ICO: It's up to Google the 'POLLUTER' to tidy up 'right to be forgotten' search links Google and its rivals are meeting privacy officials in Brussels today to discuss a recent European Court of Justice ruling that calls on search engine providers to de-link certain listings on their indexes. They are expected to look at practical implementations of the judgment from the European Union's highest court in May. It …
Interesting approach.
So if I point out (or provide an index to) crap floating in a stream, I should pay, on the grounds that you might not have smelled it otherwise?
Or perhaps if I tell you who has broken the DPA, for example if I was the Office of the Information Commissioner, and you wouldn't have noticed otherwise, I should pay the fine?
To be clear, I'm not saying that Google shouldn't clear things up - but the analogy being used is pretty poor.
The analogy is poor on multiple fronts. On one hand, this is blaming the river for being polluted. On the other hand, what they call "pollution" is what Google thinks people want to read. The reason the links appear in results is because users liked them, and have clicked on them.
By just calling it pollution and blaming it on Google, the ICO are avoiding their job, which would be to define what is pollution in the first place…
But hey, if they want Google to decide what is in the public interest, they only have to say the word. I'm sure Google will be very happy to go back to displaying everything like before the ruling happened.
By just calling it pollution and blaming it on Google, the ICO are avoiding their job, which would be to define what is pollution in the first place…
But hey, if they want Google to decide what is in the public interest, they only have to say the word.
they don't appear to want that either, just they want google to take care of it without giving up their right to complain later that google's doing it wrong...
I think the analogy would be that once the stream has been cleaned up, you would be responsible for updating the index (which is most people's primary decision making tool for choosing a picnic spot) so that the stream is no longer filed under "smells pooey"
Yes, but the stream isn't getting cleaned up. That's the problem. Google keeps checking, and keeps finding the same thing. So the solution being thrown about is: rather then actually cleaning the pollution (that's too hard, it seems the entire community is dumping raw sewage in the stream), blame the people who make an index of polluted places.
"Yes, but the stream isn't getting cleaned up."
I wish I had multiple upvotes for this.
This is where my problem with the ICOs (frankly ridiculous) take lies. What the ICO is arguing for amounts to a cover up of the informations existence, nothing more. Its similar to suggesting that ordering the press to stop reporting corruption, government ineptitude, whatever would mean the problem didn't exist - and we've been there before. The original page Google linked to is still there, and probably referenced by multiple other pages, links, quotes etc, and that is what needs cleaning up, after which the "Google problem" will go away. Its the page that needs removing, not googles link to it, but perhaps thats an epic can of worms they can't quite get their heads round.
The ICO (and possibly the court) seem to either not understand this, or to be attempting an easy 'fix' for a complex problem, tinged in the ICO's case with a little bit of nastiness - "... making millions and millions out of processing people's personal information". And thats relevant why? I'm as ready to belt Google with a heavy stick as the next man for their rampant, cynical abuses of privacy, but in this case its not about the 'personal information' processing they directly make money out of, it's about a service which makes the web both usable and useful.
Notwithstanding there's a 'public interest' factor to be accounted for, money usually cicumvents that for exactly the people it shouldn't quite nicely - remember 'Trafigura - the pollution that didn't happen' after the lawyers waded in? That's what really, really bothers me personally, that this leads to censorship for those that can afford it. It'll end up as "you can't index this article, because it mentions X as director of the Big Corrupt Polluters Ltd". So not only does X get to go on with a blemish free character in the wider world, but Big Corrupt Polluters Ltd get a tidy little image makeover as a free gift.
I blame the court in this case less than the ICOs specific take, which strikes me as very a very lazy approach on almost every level that looks suspiciously like "you deal with it, I can't be arsed, but maybe going all finger waggy on yer will persuade others I'm right" - quite in keeping with the usual bureaucratic penchant for treating the public like six year olds.
I've previously viewed the ICO as merely ineffective and a bit pointless, but it looks like I'm going to have to add 'lazy and willfully obtuse'.
I have quite an uncommon name - there only appears to be 2 of us in Internet land.
Recently, a search on my own name returns a list of results with a message at the bottom saying that some results may have been removed under EU data protection laws.
I've never requested this, I'm going to check "the other me", but I suspect my name has appeared in articles along with someone else's name, who themselves have asked for the articles to be removed from Google's results.
Thing is, doesn't this look like I have something to hide?
Well for all those upvoting the above that have nothing to hide. How about we install some CCTV in your house so we can see what you get up to? After all, if you have nothing to hide...
