txt should be able to say "don't index these pages unless you also index those pages". For example, a medical site should be able to say, you can index these several name-brand drugs but only if you also index their generic equivalents.

And as far as fairness, a better robots.txt should be able to tell Google to index the portions of its content to attract clicks, but prevent them from scraping their most valuable content. For example, Yelp may allow them to index a restaraunt's overall rating, but not summarize their reviews.

You can just have a `robots.txt` if you don't want Google to link to your content.

The issue is that sets a black and white option of not letting people find you through Google. But the alternative is letting Google scrape your content and give it to people directly from their site instead.

> But the alternative is letting Google scrape your content and give it to people directly from their site instead.

No, that is not the alternative. The alternative is to allow citations as they are allowed everywhere else, i.e. small fractions of the content with attribution.

For example, google currently may list the headline of an article and link to it. The proposal, which is based on a failed German law, is that google should pay for the privilege of sending viewers to newspapers this way.

Last time they tried google simply said "no, thanks. Then we'll not list you". Newspapers immediately caved and said they will give free access. End result: monopolies get free access and smaller potential competitors have to pay. The law completely and predictably backfired.

The, misguided IMO, hope now is that if it is an EU wide law, google will have to bow down and pay for the privilege of providing a google news like service.

I believe the newer proposal addresses that flaw. But to be honest, if Google can't build a service without theft, it should simply not operate. New solutions will be able to fill the gap.