Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
×
Patents The Courts United States

US Supreme Court: Patent Holders Must Prove Infringment 143

jfruh writes "The Supreme Court issued a ruling that might help marginally curb patent madness. Ruling on a case between Medtronic and Mirowski Family Ventures, the court rules that the burden of proof in patent infringement cases is always on the patent holder. This is true even in the specific case at hand, in which Medtronic sought a declaratory judgement that it was not violating the Mirowski patents."
This discussion has been archived. No new comments can be posted.

US Supreme Court: Patent Holders Must Prove Infringment

Comments Filter:
  • Now the next step... (Score:5, Interesting)

    by Moryath ( 553296 ) on Thursday January 23, 2014 @11:52AM (#46047161)

    placing the burden on the patent holder to prove the patent is NOT the result of:

    - Patent slamming to game the system (e.g. submitting the same fucking thing 100 different times hoping one submission will slip by an overworked patent reviewer)
    - Patenting something already patented
    - Patenting something that is already obvious
    - Submarine-patenting

    • by manquer ( 1950350 ) on Thursday January 23, 2014 @12:17PM (#46047475) Homepage

      (e.g. submitting the same fucking thing 100 different times hoping one submission will slip by an overworked patent reviewer)

      I am not a fan of the current patenting system, but this is BS, a patent application costs $10,000, if the patent reviewer is overworked it has nothing to do with the abuse of the system, even considering a cost of $200,000 to the USPTO per patent reviewer including all the overheads a reviewer has to only review 20 patents a year to make the system viable.

      • by Anonymous Coward on Thursday January 23, 2014 @12:31PM (#46047657)

        I am not a fan of the current patenting system, but this is BS, a patent application costs $10,000

        LOL...if someone is charging you $10K to file it for you, then you are getting ripped off.

        http://www.uspto.gov/web/offices/ac/qs/ope/fee010114.htm

        • by multimediavt ( 965608 ) on Thursday January 23, 2014 @01:56PM (#46048427)

          I am not a fan of the current patenting system, but this is BS, a patent application costs $10,000

          LOL...if someone is charging you $10K to file it for you, then you are getting ripped off.

          http://www.uspto.gov/web/offices/ac/qs/ope/fee010114.htm

          That $10k price is actually not bad, as the due diligence search has to be done first by the patent attorney. Typically, I've seen patents average closer $20k to file. [richardspatentlaw.com] They may also be searching and filing in other countries as well, CYA. So, if he's getting a 10k price he's doing ok. I would not try to self apply if you want the patent to stick. You need a good patent attorney.

          • by Moryath ( 553296 )

            But what ACTUALLY happens is that the big companies have their own attorney on staff, whose job it is to robosign "due diligence searches" without bothering to do them.

            • But what ACTUALLY happens is that the big companies have their own attorney on staff, whose job it is to robosign "due diligence searches" without bothering to do them.

              I don't think so, but if you could point out an example I'd be more inclined to see your point. Most Fortune 100 corporations don't have patent attorneys on staff as employees. They may retain a firm (like they do for accounting audits), but that's not on staff. There are a few exceptions that I do know of, IBM, Intel, HP and Texas Instruments I know for a fact have patent attorneys in house but they are also R&D machines! I assure you they do the due diligence, most times internationally. Because of th

        • by sjbe ( 173966 ) on Thursday January 23, 2014 @02:45PM (#46049139)

          LOL...if someone is charging you $10K to file it for you, then you are getting ripped off.

          The cost isn't the patent submission fees. The primary cost is the lawyer you'll inevitably have to hire during the process. Getting a patent successfully through is actually more complicated than it might seem at first glance.

          • That sounds like the cost for those who don't have full time lawyers on payroll.

          • More importantly, if you have your own lawyers on your payroll, the cost is negligible. Especially when you consider that those lawyers seemingly only spend 5 minutes on Google doing 'due diligence'. If they spent much longer, then far less obvious/pre-existing patent applications would be made. That, of course, doesn't excuse the patent examiner for rubber stamping it after seemingly spending another 5 minutes Googling the claims.