The problem with what Google are doing, is that it makes individuals subject to the whims and prejudices of others. I have nothing to hide, I've requested no data be removed, but if the HR bod from the company I had a job interview with believes I have, and that candidate B hasn't, then I get overlooked.
So I have nothing to hide, but may well be treated differently by others due to the way Google has handled this.
"So I have nothing to hide, but may well be treated differently by others due to the way Google has handled this."
Not if, as I suspect will happen (if not already), the warning appears for all searches for names. If that happens, Mr HR Bod from the company you had an interview with will either think all of the candidates have something to hide, or he'll put two and two together and realise the warning is effectively meaningless.
FWIW, I've just fired a fictitious name into Google's search box, and it gave me the warning.
"I have quite an uncommon name - there only appears to be 2 of us in Internet land.
Recently, a search on my own name returns a list of results with a message at the bottom saying that some results may have been removed under EU data protection laws."
I have a very uncommon name - I'm 99.9 recurring percent sure that there is only one of 'us' in internet land, and I get the same message.
"I've never requested this, I'm going to check "the other me", but I suspect my name has appeared in articles along with someone else's name, who themselves have asked for the articles to be removed from Google's results."
No, if this someone else asked for something to be removed from the results, it wouldn't affect a search on your name (unless you searched for your name and theirs. I think Google are (or were) gradually adding that message to its search results when they include a name.
I suspect it's so that it becomes a generic warning and can't be used to infer that any given individual has asked for something to be removed. (Although another possibility is that they're doing it to make people wonder, and to try to reinforce the belief that this whole business is breaking the interwebs.)
The warning shows up for practically any search result that looks like a name. This is Google's way to "be transparent" and warn users that the "pure Google results" may have been tampered with, without actually pointing the finger to the individuals who asked for results to be removed. They tend to act a bit like a Prima Donna on the subject.
But despite the warning, chances are nothing has been removed at all from results when you search for your name, and in the unlikely event that it has, it would be for articles that are about the other guy who has the same name.
This post has been deleted by its author
Sorry, but no. That is incorrect.
If there is a web page mentioning John Smith and Joe Bloggs, and Joe Bloggs asks Google to remove the link from search results for his name, the web page will still show up when searching for John Smith, as it will for any other queries that are relevant, such as "bothering a goat" or such.
"No, this isn't about blocking search terms, this is about removing links. If you are mentioned on the same page as someone else whose request to remove that page was granted, then that removed page won't show when you are doing a search for yourself.
I didn't suggest it was about blocking search terms. Blocking a search term would result in them saying "There are no results available for [what you searched for]" - and that's not what they do.
They 'forget'/remove/de-index a page for the name of the person requesting it (if they agree it's correct to do so). That means that when you search for that person, the page doesn't show up: it's no longer linked, in their vast amounts of data, to that person's name1 (and by extension, any search term that includes that person's name). The link remains in that same data, though, for every other search term that might lead to it.
That this is what happens was demonstrated only a few weeks ago - where the forgotten link could demonstrably still be found by searching on Google. It was clearly still indexed for the person the article was about, so the only likely explanation was that it was de-linked from the name of an unidentified person in the many comments.
1. In reality, the link is probably still there for the name of whoever asks for it to be forgotten, but with a flag (or some other mechanism) to identify it as a 'forgotten' link - otherwise it risks being picked up again by the Googly spiders.
You should be rejoicing and putting a few quid aside to buy the person who got those entries removed a pint or two.
Frankly, there is far too much data all all of us out there and anyway to reduce that pile is a good thing as far as I'm concerned.
Anon for bindingly obvious reasons
I don't know where you're getting that figure from but for our main site (where we don't use Google Analytics to count the hits so we know it's not them biasing the results) 97% of the search traffic comes from Google. 2% from Bing, the rest are in the noise. Google really is the only game in town.
"Google should remove links that are old, out of date or irrelevant and - significantly - found not to be in the public interest."
What is old?
Do you want Google, Microsoft and friends to decide what is irrelevant or not in the public interest?
Is there some standard test for what is in the public interest?
However, the ruling continues to be a load of crap.
How is Google the polluter?
Who said that the court told Google that they had to come to these conclusions themselves?
If there is any doubt the solution is simple: forward the complaint to the ICO and let them determine if it should be removed. Job done. No need for a single lawyer in that case.
All this 'too much work' crap is a red herring put out by Google who would presumably rather have no legal obligations whatsoever if it means interfering with their business.
It's like trying to stop the phone from ringing by removing all the phone books.
If information is out of date, then it is the out of date information that should be removed, not the links to it. The links will disappear if the information is removed.
Politicians really don't get it, do they
> If information is out of date, then it is the out of date information that should be removed, not the links to it. The links will disappear if the information is removed.
