Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
×
The Courts Government The Internet News Technology

Attorney Sues Website Over His Online Rating 207

An anonymous reader writes "The Seattle Post-Intelligencer is reporting that a local attorney is suing legal startup Avvo over a rating that was algorithmically assigned. The story covers the controversy of computers grading humans. 'Browne, who has participated in a number of high-profile cases in the state, including the defense of arsonist Martin Pang, said in an interview that Avvo is being irresponsible with the ratings and called them a fraud. And he questioned why Supreme Court justices and prominent lawyers score so low. Three other attorneys interviewed by the P-I also expressed doubts about the rating system, while News.com reported that the site "seemed to be riddled with bizarre errors."' Such practices are not new: the New York Times earlier this year reported on Google using algorithms to determine applicant suitability. But what happens when you don't like the result? Can a computer program be considered defamatory?"
This discussion has been archived. No new comments can be posted.

Attorney Sues Website Over His Online Rating

Comments Filter:
  • Just another tool. (Score:5, Insightful)

    by Harmonious Botch ( 921977 ) * on Saturday June 09, 2007 @06:47PM (#19453967) Homepage Journal
    From TFA:

    Can a computer program be considered defamatory?"
    Programs can not be defamatory. Their output may be.

    If I use a hand-held calculator to get a result, and then publish it and that publication defames someone, I can't blame it on the calculator.
    In this case, a computer is just another tool used to calculate something - perhaps a tool that many people don't understand as well as they should - but a tool nonetheless.
    You use it, you take responsibility for the results. You don't understand how it works? Hire a consultant. The fact that it is a complex tool does not excuse you if you libel someone.

    ( NB: The above paragraphs presume that there is indeed libel, a fact not yet proven.)
    • Nice sig (Score:3, Funny)

      by Anonymous Coward

      Parent's Sig: When members of a profession start referring to non-members as 'laymen', it is time to start shooting them.

      Shooting whom? The members of the profession or the laymen? If you're going to call for someone's head, you should at least be a little more specific about whose head you want served to you on a silver platter.

      • Re:Nice sig (Score:5, Funny)

        by mybadluck22 ( 750599 ) on Saturday June 09, 2007 @07:06PM (#19454075)
        We can solve this with a simple u-substitution.
        u = "start referring to nonmembers as 'laymen'"

        When members of a profession u, it is time to start shooting them.

        It becomes clear, now.
      • by camperslo ( 704715 ) on Saturday June 09, 2007 @07:14PM (#19454147)
        Shooting whom? The members of the profession or the laymen? If you're going to call for someone's head, you should at least be a little more specific about whose head you want served to you on a silver platter.

        Those who call lawyers useless aren't being very open minded. Lawyers may be a melamine-free source of protein, but I think my cat would prefer something a bit less bony than the head.

        I wonder if the computer program has a way to rate them on flavor?
        • by thegnu ( 557446 ) <[moc.liamg] [ta] [ungeht]> on Saturday June 09, 2007 @07:23PM (#19454215) Journal
          Those who call lawyers useless aren't being very open minded. Lawyers may be a melamine-free source of protein, but I think my cat would prefer something a bit less bony than the head.

          I had to get a lawyer because my landlord was a slimy, conniving liar. And it's turned out to be some of the best money I've ever spent, because before I was spending loads of time doing all the paperwork and covering my own ass. Now the lawyer does it. And he's a nice guy, to boot. Shocking.
          • Re: (Score:3, Funny)

            by Tychon ( 771855 )
            I prefer pants to cover my ass, but if a lawyer does the job well enough for you and handles your paperwork, something I'll admit my pants have never done, then I may have to look into this.
          • Re: (Score:3, Funny)

            by vertinox ( 846076 )
            And it's turned out to be some of the best money I've ever spent, because before I was spending loads of time doing all the paperwork and covering my own ass. Now the lawyer does it. And he's a nice guy, to boot. Shocking.

