Slashdot Mirror
Google Loses Autocomplete Defamation Case
superglaze writes "Google has been found liable in an Italian court for defamatory comments made against an anonymous plaintiff — the complainant's name, when googled, elicited autocomplete suggestions that translate as 'con man' and 'fraud.' Google was found not to qualify for EU 'safe harbour' protection because the autocomplete suggestions were deemed to be Google's own creation, and not something merely passing through its systems."
That's it. Clearly Italy has shown that it can't handle the Internet. Someone grab me a chainsaw, I'm cutting their fiberz.
... foreign courts are being used for foreign nations to extort money from business they did not produce and had little connection to its success.
Google should stop providing links to Italian businesses.
Oh... Italian court judges.... Well you know what SPQR means - Sono Pazzi Questi Romani (These Romans are crazy)
It seems that 'con man' no longer auto completes for me. I guess I will have to go back to pressing that tiresome enter key.
...they are sure to think that Google doesn't care.
And they shouldn't... Utter nonsense this is. But the idea of controlling speech still appeals to the world's idiots.
For justice, we must go to Don Corleone
Google defended themselves by saying that they shouldn't be held responsible for the output of an algorithm that they created. That's weak.
You left out the key part regarding searches from users. The output is determined by the input, the input is dictated by users.
what about google showing what input the algorithm had for it to give this output?
Algorithms generally don't make stuff up, I'm guessing that maybe, just maybe, some people have been calling that man a scoundrel, a con man and a fraud. THOSE people would then be responsible for their words, because unless he IS a con man, then calling him one is defamation
What? Do you know what would happen if they said that? Every special interest group in the world would have google censor everything that exists.
This isn't an algorithm they created, this is essentially their end product. So of course they should not be liable for what comes out of that. What they should be doing however, is trying to find a way to watch for abuses, but that is not really something that is their job, and more something they should do simply to improve quality of the results.
If you would like to improve your Google search experience, we encourage you to write to your local member of parliament.
Google defended themselves by saying that they shouldn't be held responsible for the output of an algorithm that they created.
Yes! That's the crux of the matter. This decision is not some Luddite rejection of the internet, nor merely holding someone liable for citing an offending publication (as the links to infringing copies == infringing copies reasoning does), but merely holding someone responsible for what they themselves publish.
It does not matter in this case that Google didn't write the defamatory material out "by hand," nor that the data upon which their automated publication was based was not their own. The point is they made their own publication (recklessly, as it turns out) based upon that other data, as you say, by means of an algorithm they themselves created.
Tough to be held responsible for the outcome of one's own actions, ain't it?
Better to be despised for too anxious apprehensions, than ruined by too confident a security. --Edmund Burke
Google defended themselves by saying that they shouldn't be held responsible for the output of an algorithm that they created. That's weak.
You left out the key part regarding searches from users. The output is determined by the input, the input is dictated by users.
And in America, Google has a blacklist of words that are never included in their autocomplete results. They knew this kind of stuff would happen.
If you're so unimportant, why does Google come up with so many pages of results for your name?
Good, inexpensive web hosting
Man, you're lucky. The top results for my name(esp. my full name) are pretty much all porn.
Monstar L
Replying to my own post:
The Google Blacklist
Obviously, all you have to do is hit return to get the results like you always could. However, even when your request isn't blacklisted, you're not getting the SAME results that you would get by hitting return. Entering "murder" into the search bar and hitting a space gets you suggestions of mostly band names. It's only after you hit return that you can learn the other sinister meaning of the word. What we have here is a demonstration of how content can be filtered, controlled, and ultimately suppressed. It is indeed a good thing that Google isn't evil.
The algorithm was designed to keep this stuff from happening. This case was nothing more than a simple oversight on Google's part. Their response was poor and arrogant.
Except Google killed themselves by editing certain material (i.e. anything that deals with Islam).
Associating words in an algorithm to find the next most likely word is not the equivalent to making a statement. It's not defamation.
Google never said "soandso is a con-man", they just offered a service that knows "When people type that name, the most likely associated term is usually con-man".
The court is not thinking this through far enough, or just completely ignorant.
Look on the bright side; if there was a place that hired programmers to work in the nude, you'd be at the top of their list.
How can I believe you when you tell me what I don't want to hear?
Wow. I forgot how many posts I made. I need to get a life.
