Other Big CJEU Case Says Google Must Put Certain Links At The Top Of Search Results

from the must-carry dept

While most of the attention today was focused on the CJEU "right to be forgotten" ruling concerning global censorship, the court actually released another ruling concerning the right to be forgotten, again around disagreement between French regulators and Google. And, as intermediary liability expert Daphne Keller notes, this ruling may turn out to be more interesting in the long run.

This case involved how Google should deal with "sensitive data," when it's a part of a RTBF request. The court does decide that a "notice and takedown" regime makes sense for such sensitive content, which is better than the possible alternative advanced by some: that the law requires Google to pro-actively stop the indexing of such sensitive information (or even to first get consent). The court points out that this wouldn't make any sense at all, given how search engines work:

In practice, it is scarcely conceivable — nor, moreover, does it appear from the documents before the Court — that the operator of a search engine will seek the express consent of data subjects before processing personal data concerning them for the purposes of his referencing activity.

But what really stands out is what appears to be a totally uncalled for random aside by the court towards the end of the ruling:

It must, however, be added that, even if the operator of a search engine were to find that that is not the case because the inclusion of the link in question is strictly necessary for reconciling the data subject’s rights to privacy and protection of personal data with the freedom of information of potentially interested internet users, the operator is in any event required, at the latest on the occasion of the request for de-referencing, to adjust the list of results in such a way that the overall picture it gives the internet user reflects the current legal position, which means in particular that links to web pages containing information on that point must appear in first place on the list.

This seems like it could be a very big deal. It's the court saying that Google is required to make sure that top search results "reflects the current legal position." In other words, if someone was exonerated after being accused of a crime, that must now be the top link. As Keller notes, this is going to have some strange consequences that probably won't be very good:

Having a court come in and tell a search engine what is "required" to be the top search result, tossed off without much detail or thought in a world where getting certain links to certain places on search engines is literally an entire industry, is going to have pretty significant consequences -- nearly all of them I can guarantee you the court did not even begin to think about.

Filed Under: cjeu, cnil, eu, links, must carry, right to be forgotten, rtbf, search engines, search results, seo
Companies: google


Reader Comments

Subscribe: RSS

View by: Time | Thread


  • icon
    Anonymous Anonymous Coward (profile), 24 Sep 2019 @ 11:32am

    Are the goose and the gander on the same page?

    Let us say that George Smythe has been acquitted of some dastardly deed and this then pops to the top of the search list. Let us also assume that my name is also George Smythe. That would mean that anyone searching for my name could assume that I had been, at the very least, accused of some dastardly deed. How well will that play out?

    I would like to know how those in favor of the RTBF feel about smearing others, intentionally or not?

    reply to this | link to this | view in chronology ]

    • identicon
      Anonymous Coward, 24 Sep 2019 @ 12:23pm

      Re: Are the goose and the gander on the same page?

      What I want to know is why do people assume that people name are unique when government assign them a number to distinguish the from other people with the same name?

      reply to this | link to this | view in chronology ]

      • icon
        Anonymous Anonymous Coward (profile), 24 Sep 2019 @ 12:32pm

        Re: Re: Are the goose and the gander on the same page?

        Those numbers don't tend to be public knowledge, but names are. Are you recommending the publication of personally identifying numbers? Do you have any idea that would do for identity thieves?

        reply to this | link to this | view in chronology ]

        • identicon
          Anonymous Coward, 24 Sep 2019 @ 12:49pm

          Re: Re: Re: Are the goose and the gander on the same page?

          Those numbers should not be published, just like people should not assume search results on a name refer to the same person, or the person they are researching, without further evidence.

          reply to this | link to this | view in chronology ]

          • icon
            That One Guy (profile), 24 Sep 2019 @ 12:58pm

            'Solve' one problem and create another

            While good advice that would seem to run counter to the idea behind rulings like this, where the assumption is that people will take whatever the first results are as true at face value, and as such it's important that the more recent stuff is at the top.