            Defense Lawyers = Good
            Other Person's Lawyer = Bad
            Prosecutor Lawyers = Really Bad
            Corporate Lawyers = Really Evil
        • Re: (Score:3, Funny)

          by flyingfsck ( 986395 )
          "a melamine-free source of protein" Slime has protein?
    • by pimpimpim ( 811140 ) on Saturday June 09, 2007 @06:58PM (#19454021)
      As far as I understood it, a person is rated by a website with some rating procedure the website worked out. So your example doesn't really hold. Someone else used a method on the calculator to calculate the result, and he complains about the method.

      I don't know what to think of this. As a restaurant, you can earn michelin stars based on the grades you get from probably several testers. Did you ever hear of a restaurant that sued michelin for loosing a star? It doesn't seem to make sense.

      Rmember, any person, magazine, or website can grade services, but they will only be taken seriously when they have a decent method. Systems like this work when not only the public has faith in the method, but also the people involved (e.g. the chefs), respect the way the testing was done. If restaurants would think that the michelin system is fake, they wouldn't be proud of their stars in the first place.

      Since there seems to be quite a group of people that do not agree with the method used by that website, they can of course try to sue them, but I figure that the website will be rendered useless within the trade fast enough that they might as well just ignore the score of the website all together. Sueing might even be counterproductive, I didn't know before that the website existed in the first place. And I think he has very much the right to do so. He never asked to be rated, but at the same time the rating will be of high economic importance to him (getting more high-profile jobs, etc.). Now you could compare this with a restaurant getting the famous michelin stars, it can make or brake the restaurant. Now In cases like that, you need to be able to ask for a second opinion.

      • by Harmonious Botch ( 921977 ) * on Saturday June 09, 2007 @07:22PM (#19454209) Homepage Journal

        As far as I understood it, a person is rated by a website with some rating procedure the website worked out.
        All by itself? Just a website doing things on its own?
        Some human made a decision somewhere.
        • by Mistlefoot ( 636417 ) on Sunday June 10, 2007 @12:21AM (#19455699)
          And if my algorithm says that if your first and last name start with P you must be a PedoPhile can I call you one and hide behind math?

          I think not.
          • Re: (Score:3, Interesting)

            by Fordiman ( 689627 )
            Sure you can. And every person named Peter can call you a fake, and every person who sees your method can tell the world that you're a fake.

            The problem here isn't the math, or the low ratings. The problem is idiots like you who don't understand that an industry rating depends on the faith of the people and the respect of the industry.

            If the site gets no faith or respect, the site will fail.
        • Re: (Score:3, Funny)

          by inKubus ( 199753 )
          You've never been a developer have you? When I read this article, I immediately thought: "Great, something else they can blame IT for."
      • by Sancho ( 17056 ) on Saturday June 09, 2007 @07:32PM (#19454293) Homepage
        I'm thinking of a number between 1 and 100. Can you guess what it is? Give up? It's my Pagerank score, and mine happens to be 86.

        That's what this situation reminds me of. Credit scores (which is what the above lines reference) are calculated by an algorithm, and they can be wrong. If they are, you can get the company to adjust them, provided that you can provide evidence that their input is faulty. I see no reason that this case should be any different, provided the inputs (reasons for the ranking) are well known.

        In the case of Google, however, they are providing a ranking rather than a rating. It is much, much harder to objectively rank webpage relevance based on search terms, and even harder to know whether a given ranking is justified when you don't even know the algorithm used.
        • I hate those commercials. In my head, I think "He's thinking of a number between 20 and 80. Do you know what it is? It's his goddamned IQ!"
        • by timeOday ( 582209 ) on Saturday June 09, 2007 @10:18PM (#19455113)

          Credit scores (which is what the above lines reference) are calculated by an algorithm, and they can be wrong. If they are, you can get the company to adjust them, provided that you can provide evidence that their input is faulty.
          If, on the other hand, the credit rating company has all its facts right, but you don't like their algorithm and think you're still credit-worthy even though they don't? Well you're screwed. How to combine numbers is their prerogative. Think about how that applies to this case.
    • Re: (Score:3, Funny)

      by Dunbal ( 464142 )
      When members of a profession start referring to non-members as 'laymen', it is time to start shooting them.

