Website liable for Google-generated page summary A Dutch website has been sued — and sued successfully — for the way Google summarized one of its pages. As reported by De Telegraaf and the Dutch blog 24 Oranges, a judge has ruled that the website Miljoenhuizen.nl is liable for a Google-generated page description, or “snippet”, that may or may not give the false impression …
Certainly their appeal should centre on the fact that they've taken the 'offending' page down and yet Google still shows the 'offending' description. Google's cache of the page has an abrupt "Article not available" message and yet the snippet shown still has the previous text.
Interestingly, the live page is quite different to what you see if you try and run it through Google Translate. It seems that Google's IP addresses are no longer welcome. I wonder why...
Bonkers attitudes from all sides, methinks.
I remember, way back when the Internet was accessed through modems and only available in sepia-tone, the de-facto response to anyone who claimed 'they read it on the Internet' was, 'you shouldn't believe everything you read on the Internet!'
Seems now we are expected to believe content that's not even generated by humans.....
....never heard such utter rubbish in my life. The judge should be....shot like a .....stuffed animal...and his balls removed....and served roasted with a dash of pepper
Sorry, seemed to taking snippets from a page with current news, a toys R us advert, a recipe and a guide to child behaviour.
This is, as the first poster says, bonkers.
So the judges has ruled that one cannot create a webpage that contains two words that may concievably be placed near eachother if the page is concatenated by an algorithm, with the source words that driver the algorithm, yet to be provided by an unknown human user, if the resulting computer generated "sentence" may offend another unknown human.
I love it when "judges" have no understanding of how technoology works :)
I read about this in the Dutch press a while back. The verdict does surprise me, and not exactly in a good way. Sounds like the judge missed a couple of important bits about who does what on the internet. I really do hope they appeal, otherwise anyone hosting a website has got his work cut out trying every combination of words in Google that might be slightly sueable.
In order to avoid a zonking great law suit web site owners now have to check that Google doesn't misrepresent their site in its *generated* summary. I assume that the same also applies to other search engines as well.
Bonkers. Totally bonkers. Judges, like politicians should be kept away from IT like small children are kept away from matches.
if the only sanction the court has imposed is to reword the page then its probably not that unreasonable, its often worth looking not just at the guilty/not-guilty bit but what the judge imposes.
I've had a check and can't see so its possible the judge is still about to slap a £10k fine on the original site, but if its as simple as please change the words then whats the big deal?
Does any vaguely intelligent person form opinions based on what they see on the google summaries? This has to be a step dumber than using wikipedia as your sole source of knowledge.
Continuing in a similar vein, this would imply that anything a reasonably skilled splicer could do with things an individual had said would result in them (the individual, not the splicer) being sued. Crazy!
Google "!snippets" don't just contain words from the webpage that is the search result ... Google algorithms can return matches on the basis of words occuring in the page and/or in pages that contain links to that page - in this latter case it is possible for the snippet to contain text from the pages that link to the result. Quite often when searching for multiple words/phrases I'll find a result where the snippet shows text that seems to inidicate its what I'm looking for but when I go to the page its not there and then if I go via the google cache link I find it tells me that that phrase is only found in pages linking to this page.
So, does this mean every webpage author can be liable for text in any other webpage that has been indexed by google?
So if a newspaper has a story about someone (famous) on top of one page and an article about bankrupcy at the bottom, should we be allowed to sue the newspaper because it was skim read and created a totally (false) report after joing the two different (unrelated) items together?
Or maybe if I fold the page in a certain way (like you used to fold pages in MAD comic) I can make up my own stories and sue the newspaper for false allegations!!
This is proper mental!
Where in one of the theads a garage was being slaged off for being poor, however some recomended another local garage in the same thread and subsequent google searches for garages in that area (not bad or good ones just ones in the area) turned up the page summary with the good garage aparently being called poor.
When the garage reported this to us we deleted the offending thread (it was old), and informed google via their automatic system that that page should be removed from the their cache, it happened within 48 hours. I pressume the speed of this was because we use google analytics so they knew it was the websites admin that was asking for the changes.
In our case it was quickly and easily sorted out with everyone being happy,
I hope the Dutch media picks up on this case and highlights the shocking level of incompetence displayed.
Also, regarding folding tabloids to create new stories, It's always seemed to me that they do this themselves. Often I've glanced at a paper and seen Public figure X accused of Something Bloody Awful only to notice on closer inspection that these are really two independent stories simply positioned rather unfavourably on the page, the cynic in me suggests these frequent juxtapositions are all about sales.