            For the people who aren't likely to do more digging before making making an evaluation of a person, and thereby need to see the newer stuff first, someone having the same name as a person accused of a crime is likely to be lumped right in with them, such that the court's 'solution' would seem to have replaced one problem with another.

            reply to this | link to this | view in chronology ]

          • identicon
            Anonymous Coward, 25 Sep 2019 @ 11:16am

            Re: Re: Re: Re: Are the goose and the gander on the same page?

            "Shouldn't" is entirely subjective. People shouldn't do a whole lot of things, but they do anyway.

            reply to this | link to this | view in chronology ]

            • icon
              Wendy Cockcroft (profile), 26 Sep 2019 @ 3:20am

              Re: Re: Re: Re: Re: Are the goose and the gander on the same pag

              That's true. I can't help thinking that these judges ought to do what some American ones do; educate themselves on the subject. They don't know how search engines work.

              In any case, someone has only to do a search result with "John Smythe court case" and that will bring it up, no matter what SEO, etc., has been done to bury it. So anyone wishing to check someone out has only to add the word "court" or "offence" or "crime" to someone's name to bring up the required search results. Even if Google removes the link from the search results, people posting and sharing the link will put it back on. This is a stupid ruling.

              If someone has done something stupid back in the day, they need to own it, move on, and show how they've become a wiser person. The ability to learn from mistakes instead of doubling down on them is a desirable trait, isn't it?

              reply to this | link to this | view in chronology ]

    • icon
      Peter (profile), 24 Sep 2019 @ 11:21pm

      George Smythe the plumber might very much prefer ...

      ... to see good rankings for his business at the top of the search results rather than reports of a court acquitting him from accusations of ripping off his customers.

      reply to this | link to this | view in chronology ]

      • identicon
        Federico, 25 Sep 2019 @ 4:54am

        Re: George Smythe the plumber might very much prefer ...

        No problem. The plumber website now only needs to have an "about" page which at the very end mentions in passing their acquittal from a certain accusation in year X, and then the page can be selected as the one and only legitimate first search result. For extra safety, in case someone looks for "plumber crime" and risks finding the older news, the page should be the first result for anyone searching "plumber" as well.

        reply to this | link to this | view in chronology ]

        • identicon
          Anonymous Coward, 25 Sep 2019 @ 9:08am

          Re: George Smythe the plumber might very much prefer ...

          So what happens if they're searching something else, like "plumber" AND "princess"?

          Does it just return the plumber link first, effectively as a disappointing reminder that our princess is in another castle?

          reply to this | link to this | view in chronology ]

          • identicon
            Anonymous Coward, 25 Sep 2019 @ 10:27am

            Re: Re: George Smythe the plumber might very much prefer ...

            Turns out Joe is not actually a plumber, why did they call him a plumber in their political rallies? idk ... must be more of that red neck bullshit.

            It is a search engine. It takes your search string and looks thru a huge mess in order to return something relevant to your input. If your search criteria is weird, then the results will be weird. Try it out and see.
            Granted, the results are skewed toward big business because they bribe the engine management who then directs the minion coderz to basturdize the algorithms.

            reply to this | link to this | view in chronology ]

  • This comment has been flagged by the community. Click here to show it
    identicon
    Undetectable Deepfaxes, 24 Sep 2019 @ 12:24pm

    So, gov't CAN tell GOOGLE what to publish and where.

    There goes one of your frequent assertions, kids.

    However, I don't see much programmatic trouble here. Soon as a hit in the exception list, it inserts the required link, and done.

    That exception list in first place is the major obstacle. Of course, as I pointed out last week, the Ruling Class simply isn't going to allow their information to be found and exposed, so exceptions are de facto intrinsic to "search engines".

    reply to this | link to this | view in chronology ]

    • identicon
      Anonymous Coward, 24 Sep 2019 @ 12:33pm

      Re: So, gov't CAN tell GOOGLE what to publish and where.

      “There goes one of your frequent assertions, kids.”