            I think layman is a lot better than vulgar, don't you, you vulgar person?
    • by buswolley ( 591500 ) on Saturday June 09, 2007 @07:12PM (#19454129) Journal
      Programs can't be defamatory? Well their code can be?

      Have you read some of the comments people put in source code?

    • by QuietLagoon ( 813062 ) on Saturday June 09, 2007 @07:43PM (#19454373)
      Programs can not be defamatory. Their output may be.

      I disagree. Programs are merely an extension of the human(s) who designed them, the program does what the human(s) told them to do. Therefore what a program does is the full responsibility of the human who designed it.

      I always laugh when a programmer tells me, "there's a bug in my program." My first questions is always, "well, who put that bug there?" Programmers talk of bugs as if they just magically appear, and are not the result of the programmer's error(s).

      The comment that programs cannot be defamatory smacks as specious at best. Of course, programs can be defamatory. Programs are written by humans, programs are computerized extensions of humans.

      • by SnowZero ( 92219 ) on Saturday June 09, 2007 @08:33PM (#19454637)
        Agreed. You can encode whatever you want into an algorithm, and it can certainly be biased and even illegal. For example, in some applicant rating system for a hiring "screen" program, if you have:

        if(applicant.sex == female) rating = 0.0;

        That would certainly violate equal opportunity laws, unless the company could prove that the output of said program was not used at all.
        • Funny you should mention the trivial case above, I was going to post the same as an example of a clearly "illegal" program, but GP made me think of something else... What if we modify it to be:

          if(applicant.sex == female && will_halt(random_bytes())
          || applicant.sex == male && !will_halt(random_bytes()) rating = 0.0;

          -- making the one having to prove the violation of the equal opportunity law (presumably NOT the company using the test!) also being able to violate the halting theorem [wikipedia.org]!

          Should we j
          • mmh i dont think so. Regardless of the outcome of the haltin theorem you have: A = appplicant.sex B = will_halt(random_bytes()) then your logical statement is AB + A!B --> rating =0 A(B + !B) --> rating =0 and since (B + !B)== 1 (always, regarding of the validity of the statement B) A --> rating =0 which is the original post.
            • by PaulBu ( 473180 )
              Note that there are two calls to random_bytes(), thus, depending of the actual implementation of will_halt(), it if actually gets into the infinite loop when it can not decide (and test would have to be aborted) the code above gives some (50% or 25%?) preference to females over males (assuming that the halting theorem is true, of course, and there is 50% chance of running into a non-halting program, which is most likely untrue!). In other words, there are two uncorrelated B and B' and one has to make other
          • actually they could just make the program run N times with N being sufficiently large, such as 5000 times, and show that statistically the program created a bias against women.
            • by PaulBu ( 473180 )
              We are talking about the Law here, not statistics!

              If it is statistically proven that white guys rape black girls significantly more often, or asian girls murder russian ones significantly more often (tried to put as many random inversions as I could, and, I'd guess, still failed to satisfy the PC crowd :) ) -- it should (in theory) have no weight on whether this or that *actual* violation took place...

              But good try, next one, please! :)

              Paul B.
    • Re: (Score:3, Insightful)

      by Snaller ( 147050 )
      "You use it, you take responsibility for the results. You don't understand how it works? Hire a consultant. The fact that it is a complex tool does not excuse you if you libel someone."

      Ranking someone based on your critera is not libel.
    • by 2.7182 ( 819680 )
      My Lisp code says you are a troll! (Just kidding....)
    • by Torvaun ( 1040898 ) on Saturday June 09, 2007 @09:10PM (#19454811)
      "Programs can not be defamatory. Their output may be."

      Oh yeah?

      int main()
      {
         int x = 2; //Anonymous Coward is an incredible tool.
         return 0;  //The number of balls Anonymous Coward has.
      }
    • by Firethorn ( 177587 ) on Saturday June 09, 2007 @09:18PM (#19454859) Homepage Journal
      The fact that it is a complex tool does not excuse you if you libel someone.