          • Doesn't sound like every patent [newscientist.com] will cost $10K.
      • by sjames ( 1099 )

        There is no $200,000 involved. They just keep resubmitting revised applications with trivial changes over and over on the same $10K fee.

      • It's not BS.

        The USPTO lowered its standards [at the behest of Congress] to lower its standards to reduce its backlog. If an application is denied, it can be refiled [many times]. The only way to truly clear it is to approve it [and toss it into the court system]:
        http://www.techdirt.com/blog/i... [techdirt.com]

        $10,000 times 20 is a trivial amount [on a corporate level] compared to an NRE budget for a legit R&D outfit. There seems to be plenty of [unscrupulous] VC money to back such refilings to get something that ca

    • by SQLGuru ( 980662 )

      I'd also like software patents to have a working implementation which is what's actually covered by the patent.....much like physical patents. If I patent "sorting" (and yes, many software patents are that vague), it doesn't allow for bubble sort, selection sort, insert sort, quicksort, etc.

      • by Sarten-X ( 1102295 ) on Thursday January 23, 2014 @01:10PM (#46048055) Homepage

        Physical patents do not require a working implementation, and haven't in quite some time. Rather, they only require enough detail that someone skilled in the art could create a working implementation of what's claimed using only common knowledge and the patent itself.

        For software, that means that if you're claiming a particular sorting algorithm, you have to include enough details about how that algorithm works that another programmer could duplicate the sort... but if you're not patenting the comparison operation, you don't have to include that, even if your algorithm requires some particular comparison to work. That's a trade secret.

        Also note that the title and abstract really don't matter. Most of the "vague" patents reported on Slashdot and other anti-patent sites have vague titles, but the claimse themselves are usually pretty detailed about the specific situation where the method applies.

        Continuing the analogy, a method for sorting 3D models might have a title of "Comparative sorting method", and an abstract that describes the need for sorting, both of which are sure to annoy Slashdotters. The patent claims could then detail a particular way of framing the problem, where we sort models based on their ratio.of filled-to-empty space, with a weighted preference for the appearance of space from particular projections. The algorithm itself might appear at first glance to be, say, a quick-sort, but with a few little tweaks to accommodate the multiple sorting criteria. As above, the actual method for computing filled space may be a secret that's left out of the patent, even though the patented mechanism is really useless without it. The end result is a patent that looks like "quick sort... now in 3-D!" but really describes a useful, novel, and non-obvious solution for a very specific problem.

        • ... because its not obvious to take an existing algorithm and do 'a few little tweaks' to make it suit your application?

          • Not necessarily, no.

            Ideally, it depends on how much effort or insight is required for those tweaks, but there's no quantitative metric by which to judge insight or effort.

            • Not even obvious to someone skilled in the art of software development or mathematics?
              Someone with a computer science degree?

              Prior art is not limited just to the references being applied, but includes the understanding of one of ordinary skill in the art....

              http://www.uspto.gov/web/offic... [uspto.gov]

        • by SQLGuru ( 980662 )

          And yet, if I implement the same type of feature (with no knowledge of the magic sauce), I'm in violation of the patent.....even though it was obvious enough for me to come up with the same solution. Either the patent should include everything you need in order to implement the solution (so that I know when I'm violating it), or it shouldn't be a patent.

          • Yes, you would be in violation of the patent if you duplicate everything in the claims, but only if you duplicate what's in the claims. If you're using their magic sauce in some other context, such as a 3-D model catalog rather than a sorting algorithm, you're in the clear. If you can make their algorithm work without one particular claimed step (or doing that step differently), you're still fine. The patent itself does contain everything you have a risk of violating.

            That someone's trade secrets are obviou

        • Physical patents tend to include a great many details about how the systems described work. Even without illustrations, they're sufficiently documented that they actually CAN be constructed from the claims and any other supplemental materials.

          Contrast to most software patents which list some arbitrary "using undisclosed comparison algorithm, verb the noun". The actual verb and noun are irrelevant because there is no way to get past the comparison algorithm. As an example, the patent you describe is compl
          • Let us hope the lawyers read patents more thoroughly than you read my post.