      Whatever helps keep the voices in your head to a reasonable volume bro.

      reply to this | link to this | view in chronology ]

      • icon
        That One Guy (profile), 24 Sep 2019 @ 12:51pm

        Look, it's easy, just like this

        If people are going to respond to Blue the least they can do is strip out/replace the subject line. While it's not as bad now as when they used to put entire sentences up there so every one of their marks was repeating their comments by responding it's still rather counter-productive to have the comment hidden so that people only have to deal with their garbage if they want to yet still have whatever inanity subject-line they went with repeated with every reply.

        reply to this | link to this | view in chronology ]

    • icon
      Stephen T. Stone (profile), 24 Sep 2019 @ 12:36pm

      So, gov't CAN tell GOOGLE what to publish and where.

      Outside of the United States, where the First Amendment doesn’t apply? Probably, yeah.

      Within the United States, where the First Amendment applies and the government cannot compel speech from people and (in general) from corporations? Yep, the government can’t do that.

      reply to this | link to this | view in chronology ]

    • icon
      Gary (profile), 24 Sep 2019 @ 12:37pm

      Re: So, Troll can't read

      There goes one of your frequent assertions

      Please cite the "frequent" assertion you mean?

      And by "Government" are you suggesting the US Government should toss out the first amendment and tell Google what to say? Do you hate free speech so much you'd let government toadies censor you?

      reply to this | link to this | view in chronology ]

      • icon
        Stephen T. Stone (profile), 24 Sep 2019 @ 12:45pm

        Do you hate free speech so much you'd let government toadies censor you?

        He would if they invoked copyright as the reason for that censorship. 😁

        reply to this | link to this | view in chronology ]

      • icon
        PaulT (profile), 25 Sep 2019 @ 12:22am

        Re: Re:

        He's saying that so long as one government is fulfilling his authoritarian fantasies, then they should be doing that in the US as well, constitution be damned.

        Unless the government are compelling action he disagrees with, in which case the constitution trumps every other idea on the planet and must be defended with blood.

        reply to this | link to this | view in chronology ]

        • icon
          Scary Devil Monastery (profile), 27 Sep 2019 @ 3:35am

          Re: Re: Re:

          "Unless the government are compelling action he disagrees with, in which case the constitution trumps every other idea on the planet and must be defended with blood."

          Except when the constitution doesn't agree with his views in which case he takes creative liberties in rewriting it and presenting it as prima facie.

          The best example of which would be when Baghdad Bob/Blue keeps trying to present article 8's "promotion of science and the arts" as a wholesale endorsement of copyright when in reality it's that one amendment in the US constitution the founding fathers found so controversial they made it an OPTION congress could yea or nay in a simple session.

          reply to this | link to this | view in chronology ]

    • icon
      Toom1275 (profile), 24 Sep 2019 @ 4:12pm

      Re: So, gov't CAN tell GOOGLE what to publish and where.

      Rational, logical thought is nice; I recommend you give it a try.

      At least once.

      reply to this | link to this | view in chronology ]

    • identicon
      Anonymous Coward, 24 Sep 2019 @ 5:58pm

      Re:

      There once was an out of the blue
      Who hated the process of due
      Each judge that he'd paid
      Was DMCAed
      And said, "Wow, what an ignorant motherfucker!"

      reply to this | link to this | view in chronology ]

    • identicon
      Anonymous Coward, 24 Sep 2019 @ 6:12pm

      Re:

      So, gov't CAN tell GOOGLE what to publish and where.

      No, they can't. While the court makes a weird side ruling in this case, the overall ruling is "Google can do whatever the hell they want to with their search results outside the EU". Full stop.

      There goes one of your frequent assertions, kids.

      And that would be??????

      However, I don't see much programmatic trouble here.

      Obviously. You're not a programmer, nor do you know how any of this works, as you've amply demonstrated.

      Soon as a hit in the exception list, it inserts the required link, and done.

      What exception list? What required link? Is it the same link for every person? Unlikely, in fact certainly not. So now you're saying Google has to build a database on each and every person and know which links containing their information are "most relevant" and which ones aren't. How does it know that? Does it go by date? That wouldn't work. Somebody could write a fake story with a more recent date and that definitely wouldn't be "most relevant".
      Congratulations, you've just presented an argument for Google collecting even MORE data on the world than they already do.

      That exception list in first place is the major obstacle.