      But is it libel? I see very little difference between this and consumer reports - Both take data and attempt to draw a simple rating/conclusion from it. If the lawer wins his case it undermines the whole independent review and ratings system - because anybody giving a negative rating would be open to a lawsuit.

      How they weigh their ratings is up to them, just as it's my choice as to how heavily I weigh their ratings (anywhere from not bothering to read them to simply attempting to get the highest rated).

      Errors in the data is one thing - for it to be true libel it would probably have to be deliberate. As they get more data into the system, improve accuracy, etc... The ratings should become more accurate.
    • by Joebert ( 946227 )

      Can a computer program be considered defamatory?"

      No, but the criteria the computer program uses to determine the rankings can be.
    • by neoform ( 551705 )
      Does that mean I can sue the mirror company for making me look so fat and ugly?
  • by WrongSizeGlass ( 838941 ) on Saturday June 09, 2007 @06:53PM (#19453999)
    ... if John Henry Browne does sue Avvo's computer ranking program it will make him a bad lawyer and thus the ranking will have been a self fulfilling prophecy.
    • Re: (Score:3, Funny)

      by dwater ( 72834 )
      > ... if John Henry Browne does sue Avvo's computer ranking program it will make him a bad lawyer and thus the ranking will have been a self fulfilling prophecy.

      Unless he wins (this is in the US, right?), in which case they'll be obliged to increase his rating.

      "Think I'm a bad lawyer? I'll show you!"...."See, I told you I was a good lawyer."
  • imagine what would happen to consumers who were negatively affected by soundscan's demonstration of the buying patterns of human beings and its effect on the decline of popular culture in music? we could recoup millions in damage to lost souls.
    • Re: (Score:2, Funny)

      by loudersoft ( 1113511 )
      *damage: people who bought britney spears and avril lavigne records because they thought higher soundscan numbers meant "cooler" only to later discover that they had fallen victim to the fallacy of computer data
  • by MightyYar ( 622222 ) on Saturday June 09, 2007 @07:00PM (#19454033)
    There's nothing wrong with computer rankings, but they inputs are very, very important. We shouldn't pretend that they are different somehow from human rankings, since humans still carefully select the inputs.

    An example that most here can relate to is the US News and World Report college rankings. It's a whole other topic in itself, but suffice it to say that there is a lot of discussion about their inputs and how it has influenced the way colleges operate. Most colleges try to get many small donations instead of a few big ones, because the rankings weigh number of donors more heavily than total amount donated. They encourage many, many applications from just about anyone because they get ranked based on the number of applications that they reject.

    Once people learn what the inputs are, they just game the system.
    • by Dymus ( 595974 ) <dymus&hotmail,com> on Saturday June 09, 2007 @07:31PM (#19454275)
      And it isn't terribly difficult to determine what the inputs are. I am ranked at 8.1 (only having been licensed for two years) which is higher than all but one of the attorneys practicing in the same field as me at my firm (ranging between 25 and 6 years of practice). After reviewing all of our AVVO pages, it appears my rating is boosted significantly because AVVO discovered I received two CALI awards in law school. Anyone in the business knows these awards are more or less meaningless in practice (especially since the two I got have nothing to do with my practice area), but they appear to have a significant impact on my score. Based on our informal analysis, it looks like you start with a baseline of around 5, get a boost of about 1 for every award they found for you, and reduced a ton if you have any bar disciplinary actions pending. I played around with it by updating my profile with organizational memberships and even got endorsed by an attorney friend of mine and it had no impact on the score. The key seems to be how much the AVVO system can find on you on its own, without you updating it. So more important than what the inputs are may be determining where AVVO is farming its data from. Overall, it seems to be that the criticism the ranking system is receiving is pretty valid.
  • Ugh... (Score:4, Insightful)

    by rakslice ( 90330 ) on Saturday June 09, 2007 @07:01PM (#19454035) Homepage Journal
    Is it just me, or does it seem like there are a lot of legal professionals who normally have no problem applying existing law to novel situations but who turn into drooling idiots as soon as a computer program or computer network becomes involved?

