Stories
Slash Boxes
Comments

News for nerds, stuff that matters

Slashdot Log In

Log In

Create Account  |  Retrieve Password

Amazon Patents Customized 404 Pages

Posted by CmdrTaco on Wed Jan 30, 2008 08:50 AM
from the makes-my-heart-hurt dept.
theodp writes "Among the patents awarded to Amazon CEO Jeff Bezos on Tuesday was one for his invention of Error Processing Methods for Providing Responsive Content to a User When a Page Load Error Occurs, which covers displaying alternate web pages in response to HTTP 404 page-not-found errors. So is this the technology that causes Amazon's Home Page to be displayed when Bezos' MIA Patent Reform Page can't be found?"
+ -
story

Related Stories

This discussion has been archived. No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
 Full
 Abbreviated
 Hidden
More
Loading... please wait.
  • The Plan (Score:5, Funny)

    by suso (153703) * on Wednesday January 30 2008, @08:50AM (#22233614) Homepage Journal
    They were going to sue everyone on the planet, but they couldn't find their websites.

    And by the way, what constitutes "customized" when its open source software?
    • Re:The Plan (Score:5, Informative)

      by Actually, I do RTFA (1058596) on Wednesday January 30 2008, @11:33AM (#22235216)

      And by the way, what constitutes "customized" when its open source software?

      Their patent doesn't have anything to do with 404 pages. Their patent covers the specific case of having multiple error pages corresponding to cashed version, or closest name for a page, etcetera -and- a client side component that says failures load alternate version X. The client-side component may be prt of the browser. But the important thing is that the error type is user-settable./p

      • Re:The Plan (Score:4, Informative)

        by ashridah (72567) on Wednesday January 30 2008, @12:17PM (#22235824)
        closest name for a page
        So... mod_speling for apache would be an accurate representation of prior art of some of that patent, then...

        ash
        • So... mod_speling for apache would be an accurate representation of prior art of some of that patent, then...

          Nope. Or rather, only if the user got to choose whether the server would use mod_speling -or- an alternate method. The patent is about letting the client choose one of N (N >= 2) methods of error handing, like mod_speling. It doesn't appear to cover any specific method of error handling.

  • 404'd! (Score:5, Funny)

    by boristdog (133725) on Wednesday January 30 2008, @08:53AM (#22233630)
    404'd! [homestarrunner.com]
    • Old news (Score:5, Insightful)

      by yog (19073) * on Wednesday January 30 2008, @10:17AM (#22234396) Homepage Journal
      Bezos has stated in the past that he is patenting software methods as a defensive measure. "We're not saying we have bad patents," Amazon.com spokesman Bill Curry said. "We feel very good about our patents... [Bezos] makes the point very emphatically in the letter that we cannot unilaterally disarm in a world where there are big ugly players who aren't disarming." [wired.com]

      It's like road rage. When people are cutting you off and breaking all the rules, you have to tailgate and cut them off as a defensive measure (sometimes, at least). Nice guys finish last. The entire system is broken and the Patent Office really needs more legislative direction because it has strayed from its original mission.

      I think software and business methods should not be patentable in the same way that physical inventions are. Also, I question the concept of selling patents. We end up with these litigious patent holding companies that have no technical abilities of their own, only a lot of lawyers.

      A few years ago I looked into making and marketing a telephony device that would be an incremental but useful improvement over existing equipment, and discovered that so many methods related to telephony and voicemail are patented that practically speaking there was no way to make a device without infringing. "A method for playing back a telephoned message by pressing a button"--give me a freaking break. No wonder the U.S. has slipped behind in technical innovation, when much of the incentive for incremental product improvement has been removed by the threat of instant litigation. Thank goodness the Asians still believe in incremental improvement.

      I'm OK with Amazon patenting stupid obvious things, as long as they don't enforce those patents, which I believe they have done very little of, and as long as Jeff Bezos continues to crusade for patent reform. Just my 2c!
      • Re: (Score:3, Insightful)

        That's all well and good until someone steps in and replaces Bezos. What happens when we get someone who sees patents as a profit center and not just a way to prevent other wackos from patenting it? What happens when they decide to sell those patents to some patent warehouse owned by lawyers? Needless to say, it could get ugly fast. The fact is, these patents should never have been granted in the first place. They're nothing but harmful to American business.
      • Re:Old news (Score:4, Insightful)

        by seebs (15766) on Wednesday January 30 2008, @12:18PM (#22235834) Homepage
        Bezos has said that, yes.

        Would you call his one-click lawsuit from a few years back a "defensive measure"? I would not.

        The fact is, he's lied on every last thing he's said on the issue, and swallowing these lies given his clear, repeated, public contradictions of every such claim is just pathetic. I mean, it's long past "stupid".

        You believe they have done "very little" of enforcing patents on "stupid obvious things". How much is okay? Would you say that demanding an injunction against a competitor running their existing web site, at all, during the holiday season is "very little"? Reasonable for you? You think it's no big deal to demand that someone suddenly, on no notice, stop accepting any orders on their web site until they revise their system not to conflict with a "stupid obvious" patent?