Google defended themselves by saying that they shouldn't be held responsible for the output of an algorithm that they created. That's weak.
You left out the key part regarding searches from users. The output is determined by the input, the input is dictated by users.
And in America, Google has a blacklist of words that are never included in their autocomplete results. They knew this kind of stuff would happen.
They filtered out stuff that more than likely is going to direct the user to illegal content, that obviously isn't the case in this situation.
*sigh* Same bullshit argument comes up every time. The people who believe and act on hearsy are the idiots, and so are the people that stand up for them.
For justice, we must go to Don Corleone
I hope no translation is necessary... But I'm sure you can Google it.
For justice, we must go to Don Corleone
That's odd. I typed "Italy" into Google's search bar, and the only autocomplete suggestion it has is "Italy is full of douchebag assholes"
Perhaps the Italian Justice System should 'go stick its head in a pig'.
Libel and slander are something that Google should be held liable for no different than anyone else.
AFAIK, in order for something to appear in googles suggestions, someone else has to have searched it first. Google isnt creating the suggestion, its simply remembering the search that someone else did and offering it up.
This really isnt any different than google results turning up libel and slander. Google isnt creating it, its simply indexing it.
Obviously they should refrain from disparaging words and remarks in the auto-complete as they cannot possibly predict or verify if such words are libelous or slander against a person.
I find it surprising that you (and the judge) consider autocompleted text to be potentially libellous - as I mentioned in another post, Google's autocomplete function isn't saying (or even implying) that "x is a fraud" is a factual statement, it's saying that "x is a fraud" is a string of words being searched for regularly.
They make no judgement on the veracity, or even the coherence, of the sentence - they simply present it as something that people have been typing in to the search box. To me, that seems very, very different from Google making or publishing the statement in question.
You know what? It's not google who makes baseless accusations. BTW, how the fuck is an autocomplete suggestion an accusation I don't know, but let's set this aside. It's a typing aid, damnit. It's a piece of code. It's not programmed to target anyone in particular. Heck, for all I care, the output is precisely a statement of mathematical fact -- an output of a generic autocomplete algorithm, whose input was the snapshot of google's database at a certain point in time. You may not like it, but facts aren't libel, and especially not an output of a relatively simple and impartial algorithm. Sure, google's employees coded it up, but the jump from "autocomplete code" to "libel" (it's not slander, duh!) is a long one.
It's like saying that just because your last name is Liaraan, by chance the same as that of the Liaraan, Anders, the crook, whose dictionary entry succeeds the entry for liar, n., you can sue Liber Load and Co, publishers of said dictionary, for libel. I mean, couldn't they just put the Liaraan's entry somewhere less, um, obvious?
If recent slashdot stories are anything to go by, Italian law is seriously fucked up. Seriously.
A successful API design takes a mixture of software design and pedagogy.
Well what if searching for this his name on Google results in the top 10 hits being titled along the lines of "This dude is a con man and a fraud!!"? Is Google responsible for *that* algorithm? After all, the autocomplete algorithm is just another search algorithm, except instead of searching through pages it searches through past inputs.
That's just idiotic. By that rubric, they should also be liable if something comes up in the search results that someone deems offensive. Because they're "publishing" the search results and the titles of all the pages found. Your logic means search engines can't exist. Congratulations on finding a way to ruin the internet.
the italian courts couldn't possible rule against google... unless... *types Berlusconi into google*... I see.
[Brian is writing graffiti on the palace wall. The Centurion catches him in the act]
Centurion: What's this, then? "Romanes eunt domus"? People called Romanes, they go, the house?
Brian: It says, "Romans go home. "
Centurion: No it doesn't ! What's the latin for "Roman"? Come on, come on !
Brian: Er, "Romanus" !
Centurion: Vocative plural of "Romanus" is?
Brian: Er, er, "Romani" !
Centurion: [Writes "Romani" over Brian's graffiti] "Eunt"? What is "eunt"? Conjugate the verb, "to go" !
Brian: Er, "Ire". Er, "eo", "is", "it", "imus", "itis", "eunt".
Centurion: So, "eunt" is...?
Brian: Third person plural present indicative, "they go".
Centurion: But, "Romans, go home" is an order. So you must use...?
[He twists Brian's ear]
Brian: Aaagh ! The imperative !
Centurion: Which is...?