    • by suv4x4 ( 956391 )
      Is it just me, or does it seem like there are a lot of legal professionals who normally have no problem applying existing law to novel situations but who turn into drooling idiots as soon as a computer program or computer network becomes involved?

      Apparently not. They'd be drooling idiots if they just said "well, it's a computer rating. They're expected to be wrong, I'm fine with it" and did nothing about it.
    • Re: (Score:3, Interesting)

      by mangu ( 126918 )
      ...who turn into drooling idiots as soon as a computer program or computer network becomes involved

      It happens because human laws have a lot of flexibility that computers don't have. When human logic turns against them, people respond with "don't be a fool, you know what I mean". You can't do that with computers.

      It's a simple fact that this guy is not as good as some other lawyers, according to a given set of objective factors. If they change the algorithm to improve his rating, then what about other lawyer

      • Computer algorithms aren't flexible,

        What? Of course they are flexible. They can be programmed to calculate whatever you want them to, and they can be changed on a whim.

        they cannot be bent so that everyone comes out looking good.

        Huh? Of course they can, you just program them to make everybody look good.

  • by jfclavette ( 961511 ) on Saturday June 09, 2007 @07:01PM (#19454039)
    Uh....

    switch (person.getSkinColor()) { ... } ?
    • switch (person.getSkinColor()) { ... } ?
      It really depends on what you're doing with it. For example, if the program is recommending what hue of skin cream (or made-to-measure suit, or something else like that) will work best with your colouration, then that program that requests that info for such purposes can hardly be inherently defamatory. On the other hand, it is most certainly possible to use a program to defame someone: it's not that hard to conceive of ways to do it (e.g. by spamming libellous statements all over the place.) But all this is programs, not algorithms; I've no idea what a defamatory algorithm might look like, and I don't think it is possible to make mathematics work that way.

      As an aside, I really doubt that skin colour is going to be expressible as a simple enumeration or integer, so you won't be able to switch on it...
      • You seem to be using a non-standard definition of an algorithm (programs are a superset of algorithms, the primary difference being that algorithms must terminate while programs may or may not). There's no reason an algorithm can't take somebody's skin color as an input and do something based on that input.

        Anyway, I think the only way an algorithm (or program) would be considered defamatory is if you didn't precisely specify what the output meant. For example, if your algorithm said white lawyers got a valu
        • "programs are a superset of algorithms, the primary difference being that algorithms must terminate while programs may or may not"

          Program = Algorithm + Data. - Knuth [wikipedia.org].

          "if you didn't precisely specify what the output meant"

          If you expand that to include input, you will have described what is otherwise known as - GIGO [wikipedia.org].

          BTW: A subjective judgement as to wether the "garbage" output by a particular program is/isn't defamatory is outside the realm of computer science.
    • Uh....

      switch (person.getSkinColor()) { ... } ?

      Fixed:

      try {
            switch (person.getSkinColor()) { ... }
      }
      catch( CaucasoidFeaturesException ) {
            switch (person.getReligion()) { ... }
            switch (person.getCountryOfOrigin()) { ... }
      }
      finally { person.warilyAccept() }
  • Defamatory? (Score:5, Insightful)

    by piojo ( 995934 ) on Saturday June 09, 2007 @07:09PM (#19454109)
    Well, if they made it clear to viewers how their ratings are calculated, they should not be responsible for harm done. In that case they would just be stating facts (e.g., rating = this lawyer wins X% of his cases - this lawyer charges %Y percent over the industry average for their type of cases...) But if they don't tell people where the ratings come from, then I wonder: how is writing a shitty algorithm that says defamatory stuff about people any better than just saying defamatory stuff about people. People are responsible for the computer programs that they knowingly use.
    • Re: (Score:3, Interesting)

      Mod parent up, that's a great point. If you can't defame a person using a Computer Program output based on an algorithm then people can just design algorithms to defame people and have no repercussions.