        The fact is, Bezos is part of the problem, and actively so, and all his "crusading" for patent reform has consisted of, purely reactively in response to negative outcries over his abuses, saying sets of things that his critics would like him to say... And then doing nothing about it, and continuing to use the system, as is, to his advantage. Including filing suits.

        You know why so many 419 scams have phrases like "in God's name" and "we are devout Christians"? Because there are millions of people who will reflexively assume that anyone who claims to be Christian is honest and trustworthy, as long as they use a few of the right buzzwords. Bezos has found the corresponding hole in your cognitive system; you simply can't be bothered to investigate the truth of his claims. Why? Because, if they were true, they would be exactly what you wish he'd think.

        It ain't so. Amazon is an abuser of the patent system. Amazon is a spammer. Amazon is everything we hoped they wouldn't be, and they rely on our wishful thinking to convince us that, really, they're a great company, when they are actually systematic scumbags. They spam, and then they get caught and "fix" it. They abuse the patent system, and they don't even stop abusing it, they just say it's "defensive". They have filed suits against competitors who were not using patents against them since they first claimed this was defensive.

        Why do you keep trusting them? What's your emotional investment in never, ever, considering the possibility that they lied to you?
  • cat /etc/httpd/conf/httpd.conf |grep -B 5 404

    # Customizable error responses come in three flavors:
    # 1) plain text 2) local redirects 3) external redirects
    #
    # Some examples:
    #ErrorDocument 500 "The server made a boo boo."
    #ErrorDocument 404 /missing.html
    #ErrorDocument 404 "/cgi-bin/missing_handler.pl"

    Customizable error responses come in three flavors:
    # 1) plain text 2) local redirects 3) external redirects
    • Not prior art (Score:5, Informative)

      by codegen (103601) on Wednesday January 30 2008, @09:01AM (#22233690) Journal
      The patent is not about the server serving custom
      error pages (which is your post), but about a client
      side process that communicates with a separate error server
      to generate the appropriate response. So I would guess it
      is a intended to be a plugin for a browser.

      But then this is slashdot, why bother to read the article.
    • Reminds me of Marvin:

      The requested patent is totally fake.
      No patent here.
      Even tried multi.
      Nothing helped.
      I'm really depressed about this.
      You see, I'm just a web server...
      -- here I am, brain the size of the universe,
      trying to serve you a simple patent,
      and then it doesn't even exist!
      Where does that leave me?!
  • by bleh-of-the-huns (17740) on Wednesday January 30 2008, @08:58AM (#22233668)
    I mean seriously, in the last 2 weeks, we have the Minerva Industries patent on smartphones, and now this.. Who the hell is working in the patent office.....

    I am sure we can find some prior art.. the most annoying being angelfire and geocities from way back when.
    • by rucs_hack (784150) on Wednesday January 30 2008, @09:09AM (#22233770)
      Its not all the patent offices fault. They have to follow the rules, and those rules were not set up for the torrent of patents they receive these days.

      If you get too many patent applications, the process of establishing if prior art exists also gets swamped. Thus without a special effort, patents which have prior art can still get granted.

      I've skimmed the patent in question, and it sounds like a new thing to me. There may be bits and pieces that invalidate some of what it does, but since the USPTO allows patents for software products (which has always struck me as dumb), this is probably valid.
      • by Anonymous Coward on Wednesday January 30 2008, @09:43AM (#22234110)

        Its not all the patent offices fault. They have to follow the rules, and those rules were not set up for the torrent of patents they receive these days.


        They should use Comcast, then. That should slow the torrents down a bit...

        *cower (Rank 8)*
      • by Hatta (162192) on Wednesday January 30 2008, @10:14AM (#22234368) Journal

        If you get too many patent applications, the process of establishing if prior art exists also gets swamped. Thus without a special effort, patents which have prior art can still get granted.


        Um, easy solution, don't issue any patents until you're sure there's no prior art. If there are too many patents submitted, tough shit, no patents for anyone. That would motivate reform!
          • Re: (Score:3, Interesting)

            Sure enough that you're willing to risk some sort of liability if you issue the patent and prior art is later discovered.
    • Who the hell is working in the patent office.....

      I always wondered whether the "According to Jim" writers were working day jobs during the strike. Now I know.

  • RTFA you tards (Score:4, Insightful)

    by Anonymous Coward on Wednesday January 30 2008, @09:03AM (#22233720)
    It's not a patent on 404 handlers. It's a patent on a client-side component that detects errors including, but not limited to 404s, then relays the error to an external server and receives an alternate URL or resource to serve the user. Blame the patent office for being idiots if you want, but this time *you* are the idiot.
  • It's *CLIENT* based (Score:5, Informative)

    by ps236 (965675) on Wednesday January 30 2008, @09:05AM (#22233736)
    If anyone else has read the patent, they'll realise that it's a CLIENT-SIDE component that's involved.

    Most (all?) 'automatic redirect' systems I've seen are server based - the server runs a script which says 'That page couldn't be found, did you mean any of these...'