Brian: Aaaagh ! Er, er, "i" !
Centurion: How many Romans?
Brian: Aaaaagh ! Plural, plural, er, "ite" !
Centurion: [Writes "ite"] "Domus"? Nominative? "Go home" is motion towards, isn't it?
Brian: Dative !
[the Centurion holds a sword to his throat]
Brian: Aaagh ! Not the dative, not the dative ! Er, er, accusative, "Domum" !
Centurion: But "Domus" takes the locative, which is...?
Brian: Er, "Domum" !
Centurion: [Writes "Domum"] Understand? Now, write it out a hundred times.
Brian: Yes sir. Thank you, sir. Hail Caesar, sir.
Centurion: Hail Caesar ! And if it's not done by sunrise, I'll cut your balls off.
Live by the sword, die by the sword.
You can't censor some things then cry for "safe-harbor" in other areas. If you're simply a transparent conduit for user-provided data, fine, but Google clearly isn't transparent.
I just tried a couple good-old Anglo-Saxon words. I didn't get any suggestions for "assh" or "fuck" either.
What's more confounding is Google's implementation of the policy cited in TFA by the plaintiff's attorney:
"Google argued that it could not be held liable because it is a hosting provider, but we showed that this is content produced by them (and by the way, they do filter out certain content, including terms that are known to be used to distribute copyright-infringing material), although through automated means," Piana wrote.
However Google's implementation of this promised screening policy turns out to be pretty inconsistent. Using google.com, it's still possible to search for "[name of movie] torrent" and get a list of torrent sites. So I tried a similar search at google.it. The results are mixed. Entering "la strada" prompts for "la strada streaming" (and even "la strada streaming megavideo"!) and takes me to a page of links. "la strada torrent" returns nothing. I tried a couple of other films. "Streaming" is often proposed, but the word "torrent" never appears in the suggested list.
So, if they are implementing the policy announced last December, that implementation differs by country and, within Italy at least, differs by your choice of infringement method.
But google no longer returns autocomplete suggestions for certain things (like torrent and rapidshare) so it is no longer just doing that. It now has some imput from people at google, which means they can now be held responsible (however stupid that may be).
Part of me wishes for Italy to be cut off from search results after the last few cases. Temporarily enraging the populace is a good way of getting things changed (even if it's abusive of their market position).
Another part of me wishes Google would stop censoring stuff and just go back to before where they weren't liable.
The problem is now they have a filtering solution, but they weren't using it well enough.
It would be like an ISP setting up a copyright infringement filter and having it not block everything that's infringing. The blame gets shifted to the ISP for not doing enough.
By modifying the computer generated results to remove keywords they show they can do it but they aren't doing it.
If I search my name, I get the results of a guy who has eerily similar interests and has the same occupation as I do. I have actually had acquaintances ask me if I have moved recently as he lives in another country. The good news is that he is relatively famous within the industry we work, so I can ride on his coattails. :)
Italian Justice == Oxymoron.
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
They should have said "Crap, we screwed up. We'll fix it right away."
But it's impossible to fix. Software isn't cable of discerning "truth", so no algorithm can tell you if something is libel or not.
Almost exactly this. There is a slight difference from the Google results as Google suggests have no links to their source. I find myself considering what would stop the Plaintiff from seeding those suggestions himself?
FTA: "Google lost its bid to claim the protection of the E-Commerce Directive's safe harbour provisions, which partly shields hosting and ISPs from liability for content held on or transmitted over their systems."
Obviously the suggestions shown were generated by Google and transmitted over it's network, as opposed to indexed content not owned by Google. So know wonder Google lost, maybe the lawyer(s) should have argued that the content produced by the auto complete feature is solely based on data supplied by it's users and the internet, maybe the same way a newspaper generates headlines based on what sells!.
Google as big brother, good?
look sig is kool
Google already censors auto complete to avoid offending people. Try typing queries like 'why are nig' or 'how do we get rid of'. You know what would pop up if they didn't have certain filters in their algorithm.
I just tried searching for Silvio Berlusconi on google.it, assuming he is the unnamed plaintiff. It looks like they've already deployed a fix.
young and needed the money?
How about google try to prove that the person is indeed a con man and fraud, probably peddling fake degrees. Show them the power of the internet!
admitting fault is a retarded idea. that would set a dangerous precedent.
especially if there is no fault to admit.
google searches the internet. if the internet thinks such-and-such is a crook, google will report it as such, and doing otherwise would be dishonest.