            The innovative part of the described patent isn't the comparison. It's the modifications to quick-sort that let it work accurately on the multiple criteria the comparison operation is providing. A quick-sort for 3-D models that uses a single criterion would be outside the patent claims, regardless of the comparison used. Similarly, a merge sort for 3-D using multiple criteria wouldn't fit the patent claims, so it'd be safe, as well.

        • but really describes a useful, novel, and non-obvious solution for a very specific problem.

          Although this is rare in my experience of reading software patents. I've never read a claim that revealed anything novel or non-obvious.

          I do know a piano designer who goes through old patents dating back into the 1800s looking for ideas. So the system has been good, just right now, with software patents, stupid things get patented.

          • Obvious in hindsight is not what's being considered.

            • Obvious in hindsight is not what's being considered.

              Just to be clear, are you one of those people who thinks Amazon's one-click-patent is non-obvious?

              • The separation of delivery information from the ordering process, for the purpose of immediate order placement, was not obvious in 1997 when the patent was filed. In the 16 years since then, it has become obvious, as web-based workflows have matured.
                • The separation of delivery information from the ordering process, for the purpose of immediate order placement, was not obvious in 1997 when the patent was filed

                  Seriously? You believe that?

                  • Obviously, you don't. Please show me any documented example of such a system prior to 1997.

                    • Yes, it's actually very obvious. Anyone trying to solve the same problem would come up with a similar solution.

                      Or you think you couldn't have come up with that solution?
                    • That's not a citation [wikipedia.org].
                    • Indeed, but you've avoided the question.

                      From which I can only conclude that you would be capable of finding that solution, but are too proud to give up your position.
                    • No, because the client's not initiating the order.
                    • You are claiming the patent is obvious. The burden of proof [wikipedia.org] is on you to prove that it's obvious. Surely there must be some example of a similar system, if it was so obvious in 1997?

                      Instead of actually proving your claim, you're asking me to prove that something was not possible [wikipedia.org], to which the reexaminations of the patent in question serves as evidence of the adequate non-obvious nature of the surviving portions of the patent.

                      Please review your rules of debate and try again.

                    • Please review your rules of debate and try again.

                      This isn't a debate, the answer is obvious. This is a psychological analysis, not a debate. I am trying to understand what kind of person you are that you would hold such an unusual viewpoint.

                      Do you also consider the playing-with-a-cat-with-a-laser-pointer patent to be non-obvious?

                    • You could have just asked, you know. I am a person who values logic and application of a consistent rule set, well-educated in the fields of law and computer science.

                      Of course, now you're claiming that such a viewpoint is unusual. You'll have to prove that, too.

                    • So, do you consider the playing-with-a-cat-with-a-laser-pointer patent to be non-obvious?
                    • So, do you usually avoid backing up your claims?
                    • ok, so we'll assume: you do think it is obvious. So clearly, you agree that some patents can be obvious, and still get through the patent office.

                      Now let's see this: do you understand that there is a difference between being obvious, and having prior art? That is, do you understand that an invention might have no prior art, but still be obvious?
                    • Begging one question to avoid another? That's not right.
    • by iamgnat ( 1015755 ) on Thursday January 23, 2014 @01:08PM (#46048025)

      I'll be the contrarian here and state the belief that this ruling isn't so good.

      The major issue, of course, is that there is massive abuse to the system, but if you look at what the system is supposed to do I think this ruling turns things more decidedly in the favor of large companies.

      The idea of the patent system was that anyone could patent their grand idea and then have legal backing to protect it in court from someone that uses the idea without consent. The filing fees were also designed to be low to keep the barrier of entry low enough that "the little guy" could get the same protection as the big corporations.

      Prior to this ruling (ignoring the shake downs by trolls) an individual or small company had a chance of winning a patent case against much larger entities (motions and legal wrangling aside) as the process of discovery forces the defendant to show their cards and prove they aren't infringing with no upfront cost to the plaintiff.

      With this ruling, if you come up with the next great search algorithm (software patent absurdity aside) and Bing/Google/Yahoo steals it you now have to foot the bill for the discovery. Without the court order you also aren't going to get very far in that process as they aren't exactly going to welcome you into their office, sit you down at a console, and give you access to their code.