That's the ideal, but good luck getting any information removed from some anonymous website based out of Kerplakistan.
It seems to me that this is a taste of the problems the copyright lobby have been complaining about and fighting for years.
So you'd be OK with ihatevelv.com posting "velv smells" in some jurisdiction with bullet-proof hosting?
And Google making it the top link for searches on "velv" because the host pays for advertising?
And the papers printing "Official: velv smells", <very small print>according to ihatevelv.com</very small print> ?
because, hey, it's not their fault. And even if it's not true now, any prospective partner should know that velv is really a stinky person underneath who just cleans up a bit better now, right?
Google should grow a pair. They got away with not being classed as publishers so they couldn't be sued for republishing libel, but if they're digging in bins and displaying their findings in front of the town hall, eventually enough people will get pissed off to take them down.
Google have to redact now because they are now a Data Controller of stuff in search results, and thus it is the law to do so.
I'm sure Data Controllers also have the duty to ensure personal information they hold is *accurate*.. so.. are they now ordered to clean up who the web thinks you are? It's only the redaction that must occur at *your* prompting.
Are they now obligated to collect more data about you to ensure they *are* accurate now?
Will the EU assist them in this obeying of the law? The EU already have your data, and obviously think Google is worthy of Data Controller status.. *maniacal cackling*
There are businesses in the EU who create or hold information dossiers on people (credit reference agencies, headhunters, etc). Those businesses are subject to strict laws about personal data processing (including rights to have old or incorrect information removed from the dossier), which create non-trivial costs for them.
The decision seems to be based on the interpretation that a Google search of a name creates, in real-time, a similar sort of dossier on a person. You can argue whether that is a sensible interpretation, but I do have some sympathy with it: I can see a future (with some smarter Google algoritms) where a Google search could replace a credit reference check.
If that interpretation is valid, then clearly Google need to be subject to the same data protection laws and processes that the other dossier-makers are subject to. Including the right to have false or old information left out of the dossier. And they should clearly have to bear the cost of that, just like the credit reference companies do. Just because their process constructs the dossiers in real-time instead of cumulatively over the years, doesn't change the rights of the subject of the dossier.
That's the problem, Graham.
As daggerchild says above; what credit agencies and the like are asked to do - by the same law now being applied to searchable web-indexes (not just Google) - is to make sure the information they hold about people is accurate
That's how much of a mess everyone appears to be getting themselves into.
*
It's also interesting to note that it is no longer - officially? - Just Google in the firing line.
That being the case - as someone put above, bing can barely claim an EU market % - will El Reg's search engine be next?
"As daggerchild says above; what credit agencies and the like are asked to do - by the same law now being applied to searchable web-indexes (not just Google) - is to make sure the information they hold about people is accurate"
In the original test case the information was accurate and that was never disputed. Thus we have a law requiring the hiding of fact rather than the correction of error. That's not a good thing.
Robert & Donn, you may not agree with laws about censorship and about having to remove factually correct information from dossiers, but that is the law in the EU. I realise it is not the US way, but in the EU personal data is strictly regulated and being "fair" to people trumps freedom of speech -- not the other way around. For example, it is a true fact that a person who lived previously in my house went bankrupt. However, as that person is in no way related to me, credit reference agencies are not permitted to record that information, even though it is true, as that might adversely affect my credit score.
If that information was on the internet, and someone did a Google search of my name and used that information when making a decision to give me credit, shouldn't I be able to prevent Google making that visible? If not, wouldn't that allow Google to compete unfairly against regulated credit agencies?
Today, the credit reference agencies know the rules and apply them: they make the decision, not a court, unless you disagree and sue them. Google will need to set up a similar process -- after a few borderline cases are decided by the courts, it will all settle down.
"For example, it is a true fact that a person who lived previously in my house went bankrupt. However, as that person is in no way related to me, credit reference agencies are not permitted to record that information, even though it is true, as that might adversely affect my credit score."
Well, that's different as the information here is not about you, what's being banned is not the information but of the unjustified association of two unrelated facts. If you had gone bankrupt, this law would protect you, and that is wrong.
To broaden out the problem: when we send people to prison it is ostensibly to "make them pay their debt to society" and we're supposed to put it all behind us. There are good reasons for this in the general case.
However, all those trials were reported and are now in newspapers' digital archives where they can be searched for, if not by Google then using the papers' own website search functions.
There is nothing in this ruling that protects those newspaper databases except a naive hope that no judge will ever rule that those records are not in the public interest. Hoping that no judge will ever do something stupid is like hoping that a catwalk model won't do hard drugs.