I typed "anonymous" and the fifth suggestion was Sony. I suppose this is the one that translates to "fraud"?
4 We regret to inform you, that your Google search experience due to the actions of Carlo Piana and by order of the court of Milan. The auto complete function has been disabled for the residents of Italy, due to autocomplete results raising claims of defamation.
One reason you hire competent local counsel is to save yourself from doing something profoundly stupid when you lose a case in a foreign court.
The Italian judge, I suspect, would regard a stunt like this as profoundly disrespectful of Italian law and courts and quite good evidence for a charge of an on-going defamation with malice - an attempt to have your revenge on the plaintiff.
Nor would I expect a member of the Italian Parliament to be any more charitable.
It is useful to remember that Google is not as universally loved and trusted by others as it is loved and trusted by the geek.
Pull the fricking plug. No google for Italy.
The problem is now they have a filtering solution, but they weren't using it well enough.
No, because they are blocking the terms that likely result in illegal content, not the content itself. They can't be expected to check the validity of the results in all legal jurisdictions prior to suggesting a seemingly innocent search term.
They make no judgement on the veracity, or even the coherence, of the sentence - they simply present it as something that people have been typing in to the search box.
That is arguably true. The question is whether or not they make it clear. I understand it. You understand it. But does the average person understand it? And if the average person is expected not to understand it, perhaps it is not a good idea to generate it. I don't know the details of this case, but I can definitely understand the need for an explanation somewhere. I don't see one anywhere.
Having said that, the number one completion for "Google is" is "evil". Number three is "gay". What is one to believe?
Like a shining beacon they show the rest of the world just how destructive and arbitrary a bad justice system can be. Ditto for Brazil.
It doesn't quite work like that.
If you type in "Charlie Smith is" an autocomplete comes up with [starring in a Broadway play], then would you assume that Charlie Smith is starring in a Broadway play? I certainly would.
By the same token then, would you assume that Charlie Smith may have done something disreputable is it came up with [a con man]?
With libel standards of proof depend a little bit in different countries, but generally repeating something that's untrue and you had no reason to suspect is true is still libellous because it still does damage to a reputation and it's seen as the responsibility the reporter to at least make some effort to verify the allegations. Blaming an algorithm isn't a particularly good excuse in this case.
Was the "complainant's name" Berlusconi?
I'd like to have the option to turn autocomplete off. I've found that it often leeds to suggestions that are inappropriate for children.
So we should ban broom handles since it's possible that someone might shove one in your arse, right?
Il n'y a pas de Planet B.
It's not about censorship. It's about the fact that Google is doing more than just presenting the top X most common searches that include the word or phrase you're currently typing - they are already actively removing searches from that list of suggestions. That puts them on dodgy ground in a case like this.
It's official. Most of you are morons.
I wish I had mod points, so I'll reply...
As you said, this has absolutely nothing to do with free speech. I ask those who are criticizing the verdict: what if YOU woke up one morning and found out that Google is suggesting you are a paedophile or a crook, just because a group of determined individuals have been intentionally searching your name with those keywords? From the proceedings, it seems that this guy hasn't even been on trial for the crime of fraud and probably is the victim of his own competitors; he asked Google to remove these keywords and Google didn't comply, so he hired a laywer.
Hack your mind out of its sandbox.
That does not answer the question facing Google and the courts though.
Part of the question (for a rational legal system), is not whether Google originally made the "statement", but whether it is publishing it. And fairly clearly Google is publishing the "statement". So it does not matter that other users typed the search term first. Similar to the way a newspaper can't defend against libel by saying they just reported allegations by someone else that you are ... [insert defamatory claim here].
A more important question is whether a list of search terms on a search page, taken as a whole and in its context, can be understood as a statement or allegation of fact. And here the court seems to have strayed far from the realms of reality.
I don't see that such a list could rationally be interpreted as a statement which one might believe. Thus, no libel. Then again, I'm not a lawyer, so maybe I'm letting commn sense get in the way of a fast buck.
Paul "Say no to feeping creaturism"
... is the type of litigation culture that is flourishing in Italy/other EU countries. It only goes to show how disconnected current legislation is from modern technology.