      You mean impossible.

      Of course, as I pointed out last week, the Ruling Class

      What ruling class is this again?

      simply isn't going to allow their information to be found and exposed

      Oops. Guess somebody should have told them that since their information can still be found all over the internet. Somebody probably should have also told the court that too since they ruled in favor of Google.

      so exceptions are de facto intrinsic

      They are no such thing.

      "search engines"

      What's with the scare quotes? Are you saying search engines don't actually exist? What are you implying here? Be specific so we can continue to laugh at you.

      reply to this | link to this | view in chronology ]

      • identicon
        Anonymous Coward, 24 Sep 2019 @ 8:24pm

        Re: Re:

        So now you're saying Google has to build a database on each and every person and know which links containing their information are "most relevant" and which ones aren't

        Yeah... blue is seriously not great at thinking through his demands. If someone proposed a plan to him that involved Google having a database on everyone he'd shit enough bricks to help Trump build his wall.

        reply to this | link to this | view in chronology ]

  • identicon
    Paul B, 24 Sep 2019 @ 12:42pm

    Don't Tell Trump

    How can you tell what's true?

    Headline: Trump exonerated by Muller Investigation

    The President and Fox news says its true.

    reply to this | link to this | view in chronology ]

  • identicon
    Sok Puppette, 24 Sep 2019 @ 1:41pm

    It's the court saying that Google is required to make sure that top search results "reflects the current legal position." In other words, if someone was exonerated after being accused of a crime, that must now be the top link.

    Or, y'know, it could be the court not being absolutely perfect in drafting its decision.

    What you quoted could easily be interpreted to mean that if, say, somebody had been exonerated in a criminal case, that link would have to appear first among links mentioning the case. If the criminal case were minor 20-year-old bullshit, all of the links about it might still be on the tenth page, below more relevant material. Or nonexistant, for that matter.

    And even if you don't accept that interpretation, I bet they will correct it quickly if it ever becomes a problem.

    Now, whether it's possible for a computer to meet even that standard is another question. I sure don't know how I'd write code to figure out the current legal interpretation of anything. Maybe they'll accept "best efforts and quick correction". Which is still a can of worms. But why not stick to that real can of worms rather than jumping on things that are almost certainly unintentional and won't stick anyway?

    reply to this | link to this | view in chronology ]

    • identicon
      Anonymous Coward, 24 Sep 2019 @ 2:15pm

      Re:

      A search engine is being asked to make judgemental decisions, all I want is a search engine.

      reply to this | link to this | view in chronology ]

      • identicon
        Anonymous Coward, 24 Sep 2019 @ 3:11pm

        Re: Re:

        You don't want a search engine that presents the current and correct link before the obsolete and incorrect link?

        reply to this | link to this | view in chronology ]

        • icon
          Toom1275 (profile), 24 Sep 2019 @ 4:13pm

          Re: Re: Re:

          [Asserts facts not in evidence]

          reply to this | link to this | view in chronology ]

        • identicon
          Anonymous Coward, 24 Sep 2019 @ 5:37pm

          Re: Re: Re:

          There is nothing wrong with either link.

          A link is a symbolic representation of a location from which data can be extracted. A broken link is one that no longer has a valid source, or never had one. There is no such thing as a "wrong link".

          You are suggesting that the story, that has an update, is "wrong" if it does not include said update, or the updated story link should precede the original story link in a search engine results. This problem you have seems to be with the host of said article rather than the search engine that can be manipulated into providing you a link to said data. The search criteria could be changed such that you are presented with either story first ... your choice. I do not see how you have a problem with this.

          reply to this | link to this | view in chronology ]

        • identicon
          Michael, 25 Sep 2019 @ 4:59am

          Re: Re: Re:

          That's not what the ruling is actually saying. It is probably what they meant, but the ruling is actually saying that any search that results in anything relating to a legal case must put links for the latest ruling at the top.