So we're faced with the possibility that all digital records of news will come under a concerted attack at some point in the not too distant future by wealthy people with dodgy histories who want to re-write their reputations for future generations. How long before Bill Gates starts erasing the history of MS's illegal dealing with OEMs, for example? He's spending a fortune to buy a reputation already, so why not? Immortality is clearly important to him.
And, what about Steve Jobs if he was still about? Don't you think he'd be keen to get rid of all those nasty reports about how he treated his daughter? After all, that's not public interest and it's all in the past etc. etc. thank you, kindly, Mr Beaky. Luckily he's snuffed it but he's exactly the sort of arrogant bastard who would jump at the chance to re-write his official record. There's plenty of others.
This post has been deleted by its author
This guy keeps missing the point:
"The ruling states that search engines such as Google should remove links that are old, out of date or irrelevant and - significantly - found not to be in the public interest."
The problem is because Google makes the first decision we have essentially fobbed off the decision on "in the public interest" to a private company.
From the article:
"It has been widely misread in the media - partly because it has been spuriously referred to as a "right to be forgotten" ruling, but also due to Google's efforts to wrongly claim that the decision amounted to censorship of the interwebs."
Interesting bit of spin -- but it is just that: spin and misdirection.
If a government (or in this case the ICO) says that I cannot say or publish something (in this case a link to a page) then it is, most definitely, censorship. It is cut and dried censorship, with no ambiguity.
Whether the censorship is justified is something that can be debated, but the fact that it is censorship cannot be. Thus, the question is now why this article's author is trying to misdirect our attention by publishing falsehoods?
The polluter pays, the polluter should clear up.
Except that it's not clear what's pollution and what isn't. Coming to those sorts of conclusions should be up to the ICO. Good luck to anybody trying to get them to do it though given their complete inaction in the past over anything google/search related.
"All this talk about rewriting history and airbrushing embarrassing bits from your past - this is nonsense, that's not going to happen."
Except that the specific case the ECJ made this decision upon is exactly that.
This whole thing is insane:
1. Google is a search engine, not a data host. If the data were removed from the data host, it would not show up in Google searches. The ECJ saying that the newspaper can leave the data up, while saying that Google must not link to it, is incomprehensible.
2. Once some thing is posted on the internet, it's over. You could fine the company, but keeping a lid on the information is pretty much a moot point, especially if you draw attention to it by requesting that it be removed.
3. Links that are "old" or "out of date" can be very useful. If there is bad information about something or someone, and I am searching the web for information about it to decide if I want to interact with it, past transgressions are absolutely relevant, even if the actual issue has been resolved, or perhaps exactly because the issue was resolved. And if a link is "irrelevant" or "not of the public interest" then by definition, people will not search for the info, so there is no need to bother.
4. Google is not going to be able to adequately process all of these requests, especially given how unclear and subjective the criteria are.
Last, it's clearly time we looked at what data it is acceptable to publish without someone's permission and hold data hosts accountable to that. This will be an important right in the future, and one I wish existed right now. Going after a search engine like this has probably set the cause of privacy back a bit.
Google is not going to be able to adequately process all of these requests, especially given how unclear and subjective the criteria are.
Talk about processing 'all of those requests' is a red herring put out by Google who'd rather not have any obligation whatsoever to remove results, even when there is a clear reason to do so (and in any case it shouldn't be up to Google to decide what's in the public interest or not - if memory serves the court never said the Google had to fulfill this role either).
If something is unclear then the solution is simple: get Google to instruct the complainant to take the matter to the ICO. *THEY* are the ones that should be coming to such conclusions - not Google, or any other search engine.
The ICO and the technologically inept idiots employed there just don't want to do so - it might shine a light on their own incompetence on anything IT related after all.
@cored
It's not insane -- although it can be very confusing if you don't think in the way that data controllers (under EU legislation) are supposd to think.
1. Facts are facts. In general, you won't get a fact removed from somewhere like a newspaper (if it is true). There is no right to be forgotten.
However, Google is not a data store (as you point out). Google is processing data to provide a service: you type in a search term and Google collects information about that subject and tells you. That is where the legislation kicks in. If the data relates to a person, there are laws about processing it. They certainly aren't perfect but they protect us every day from people abusing information about us.
Amongst those laws are implictions on using data processing to create a profile of someone. As search engines were not envisaged in the laws, it has taken legal arguments to decide what the restrictions are on a profile created by searching the web (and it is perfectly reasonable to believe the decision is wrong -- but it has been made). The decision is that it is similar to other commercial companies which create profiles, like credit reference agencies. They are not allowed to include irrelevant or obsolete data, except in cases where dropping that data would be against the public interest (for example, someone standing for parliament is not likely to be able to get data about criminal offences dropped, even though the offences have expired -- they may be able to get them removed once they stop being a politician, however).