      Everyone in RL judges people based on algorithms in their heads, "He didn't do as well as person Y, and charges more than Z". Typical defamation cases are thrown against people who have messed up algorithms, "He wears white all the time...I hate white" or poor input, "I've only seen him lose". Why shouldn't a
    • if they made it clear to viewers how their ratings are calculated, they should not be responsible for harm done.

      Yes, but when you rate lawyers, they will find loopholes in loopholes and mess with you anyhow. Remember that even if they don't win a case, they can make you drive all over, pay attorneys, and hang out in court-houses to counter the charges. I once tried to start up a school bulletin board-like website, but I soon found out that the legal ramifications associated with children can be humongous.
  • by 313373_bot ( 766001 ) on Saturday June 09, 2007 @07:11PM (#19454117)
    1) Attorney sues website that assigned him a low rating.
    2) Attorney loses, and his rating goes even lower.
    3) ...
    4) Profit!

    Ok, 3 and 4 aren't really necessary.
  • ....Computer says nooo .
  • FICO (Score:3, Interesting)

    by Belial6 ( 794905 ) on Saturday June 09, 2007 @07:45PM (#19454387)
    Hasn't this been done to death with people suing over FICO scores?
  • one of my favorite recent (last ten years or so) Simpsons scenes:
    Online Auto Diagnosis Doctor [using AOL voice]: You've got... Leprosy!
  • "And he questioned why Supreme Court justices and prominent lawyers score so low."

    Looks like he deserves a low rating to me. Are we sure he's a real lawyer? Or does he just somehow avoid all discussions pertaining to jurisprudence?

    • "And he questioned why Supreme Court justices and prominent lawyers score so low."

      Well, it could be the whole '1' is the best, '10' is the worst type scale, but I figure that any SC justice is probably going to score low because he or she hasn't operated as an attorney for quite a while.

      As for the prominent lawyers- I can't be sure. I don't know how they rank stuff.

      Thinks I'd figure on:
      Years in Practice
      Number of cases
      Success rate
      Appeal rate
      Charge Rate
      Any malpractice/suits against the lawyer/bar judgements.
  • You gotta love the lawyers letter regarding how well he does at getting people off Sexual Misconduct charges. If I ever rape anyone in his neck of the woods I'll know just where to turn.
    • by WilliamSChips ( 793741 ) <full.infinity@gma i l . c om> on Saturday June 09, 2007 @09:22PM (#19454887) Journal
      Yeah, because nobody is ever falsely accused of rape and therefore anybody accused but not convicted is a rapist walking the streets...
    • Meh. The job of a defense attorney is to get his client off the hook. Whether the client actually did it or not isn't relevant to him. That is only relevant to the other side. The last thing we need is for defense attorneys to end up working against their own clients merely because they are convinced by the prosecutor or plaintiff. If that happened, the defendant wouldn't get a fair trial. Our system is adversarial: both sides zealously fight one another. You don't want that to stop working, I assure you.
  • Petty (Score:2, Interesting)

    So maybe this John Henry Browne deserves this rating? Perhaps he has a penchant for spectacularly losing cases for his clients thereby destroying the lives of hundreds of families. Or maybe not. According to my rating system, Supreme Court Justice Antonin Scalia would get a rotten rating as well, but that's all this is, an opinion. Even though there are undoubtedly fancy-schmancy algorithms behind avvo.com's ratings, these functions and formulae are no more concrete than a set of rules and opinions deve
    • So maybe this John Henry Browne deserves this rating? Perhaps he has a penchant for spectacularly losing cases for his clients...