          The result of this would be searching for "Chicken Soup Recipe" would have to show links to legal cases relating to chicken soup recipes before links to actual cooking sites.

          reply to this | link to this | view in chronology ]

          • identicon
            Anonymous Coward, 25 Sep 2019 @ 6:15am

            Re: Re: Re: Re:

            .. and this could result in the search engine providing a button to push when you are not interested in viewing all that legalese bullshit.

            reply to this | link to this | view in chronology ]

    • icon
      PaulT (profile), 25 Sep 2019 @ 12:25am

      Re:

      "But why not stick to that real can of worms rather than jumping on things that are almost certainly unintentional and won't stick anyway?"

      Because, intentional or not, that is the way the decision is written and presented to be interpreted. It's worth pointing out the problem up front rather than holding your breath and hoping they won't be interpreted that way.

      reply to this | link to this | view in chronology ]

  • identicon
    Ev Samuel, 24 Sep 2019 @ 2:52pm

    Is it wrong that I want to see this kind of chaos unfold here?

    I think something like this would be great. The SEO con artists messing around with things that are mandated by law, and nobody having any reasonable way of stopping them. Tying Google's hands, and making the concerns around this issue much much worse. God, imagine the new industries that would have to be created, simply to deal with something like this, and the fallout. It would be grand. Grand, I tell you.

    reply to this | link to this | view in chronology ]

  • identicon
    Bobvious, 24 Sep 2019 @ 3:30pm

    Perhaps folders of links rather than single entries.

    The court's (unwritten) assumption is that only a small number of results will be in the top links but a crapflood of system-gaming entries will generate multiple "first place" results. Will the court then be required to sort this list daily, or will search engine operators be required to make sure the list is no older than "one hour" with hefty fines for non-compliance?

    Keeping in mind that no system will be perfect, and that the list of people seeking RTBF is not currently massive, perhaps this option:

    Rather than having multiple single "first" links swamping the search results and growing down "exponentially", those could instead be placed in a single link that points to a folder of "firsts". At least the top 10 results will not then grow full of SEO abuse.

    Any attempts by SEO abusers will require those to be approved by the court first, even if that involves a continuous crapflood of approval notifications to the court. Otherwise it's like everyone trying to put up posters on the same piece of wall.

    reply to this | link to this | view in chronology ]

  • identicon
    Anonymous Coward, 25 Sep 2019 @ 6:14am

    And if you happen so search for John Smith would you then end up with the 20 000 top results of any John Smith being acquitted of crimes?

    reply to this | link to this | view in chronology ]

    • identicon
      Anonymous Coward, 25 Sep 2019 @ 6:16am

      Re:

      Wouldn't John Smith demand removal of said records?

      reply to this | link to this | view in chronology ]

    • identicon
      Bobvious, 25 Sep 2019 @ 1:47pm

      Re: the 20 000 top results of any John Smith being acquitted...

      As per the comment immediately above yours, those 20 000 results would be recognised as multiple instances of the same result and placed in a folder of links, whereby the folder is a link to another webpage of results. Much like keeping your desktop uncluttered, you place multiple files in a folder rather than spread them all out across the screen. We shouldn't just use the old library card filing system anymore.

      If the court is interested in all the "first links", then THEY can host that system, and all the other search engines can point to that instead of hosting ALL the links themselves.

      reply to this | link to this | view in chronology ]


Add Your Comment

Have a Techdirt Account? Sign in now. Want one? Register here



Subscribe to the Techdirt Daily newsletter




Comment Options:

  • Use markdown. Use plain text.
  • Remember name/email/url (set a cookie)

Close

Add A Reply

Have a Techdirt Account? Sign in now. Want one? Register here



Subscribe to the Techdirt Daily newsletter




Comment Options:

  • Use markdown. Use plain text.
  • Remember name/email/url (set a cookie)

Follow Techdirt
Techdirt Gear
Shop Now: I Invented Email
Advertisement
Report this ad  |  Hide Techdirt ads
Essential Reading
Techdirt Deals
Report this ad  |  Hide Techdirt ads
Techdirt Insider Chat
Advertisement
Report this ad  |  Hide Techdirt ads
Recent Stories
Advertisement
Report this ad  |  Hide Techdirt ads

Close

Email This

This feature is only available to registered users. Register or sign in to use it.