"I'm taking this loop off." - Jack O'Neill
Italy can seize their E.U. assets rather easily. Google should simply appeal this to Europe wide courts that'll play more fair.
Google might discourage this particular assault by 'accidentally' leaking the anonymous plaintiff's identity, along with all the sites making the accusations against him. They'd need to make sure only American citizens located inside the U.S. are involved in the leak, i.e. no evidence for the Italian court, no disbarment for their lawyer, etc.
The Christian religion has been and still is the principal enemy of moral progress in the world. -- Bertrand Russell
They could allege it was an automated task, and that they had no control over it, until they decided to manipulate it.
Now Google must be held responsible for the tool, as they clearly show they have control on its results.
This comment may contain speech figures. Reader discretion is advised.
Auto-complete presents only popular results, which is different from how search engines are expected to work. The resulting libel may be unintentional, but it is libel in effect, unless google makes it clear that is not a "web search engine" but a "internet gossip search engine".
No, it doesnt, because they STILL never created the suggestion. Answer me this-- Since google also sometimes removes google searches (china for example), are they now liable for all links indexed on their site?
Think carefully before you answer that; a judge saying "yes" would mean the end of search engines as we know them.
of hot coffee on the lap, microwaving the pooch etc... I'm moving to Italy with a Vista powered lap-top - gonna claim a cool few million euros in extreme emotional trauma damages for every BSOD I encounter. Hang on; come to think of it, maybe I could claim attempted murder! Then there's the good old 'a fatal error has occurred' message... I'm in the money!!
The bottom line is that Google is interested in making money, like any other company. Sure, they could cut off Italy -- but that would also mean cutting off a source of ad revenue. I imagine that it would cost more money to Google (in the long run) to cut off Italy than it would to comply with the ruling. Should it come down it this, I would like to think Google would in fact cut off Italy. On the other hand, I don't think it makes practical business.
Well people want to mod me into oblivion here (which is funny because troll != disagree) but:
A more important question is whether a list of search terms on a search page, taken as a whole and in its context, can be understood as a statement or allegation of fact. And here the court seems to have strayed far from the realms of reality.
There are a large number of people that have used Google like Wikipedia and attributed infallibility to its "answers". How many times have you heard, "Just google it"? Or you have attempted to explain a problem and a solution only to have the person say, "But Google says this"?
I think we can both agree Google should be held accountable for libel no different than any publisher and that they are publishing statements. Hardly a position worth a troll mod :)
We disagree on whether it is reasonable to assume that the average person would assume those statements to be true. I feel that is likely given my own experience.
I can be modded troll all day (incorrectly), but I don't think it is that far from the "realms" of reality to believe that the unwashed masses seeing pedophile or murderer next to your name would be unable to rationally conclude that the statements held no veracity.
For better or worse Google and Wikipedia have this "credibility" in common. There are an awful lot of people that will take what Google and Wikipedia publishes as fact without any further investigation or verification of sources.
So... *is* he a fraud? Or does that matter in Italy?
So there are a few good questions here:
First question what is publishing?
If I write a book and sell the copyright to a publisher and they print 50K copies and market them, clearly they not I published the book. They are going to be liable for any libelous statements in the book, I might be as well but they surely are.
Suppose I go the copy shop and run off a 1000 copies of a sheet of paper with "John Doe is a child molester" scaled on it. I then leave them all over the city park. Clearly John has a libel case against me, well unless he really is a child molester and I can prove it. Does he have a case against the local Kinko's? Could we reasonably say they "Published" it, even though all their employee did was take my money and press a button on a computer screen to give me 1000 copier credits?
Next question what constitutes making a statement and how specific does it need to be in a public Internet context?
Suppose John Doe is on trial in an an infamous child molestation case. He has not yet been found guilty. Auto complete aside someone does a search on the name and gets a bunch page hits about how its widely held that one John Doe has committed heinous crimes against children. Could the fact the Google returns these results first be construed to imply they in someway support or affirm the validity of the allegations? If he is acquitted can he sue them? Can some other John Doe sue them because people might confuse his name with the infamous one (they don't seem to be able to in the print world)? In the auto complete case there is not really even a document behind a link to look at how come this guy is able to claim its about him, surely someone else in Italy has the same name?