The two substantive parts of that are surely provided in the search itself? So if they typed in Zwartepoorte NOT bankrupt, would the snippet then still be libellous?
In other words, the libel (if any) comes from the person searching and this is an idiotic judgement. If I choose to skip every other word of a newspaper article and the result is libellous is that my problem or the papers problem?
There are many posters to El-Reg comments ... Most of them complete and utter arseholes, obnoxious freetards, and quite a few paedophiles and baby killers amongst them.
Obviously no one here's going to be offended because it's plainly clear that the second sentence is completely unrelated to the first.
"Also, regarding folding tabloids to create new stories, It's always seemed to me that they do this themselves. Often I've glanced at a paper and seen Public figure X accused of Something Bloody Awful only to notice on closer inspection that these are really two independent stories simply positioned rather unfavourably on the page, the cynic in me suggests these frequent juxtapositions are all about sales."
Tabloid papers are usually more careful than this, as they know the dangers. In extreme cases, and even with regard to completely unrelated stories, there is precedent for juxtapositions between stories on the same page being grounds for libel.
If it is thought that people's concenration span only extends to reading the google summary of a page they found by typing in two or three words into the search box - without actually following the link to see what it actually says - then I now understand why people Twitter. I'd add another sentence along these lines, but I fear losing my readers' atte...
Who's old enough to remember the Not The Nine O'Clock News sketch where the judge had never heard of a digitial watch or a video cassette recorder? This seems to be the problem we have with the legal profession and cases based on technology.
If, for example, a newspaper had misquoted another publication would the judge have found the original publication was at fault? Of course not, but as soon as the internet comes into the case common sense flies out of the window.
Does this mean that all Dutch webmasters have got to look at every possible combination of words on any of their pages and make sure no misleading snippet could be generated if somebody were to type that word combination into Google?
Presumably if this was any different search engine, this ruling would never have happened.
There's a weak argument that this is Google's fault, but when you visit Google and type in your search string, there's no expectation that you will be given anything other than links and summaries (and adverts). The people who got confused simply didn't take the time to understand the tool that they chose to use, and that's their look out.
To decide that this is the website owner's fault implies that Google is a fundamental part of the Internet, which, of course, it is not.
In my experience on three continents I found that most judges wouldn't know which end of the roll, toilet paper comes from let alone understanding newfangled contraptions like the Inter-net and P.C.s. Logic, intelligence or even common sense appear not to be prerequisites for high judicial office anymore, if they ever were.
Anyone who thinks that I'm just lawyer bashing (or who would like to have a chuckle or be shocked or both) is advised to have a squiz at the voluminous legal judgements in Western jurisdictions, which are now freely downloadable. Look at a few from Texas for a start, especially the capital appeals judgements. These guys are not of this earth!
(And if you think our Dutch friend is off with the pixies, have a look at this bird.
http://www.abcnews.go.com/WN/story?id=3724883&page=1
)
Although the ruling may seem odd, it is perfectly understandable:
1) The website owner represented himself without a laywer
2) The website owner told the judge that technically they would be able to 'fix' the problem by changing their page
As a result the judge ruled that the website owner should fix the problem because he himself said he would be able to do that. Also AFAIK the judge did not rule on guilt, just on fixing it.
If the website owner had used a lawyer, he would probably never have done point 2, and then most likely the judge would have ruled differently. So the person to blame for this ruling is not the judge, but the stupid site owner.
Yes it is stupid to represent yourself but it shouldn't be!
I once represented myself in court for a driving offence. I was guilty and went to court and said so. The judge threw the book at me although it was a first offence. The prosecution didn't look at me once.
Jobs for lawyers, that is about the only thing the legal system is good at! Why should ordinary people have to fork out extortionate rates for lawyers on what should be a simple case.
A simple, easy and widely documented way to stop search engines (not just Google) indexing any specified page on your site (or, if you're ultra-paranoid, the whole lot!).
Sure, it won't stop it indexing third party pages that link to your site (as in this case), but it could be a useful standby if your company's preferred lawyers are technophobes...
"There are many posters to El-Reg comments ... Most of them complete and utter arseholes, obnoxious freetards, and quite a few paedophiles and baby killers amongst them.
Obviously no one here's going to be offended because it's plainly clear that the second sentence is completely unrelated to the first."
Jason, if your point has any merit, it points to a fault with the way google represents the summary, not any fault with the original website owner.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