      Brown typically represents people accused of high-profile crimes such as murder and rape. Most (all?) of his clients are guilty. Given the trash he represents, it wouldn't surprise me if he has a high loss rate. Everyone deserves a lawyer, and in Washington, when you've chopped up a few people, killed a few in an arson, done a rape or two, if you've got money, you get Brown. You

  • My online rating (Score:2, Insightful)

    by Anonymous Coward
    If this comment doesn't get modded +5 insightful, I'm going to sue Slashdot.
  • by belmolis ( 702863 ) <billposer@@@alum...mit...edu> on Saturday June 09, 2007 @08:10PM (#19454509) Homepage

    It sounds like the rating algorithm isn't very good, but I don't see how this guy can win a suit for defamation. In US law, only false claims of FACT are actionable. If the web site stated that a lawyer had received a reprimand from the bar association when he had not, that would be defamatory. If it said he had cheated a client or bribed a juror and he had not, that would be defamatory. But saying: "This guy is a jackass" or "This guy is a poor lawyer" is not actionable because these are opinions.

    I'm not sure what can be made of the use of a poor algorithm. If they disclose the algorithm and say "Here is what we get when we plug in the data we have", so long as the data is accurate and they apply the algorithm correctly, they aren't making any false claims of fact. Ethically, it seems like there should be a penalty if they persist in using an algorithm that demonstrably does not produce output that is reasonably related to what people generally take to be valid measures of lawyer quality and if they deceive people into thinking that it is valid, but I'm not sure how this can be addressed legally. I think you'd have to argue that there is an objective definition of lawyer quality of which the algorithm gives a false view. I don't know if defamation has ever been proven on such a basis.

    • I suppose, if the website publishes their methodology, then the rating number becomes a statement of fact. For example, if the foo rating is claimed to be equal to 10*fraction of cases won, then posting an incorrect foo rating is a false statement.
    • Actually, you shouldn't rely too much on the idea that couching a statement as an opinion will protect you. The states vary on this, IIRC, but it has been rejected as a matter of First Amendment law. The real issue is whether it is defamatory and there's nothing about an opinion that prevents this.
      • Couching something as an opinion does not protect you if what you claim is actually a matter of fact, but a statement that really is merely opinion is indeed not actionable. That's black letter law.

  • If you think the algorithm is defamatory, pick it apart and show how it isn't accurate.

    If the algorithm has a rule that says "if lastname=Browne then rating=rating-100" that's obviously a bogus algorithm.
  • by Angst Badger ( 8636 ) on Saturday June 09, 2007 @08:44PM (#19454691)
    Let me get this right -- these AVVO folks decided it would be a good idea to base their business model on saying things about attorneys that might not be complimentary?

    This is quite possibly the first time anyone thought they could make money by being sued constantly. Anyone who thought that the dot-com bubble used up all of the reservoirs of stupidity may now rest assured that fresh reserves have been discovered.
    • If you don't lose, being sued a lot is a good way to keep in the public eye. The advertising dollars must be just rolling in.
  • by Infonaut ( 96956 ) <infonaut@gmail.com> on Saturday June 09, 2007 @09:25PM (#19454897) Homepage Journal

    This guy is a public figure. In order to successfully claim libel, he must prove malice [eff.org]. An algorithm can't have malice against him in particular. This guy is out of his depth. To paraphrase an old saying in the legal profession, a lawyer who represents himself is an idiot.

    • by Phroggy ( 441 )

      This guy is a public figure. In order to successfully claim libel, he must prove malice [eff.org]. An algorithm can't have malice against him in particular.

      Sure it can. This one probably doesn't, but if it does, he'll win his libel suit.

      Of course, if the algorithm does have malice against him, it didn't get that way by accident.
  • These "Lawyer sues X" stories are boring. Lawyers sue people. That's all they know how to do. It's like reading a front page story about bakers baking, teachers teaching, or politicians sucking.
    • Re: (Score:2, Funny)

      by rcani ( 831229 )

      or politicians sucking


      It says something about American politics that the first thing that came to mind involved Bill Clinton and an intern.
  • by kinglink ( 195330 ) on Saturday June 09, 2007 @09:48PM (#19455007)
    Seriously, can an algorithm be biased? YES! Very much so. Imagine if we had an algorithm that rated people on a 100 point scale. If your skin color is white you get 100 points to start with, black people get 0 points. That's biased.