Repeal the 17th Amendment TODAY! Also Please Read http://www.gnu.org/philosophy/right-to-read.html
http://www.piana.eu/search/node/Carlo%20Piana%20is%20an%20asshole
I8-D
Simple solution which should prevent this happening in the future, is to disable autocomplete for Italian Google. They'll need to live without that feature, but what else can you do when the Italian courts keep slamming down like this?
Must've been him mustn't it?
Or... Barbra Streisand? Definitely got to be one of the two hasn't it?!
was the 'anonymous complainant' Berlusconi, and if so, perhaps google were just being truthful?
Calling me an idiot does not make your argument any more correct and is deserving of a +flame mod.
I am standing up for the rule of law in this case, which is libel.
Now you may think the people who believe and act on hearsay are idiots but slander and libel are legally about what a normal and rational person would believe to be true.
So many times I have stated something that is a fact and based on my expert opinion only have the person question me because of a result on Google. Same with Wikipedia even though the person does not even verify sources.
So you do not need to unduly denigrate me because I am pointing out that this is a case of libel and not oppression or abridgment of free speech. Neither is the action capricious or without cause.
Save your condemnation and vitriol for the unwashed masses that give Google such veracity by default and the legal system's interpretation of libel.
This has nothing to do with what I like or dislike. I am only pointing out that:
1) I support libel and slander laws.
2) The court system determined that it was libelous.
3) That a court going through due process and determining that speech met the definition for libelous is nowhere near the vicinity of censorship, oppression, and tyrannical control of speech, thought, and expression that some would suggest.
All of your observations and opinions about the autocomplete process and the code are out of the ordinary. That is the real issue. Does the average Italian make the distinction between autocomplete representing other user search patterns and Google suggesting what you should be searching for?
That is important because determining what an average and reasonable person would think is critical to the definitions and standards of libel and slander under most legal systems.
I highly doubt that you are or I, or most of Slashdot, represent what the average person would technically understand about Google's algorithm and just what autocomplete is or what it is for. Using that as a basis to determining that Italian law is fucked up is incorrect.
If anything this is the fault of the lawyer for Google for not convincing the judge or the jury about the facts of how Google's autocomplete works.
Which by the way is where your frustration and condemnation really should go. The Italian people. If the judge and/or jury came away honestly believing that Google's feature was reasonably interpreted by the average Italian person as an endorsement or statement of fact, then it is a failure of the Italian people and not the Italian legal system.
Just like World War II Italian tanks have one gear forward and 4 gears in reverse.
All cows eat grass!
Associating words in an algorithm to find the next most likely word is not the equivalent to making a statement.
If the particular association it can be understood by a reader as imputing a meaning how is it not a statement? If it is novel, and the novelty arises from means in your control, it is, for our purposes, an imputation raised by you.
Bottom line, you are responsible for meaningful and especially original (though quotation will not everywhere be an effective shield) "associations of words" you publish.
It's not defamation.
That's not your call to make. It's the court's. The court decided it was. It is. Ain't law easy?
The court is not thinking this through far enough, or just completely ignorant.
Perhaps you are not thinking this through far enough. Who do you think should pay for the damage that Google's algorithm has caused? The victim? The taxpayer? Why not the creator of the offending algorithm!? Personal responsibility goes a long way.
And my guess is, as pertains defamation at Italian law, the court's ignorance will be relatively lower than yours.
Better to be despised for too anxious apprehensions, than ruined by too confident a security. --Edmund Burke
That's just idiotic.
Always a persuasive way to start.
By that rubric, they should also be liable if something comes up in the search results that someone deems offensive. Because they're "publishing" the search results and the titles of all the pages found.
Would that not be a form of citation? Read with greater care!
A pertinent difference here is that they are not merely referring to someone else's utterance here, but are generating an utterance de novo, based on what they may have "heard." Apparently 'gossip' is not an adequate defence to defamation at Italian law.
Congratulations on finding a way to ruin the internet.
Thanks but they are not due. Google is far better at that than I could ever hope to be. ;)
Better to be despised for too anxious apprehensions, than ruined by too confident a security. --Edmund Burke
Now you may think the people who believe and act on hearsay are idiots but slander and libel are legally about what a normal and rational person would believe to be true.
No. It only proves that 'normal' people are a bunch of nosy gossips. And they who the law is protecting. But they are the ones who should be punished. They are the problem. They are the ones who actually pull the trigger.