    I can't imagine who'd put this on the client with client-server communication going on. It sounds like a vastly over-engineered and 'Enterprisey' solution to me. It DOES have the advantage that it can look back in the browser history, but I'm not sure I see how that could benefit the user (the component COULD tell the server what's in the history though, so it could benefit Amazon!)

    • MSIE did it. (Score:5, Insightful)

      by Rob T Firefly (844560) on Wednesday January 30 2008, @09:09AM (#22233778) Homepage Journal
      Much as I'm loath to hold it up as an example of anything ever, Internet Explorer has been using client-based action to generate a friendly "This page cannot be displayed" page in place of 404 errors for years and years.
      • Yeah, but to the best of my knowledge MSIE doesn't customize the content of that page any way, or communicate with an external server before rendering it - it just prints the pretty message.

        Or maybe it does, but the customization is like, subtle, man.

        (Warning: I haven't read TFA either. :) )
        • Patents are supposed to be for things that are innovative and not obvious. It is not really a big stretch between having your browser display a 404 error and having it also go somewhere else. I'm thinking the reason no one did this previously is because of the ruckus raised when the top level domain gatekeepers like Verisign started selling redirects when you mistyped a domain name. Amazon's idea really is just a minor twist this theme if you ask me. And even if you didn't ask me.
    • It DOES have the advantage that it can look back in the browser history, but I'm not sure I see how that could benefit the user

      Especially if the user in question has some things in their history they might not want popping up in front of others.
    • Looking at the age of the supporting documents, this patent application dates back to the days when client side apps were thought to have a role in the browser. I'm thinking of things like Realnames, and various push client mechanisms that had client side authentication. Then again, someone at Amazon must see a value in it if they have persisted in getting the patent.
  • Patents (Score:3, Funny)

    by Rik Sweeney (471717) on Wednesday January 30 2008, @09:11AM (#22233798) Homepage
    The problem is though that who's really at fault at the end of the day, Amazon for trying to apply for the Patent or the Patent office for allowing it?

    It's like the loophole I found when I was in Game a few months ago:

    "£9.99 please"
    "But it says it's half price"
    "No, it's only half price when bought with something else"
    "Oh... so if I buy two of these then I get them both for less than the price of just one of them?"
    (Realising the problem) "...yeah... yeah you would actually"
    "OK, I'll be back in a minute"
  • And yes. I know it's client side bla bla, but still :) If you just redirect all your 404 pages to Amazon, at say www.amazon.com/thispagedoesnotexistyoubastards I wonder how that'd work out :)

    Considering almost everyone uses the 'back' option anyway after hitting a 404...
  • by GauteL (29207) on Wednesday January 30 2008, @09:27AM (#22233956) Homepage
    ... which is using a well known malware/phishing technique.

    It is a client component (read adware/malware) that intercepts 404 messages, calls home to find out where to redirect the user, then redirects.

    i.e. if you type in "slahdsot.org" it will search a database of misspellings and redirect you to "slashdot.org".

    or.. in the case of malware.. if you type in "myinternetbank.com" it could redirect you to "myphishingsite.com".

    I'd be surprised if there isn't prior art among the less ethical Internet inventions out there.
    • I'd be surprised if there isn't prior art among the less ethical Internet inventions out there.

      It took me just a few google searches to find a worm that did exactly that; it intercepts and redirected google searches to a results page with different ads.
      I'm sure it wouldn't be hard to find malware that intercepts 404's.
  • by SplatMan_DK (1035528) * on Wednesday January 30 2008, @09:54AM (#22234190) Homepage Journal
    An Australian man has been issued with an innovation patent for the wheel after setting out to test the workability of a new national patent system.

    John Keogh was issued the innovation patent for a "circular transportation facilitation device" under a patent system introduced in May 2001.

    (read the rest...) [newscientist.com]

    :-)

    - Jesper
  • Error 404: Please pay Amazon 2 cents to see what caused the error.
    • by nagora (177841) on Wednesday January 30 2008, @09:12AM (#22233806)
      To me, this is like placing a patent on "using the emergency flashing lights on my car to signal for help." Someone already thought of this use of my emergency flashing lights, and that's why the lights were implemented.

      Which is exactly the situation with the 1-click patent. Cookies were invented to allow a site to recognise a returning user/customer, so patenting the act of using cookies to recognise a returning customer (and by "recognising" I mean linking them to an account) should never have been allowed a patent.

      There is a VERY simple solution: don't buy anything off Amazon, and tell your friends not to too. I don't. If they want my money they can stop trying to prevent me from working.

      TWW

    • by superid (46543) on Wednesday January 30 2008, @09:26AM (#22233944) Homepage
      IANAPA, but coincidentally in another window right now I am writing a patent disclosure application. I am lucky enough to work at an office that has a cadre of patent attorneys and just last week I had a discussion about a topic just like this Amazon patent (or I think it is).

      In my case, I've "built" a system using nothing more than a set of other peoples building blocks. Each existing component is already extensively patented and/or in common use with TONS of prior art. The system I created performs a useful function that is also covered by prior art (in my case, it is a medical diagnostic tool).

      The patent attorney told me that my idea was very likely patentable because it was a "novel" (new) implementation, even though the pieces exist and a (different) end product already exists.