    Does that mean the company behind the algorithm is biased? Yeah again. In this case (not necessarily Avvo's case) this algorithm is blatantly biased, trying to rate people on their skin color.

    Does this mean it's illegal? Not unless the law has changed. If he wants to litigate then he needs to prove the algorithm is biased (and a few anomalies doesn't mean it's biased, it means it has a flaw) Avvo has to be biased in such a way they are making a profit over the difference. Avvo has to be deliberating trying to damage someone's career for it to even be illegal.

    If the algorithm is running correctly and there's no X factor (meaningless stastical values, such as the color of the skin) then there's no crime here. They might not have a perfect algorithm but they arn't claiming it.

    If the lawyer in question wanted to fix this correctly, he should bring this to the attention of the site, point out numerous cases of people being graded too harshly and then publicize the data to the public if need be. From the sound of it, there's little to no proof except some lawyers feeling they are being treated unfairly. From the sound of it, that's sour grapes, there's no defamation.

    Just because an algorithm is poorly designed it doesn't mean it's crap, errors happen even with great algorithm's first iterations. There's still a burden of proof on the lawyer and I'm not hearing any real proof yet.
  • Credit Ratings (Score:2, Insightful)

    by llthomps ( 470748 )
    So, if this attorney wins his lawsuit - does that mean I can sue FICO for my a credit rating?
  • And people use them too, frequently without having any idea of whether the algorithm is processing given input the way they're thinking it is... or having any idea what the inputs are. Labor saving devices are great, but it's nice to have an idea of just what laborious action one is saved from performing, otherwise a monkey can do your job. Hear me, HR rep?
  • I recall a story about a medical expert system, related to me by the famous AI researcher Donald Michie (from the UK), that was designed to determine whether a patient with particular heart problems needed heart surgery or not. The expert system was proved to do a better job at predicting when heart surgery would increase the quality of the outcome (life) for the patient than specialists doctors (cardiologists?).

    IIRC, the hospital chose not to use the system for fear of litigation when the expert systems d
  • No matter what your profession, half of your colleagues are below average. That could include you. But in most professions, you don't have to be the best, or even close to the best, to be better than an unknown quantity. This is why those dim bulbs still surround you.

    You know how dumb the average guy is? Well, by definition, half of them are even dumber than that! -- J.R. "Bob" Dobbs

    Mal-2
  • by Animats ( 122034 ) on Saturday June 09, 2007 @11:50PM (#19455547) Homepage

    We've had to face a similar problem as Avvo with SiteTruth [sitetruth.com], which rates web sites. The answer seems to have two parts - integrity and transparency. This means looking at information that comes from reliable sources other than the thing being rated, and showing the information from which the ranking is derived.

    Avvo is trying to do this. Avvo's information comes partly from external sources, like legal directories and records of disciplinary actions. That's less game-able than traditional web search. And Avvo shows that information, so they have transparency.

    Google is slowly coming around to this point of view. Originally, Google rankings were opaque, but now they've put in various "Webmaster Console" features to show some of the information that drives their algorithm.

    Google faces the problem that some of their metrics for detecting junk web sites are heuristic, and rely on "security through obscurity". They don't want to say exactly how obscure text can be before it's considered "hidden text", or exactly what they consider a "link farm", or they'll be spammed right up to the allowed limit. So they can't have full transparency. They're inherently limited by the approach of primarily looking at the web site itself, which the site operator can change freely, to rate the site.

    Google does look at some external non-Web information, but mostly things like how long a domain has been registered.

    Avvo has user ratings of lawyers, which probably aren't that useful. User ratings are most valuable when the universe of raters is much larger than the number of things being rated. So it's good for major movies, where there are tens of new movies and millions of fans, marginal for hotels, and weak for businesses few people have heard of. There aren't enough clients per lawyer to get a statistically valid result, and it's too easy to game when the number of raters is small.

The 11 is for people with the pride of a 10 and the pocketbook of an 8. -- R.B. Greenberg [referring to PDPs?]

Working...