What you describe may be the law. But note that it is the law that is capricious. Defending it only makes you part of the problem. It is oppressive abridgments of a person's right to speak as he wishes. You have every right control physical action that can be harmful. Speech cannot cause physical harm, and nobody has the right to regulate it in any way. Do not conflate speech and action. You are trying to equate speech with guns. Trying to apply the same restrictions. The only proper response is to tell people who do that to 'go to hell'. I would prefer not to have to include you in that group. I prefer to think you are more enlightened than that. Otherwise, eh...
For justice, we must go to Don Corleone
I'm sure the average italian male wouldn't mind to fuck all the hot chicks that pass him by on the street, but to consider it somehow OK because it's "ordinary"?!
Anyway, are you sure you support those Italian libel and slander laws? Specifically, that certain "deceptive techniques" can be libelous, like (and I cite) clever implication, suggestive combinations, disproportionately scandalized and indignant tone, etc.? That seems like a clusterfuck designed to shut up anyone you disagree with. After reading the chapter above I of course do understand why Google got in trouble, and I don't think they'll have any way out. Alas, if you seriously support that kind of law, then I guess we'll have to agree to disagree...
A successful API design takes a mixture of software design and pedagogy.
Expectations have no effect here. Again, the auto-complete presents a fact. Popularity ranking is simply a fact from the database that the algorithm uses. It does not specifically target anyone, and to consider it libelous you'd have to be seriously drugged. I'm not excluding the last possibility when thinking of Italian lawmakers and courts.
A successful API design takes a mixture of software design and pedagogy.
I'm sure the average italian male wouldn't mind to fuck all the hot chicks that pass him by on the street, but to consider it somehow OK because it's "ordinary"?!
Anyway, are you sure you support those [google.com] Italian libel and slander laws? Specifically, that certain "deceptive techniques" can be libelous, like (and I cite) clever implication, suggestive combinations, disproportionately scandalized and indignant tone, etc.? That seems like a clusterfuck designed to shut up anyone you disagree with. After reading the chapter above I of course do understand why Google got in trouble, and I don't think they'll have any way out. Alas, if you seriously support that kind of law, then I guess we'll have to agree to disagree...
A successful API design takes a mixture of software design and pedagogy.
Type 68 + 1 into Google, then tell me it has a dirty mind. Interpretation of a string of words created by statistical association without specific intention cannot be defamation.
You might as well argue that the dictionary is implying cricket players are criminals because it put the words so close together.
Type 68 + 1 into Google, then tell me it has a dirty mind.
What relevance do you imagine that might have to this discussion? Why would Google's "mind" be relevant? The question simply is: has any imputation been raised by a publication of the company that might cause damage to a person's reputation. Try to keep your replies pertinent!
Interpretation of a string of words created by statistical association without specific intention cannot be defamation.
That is obviously false. Google was found to have defamed plaintiff on precisely these grounds. Are you not paying attention?! Or do you just get off on making authoritative sounding, but obviously counter-factual pronouncements? And how have you come the the mistaken belief that tortious defamation generally requires "specific intention" or any intent at all?
You might as well argue that the dictionary is implying cricket players are criminals because it put the words so close together.
Now you are just being silly. If someone is considering doing business with you goggles your name and gets suggestions "yourname conman", "yourname fraudster", "yourname paedophile" it is possible, even likely, that these "strings of words" will lower their estimation of you. And, in fact, they may be wise not to do business with you, after all Google is telling them that yourname is associated in some relevant way with the suggested terms. And you're 3vi1 too!
People look up words in the dictionary mainly to understand their meaning. It is understood that individual words are arranged strictly alphabetically without any regard as to their relevance to one and other. Could you to find a less apposite comparison to the situation under discussion?
Turns out this law stuff ain't so easy for you non-lawyers after all.
Better to be despised for too anxious apprehensions, than ruined by too confident a security. --Edmund Burke
You're being purposely dense. The plaintiff's entire suit is based on narrow unwritten interpretation of data as a statement. If your going to mentally add '=' between the words, you might as well add 'is not' between them, or "strawberry yogurt" even.
If Google's output seems to associate their names with fraud, it's only because of input data. The persons (or charlatans) who don't like it should contest it with the creators of that data - not with Google.
Why can't the "defamed" do that? The most logical conclusion is that there's nothing wrong with the data. Do you represent the plaintiffs?