2. Your point only matters for well known people, where it is more likely that there is a public interest argument for retaining the data anyway.
3. Yes, old data can be useful. Lots of things could be useful which are not allowed. In the case of data about people, data protection overrides utility. Get over it.
4. Yes they are. Once Google stop moaning, they will put in place a process, using advice from data protection experts. A few more cases may go to court to get some grey areas sorted out. Then the process will just work.
This issue isn't publishing data, it is processing the published data and creating profiles of people. Google searches do it automatically. What I don't know is whether the same rules would apply to manual data processing. For example, if you were to look at all my postings on El Reg, gather some personal data from those (maybe I have said where I live or how old I am or something), create a Wikipedia page for me and publish that information, would I have the right to get old or irrelevant data permanently removed from that page? I don't know.
Google is not creating a profile of someone. They don't have special access to data like a credit agency.
They don't know if you pay your bills on time, or have a mortgage, or how much you owe on your car loan, they will not know you bounced a cheque unless it went to court and became a public record. A credit agency has very specific types of data from specific known sources. Google has millions of bits of random crap, that's why you need a search engine in the first place. Your comparing apples and rocks.
Only famous people will get anything like a profile because there is so much public info available on them, but you still will not know how much they owe on there credit cards.
If you google my name there is nothing that is actually me until page 17, and you would still have a hard time telling it was me if you didn't know me. On page 19 there is a story about a guy that streaked a football game, wrong city and a few years too young but might be bad if someone thought it was me... can I request that it be removed? How would google know it is/not me?
Well put, Graham.
A very good debate/discussion, I think (spoiled just a tiny bit by a bit of unnecessary negative voting?).
I think the real issue is brought out in the posts here - I'll try to summarise:
1. There is no right to be forgotten (but some in the EU would like one)
2. So an EU court has decided that Search Engines (the court made a legal definition as to what one is and does) are data processors (because, as part of their business, they hold data about people)
3. A search engine is therefore subject to laws governing data processors.
4. These laws were originally intended to stop companies - who process, store and make-available data about EU citizens - holding "out of date information".
3a Especially relevant where "time served events " - like bankruptcy - are involved.
(this was done because citizens found once they had been "black listed", they could never get credit again)
And so, the problem is that an old law, not intended for use in the industry it is being targeted at, that was written before the vast majority of people on the planet had ever heard of the Internet, is now being used to "bash Google" because drafting new laws is too hard (and to be totally fair, Google needs bashing much more often).
However, little good ever comes of bad law nor bad implementation.
There is should be a big debate for society about the trade off between the Right to know and the Right to be forgotten. And in particular who should get to decide, on a case by case basis, between the two necessary evils.
And - For non-US citizens in particular - there are issues around (US) companies holding data about them, without being held accountable to the same laws to which they are held accountable.
However: What with massive media misrepresentation - both ways - spoiled tantrums from the IT industry and poor knee jerk reactions from the EU legal apparatus; sadly I don't think there's much chance of having a quality debate (despite the fact that we have had one here).
At which point in time does history become 'out of date' I wonder? Will all events in the EU now have to come stamped with a 'Use By' date? What will the Hindus make of that?
Someone will have to be given control over what in the EU is deemed 'out-of-date', 'irrelevant' or 'not in the public interest', and thus be denied to the public. I wonder who that will be? Google? Or a court specifically instructed by the EU to make decisions that will increase the EU's power over the people?
Despite outward appearances, this is not ludicrous bureaucratic stupidity; it is carefully-crafted groundwork for a censorship regime, masquerading as well-meant lunacy. The political muppets that strut on the EU stage may be bumbling idiots but those pulling their strings are not. Anyone who seriously believes this has been done for the benefit of the people should stop taking the soma.
And the polluter in this case is whomever published what the search engines indexed. If people want to crap removed, they appeal to the publisher (or just sue) to have it removed. It is then beholden on the published to have an indexes/caches cleared at the search engine. Oh, and The Internet Archive.
I simply do not see how any of this should be down to the search engines. They just index the content of others.
by suppressing what they knew people allowed the likes of Jimmy S. (and more recently Rolfy) to get away with what they were doing for years.
Why should we allow people to erase their own online history? Most of the time it gives important clues to who people are, how they're likely to act, and what positions they can be trusted in.
I'd put anybody who makes such a request on a watch list.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