>> Turns out this law stuff ain't so easy for you non-lawyers after all.
Meanwhile right, wrong, and simple understanding of technology seem to continually baffle many lawyers.
You're being purposely dense.
No, I'm just trying to help you understand what is going on here.
If your going to mentally add '=' between the words, you might as well add 'is not' between them, or "strawberry yogurt" even.
Don't you think that's a tad disingenuous? If I were to write the words "3VI1 FUCKTARD" very few people would in fact read strawberry yoghurt in between them. The question is not whether reading '==' between the words is logically valid, the question is simply whether the association of words could influence a reader to lower their opinion of you.
If Google's output seems to associate their names with fraud, it's only because of input data.
Agreed. But the mere fact that any defamatory allegations a person makes are direct quotes offers no protection of itself (not even in most common law jurisdictions). Some additional protection (as exists at CL for fair reporting of court proceedings for instance) is needed. My understanding (and I am not an Italian lawyer, so beware) is that Google could have availed themselves of free harbour protection, but for the fact that they were not merely hosting the utterances of some other, but were actively synthesising these utterances into what amounted, in the view of the court, an original utterance capable of bearing defamatory imputation.
The persons (or charlatans) who don't like it should contest it with the creators of that data ... Why can't the "defamed" do that?
They can.
instead
The most logical conclusion is that there's nothing wrong with the data.
Up till now your overly strenuous application of logic has been the source of your misunderstanding. Here logic has apparently failed you. Remember that mere truth is not everywhere a complete defence to defamation (nor should it be IMHO, but that is another argument). While I am unaware as to whether other defendants were in fact pursued, but the most obvious (logical or otherwise) conclusion is that Google is both easier to locate and has substantially deeper pockets than other potential defendants.
Do you represent the plaintiffs?
No. I'm admitted, but not to an Italian court. Moreover I don't practise, I develop software.
simple understanding of technology seem to continually baffle many lawyers.
instead
OK, with this I can absolutely agree!!! However, and this might not be immediately obvious, there is actually no question of technology per se to be considered here. What has to be considered is whether certain "associations of words" are capable of bearing a defamatory imputation (yes) and whether Google's creation de novo was covered by safe harbour provisions (no).
This is actually a case where the judges have not been baffled by the technology. Let me give you an example of a result that really irks the geek/lawyer in me:
In my jurisdiction larceny (a criminal offence) requires inter alia a non-consensual asportation. If you arrive at a bank with a sawn-off shotgun and demand money and take it off premises that's a non-consensual asportation. If otoh, a bank teller (being an agent of the corporate person of the bank) mistakenly hands you too much cash, it is consensual (and there is case law to that effect). You will (probably) owe the bank that money, but you will not be criminally liable.
Now there was a case (citation escapes me at the moment) where an ATM handed over more cash than it should have. The recipient was convicted of larceny because the judge deemed the ATM spewing out cash to be non-consenual on the part of the bank. In my view the preferable result would have been to see the malfunction of the ATM, installed and programmed by human agents of the bank, as directly comparable to a teller mistake, as expressing the bank's (mistaken) will. But the judge, I submit, was baffled by the technology. Though the bank should be able to rec
Better to be despised for too anxious apprehensions, than ruined by too confident a security. --Edmund Burke
>> If I were to write the words...
Then you had intent, whereas an algorithm has none. Suing Google for the output of a generic algorithm is the Internet equivalent of suing gun manufacturers for homicide. [note: I don't care about the reasons that's a legal course of action, so save the time you were going to spend on that, I think it's stupid to do as long as manufacturing guns is legal.]
Simplest put, the part about "because the autocomplete suggestions were deemed to be Google's own creation" from the summary is wrong beyond words. By that logic, we can sue Google and any other image search engine for copyright infringement all day long [note: I also don't care about your opinion on those cases, they're stupid as long as robots.txt exists.]. Or, conversely, write software that minimally recompiles existing software (which would still probably be a lot more complex of an algorithm than Google's autocomplete) to sell as our own.
I'm not interested in your understanding of legal sophistry; I'm interested in pure right, wrong, and beauty. Altering an algorithm because one party does not like the data the world has associated with them is not a beautiful solution, and therefore it cannot be right.
Nothing you've said has changed my evaluation or opinion in the slightest - so we might as well agree to disagree.