      So what this ruling does, in my opinion, is give the larger companies the right to violate patents from smaller entities with near impunity. It also (as someone suggested further down regarding OSS projects) gives rise to a whole new possible "reverse-patent" trolling business scheme.

      Basically this ruling, I think, has made things worse.

      • by viperidaenz ( 2515578 ) on Thursday January 23, 2014 @01:38PM (#46048299)

        Without the court order you also aren't going to get very far in that process as they aren't exactly going to welcome you into their office, sit you down at a console, and give you access to their code.

        That's only really valid for software patents. Since you could just buy one of their widgets and pull it apart.

        If it makes software patents less appealing to file and more costly to pursue and harder to troll, good. They shouldn't be patents in the first place

      • Patent trolls had gotten to the point of just threatening people and winning because only the biggest players could defend themselves. Couple this with the fact the PTO giving out dubious patents, and the low cost of suing someone how can things get any worse?

      • I disagree, based on presumed innocence. If you accuse me of infringing on your patent, you should be required to provide proof, as would be the case in any other legal proceeding. If I'm charged with a crime, the burden is on the prosecution to prove my guilt. Consider also a civil case. if you accuse me of being negligent in keeping my floor clean, causing you to slip and fall, the burden is on the plaintiff to prove that the floor really was a mess and I was negligent in not cleaning it.

      • Agreed. The ruling certainly doesn't help the small, private inventor, but does benefit big corps. Not surprising from the court that made corporations people and money speech.
      • So in other words working exactly as intended?

        I am sorry, did you wake up in a different country or something??

      • by pavon ( 30274 ) on Thursday January 23, 2014 @02:27PM (#46048865)

        If you sue someone for patent infringement you have always had the burden of proof, even before this ruling. All this ruling is saying is that if you threaten to sue someone, and they go to a judge first asking you to put-up or shut-up, the burden is still on you as the patent holder, same as if you had sued them.

        Secondly, this ruling does nothing to limit the discovery process. As a small inventor suing a big company you still have the same subpoena powers during discovery as you did before.

        In other words, the Supreme Court simply reaffirmed that accused infringers are innocent until proven guilty, regardless of the procedural nuances of how the lawsuit is initiated. None of the concerns you voiced will become worse due to this ruling.

        • infringers are innocent until proven guilty

          They're "innocent" except they still have to reveal their trade secrets to competitors. Which will _never_ be used outside the discovery process. Wink, wink, nudge, nudge.

      • I think you would be correct in a perfect world where patents were actually working as intended however reality is far different.

      • The idea of the patent system was that anyone could patent their grand idea and then have legal backing to protect it in court from someone that uses the idea without consent. The filing fees were also designed to be low to keep the barrier of entry low enough that "the little guy" could get the same protection as the big corporations.

        This is completely false. Patents were never about the "little guy". Their purpose is to benefit society by providing an advantage to disclosing the secrets of invention so society can learn. Prior to patents, technology was often a closely guarded secret, belonging to individuals or trade guilds, secrets that were often lost with the deaths of the people involved. By making disclosure a more attractive option than secrecy, society could benefit by learning from the details of the inventions.

        That is the ide

      • by DM9290 ( 797337 )

        Prior to this ruling (ignoring the shake downs by trolls) an individual or small company had a chance of winning a patent case against much larger entities (motions and legal wrangling aside) as the process of discovery forces the defendant to show their cards and prove they aren't infringing with no upfront cost to the plaintiff.

        With this ruling, if you come up with the next great search algorithm (software patent absurdity aside) and Bing/Google/Yahoo steals it you now have to foot the bill for the discovery. Without the court order you also aren't going to get very far in that process as they aren't exactly going to welcome you into their office, sit you down at a console, and give you access to their code.

        If a company files a motion against you to for a declaratory judgement that it is not violating your patents, that motion would only be able to cover the material that it disclosed to the court. A judgement could never cover anything that they refused to disclose to the court.

        You can't get a court to rule that something it has no knowledge of was legal. It has no jurisdiction to make such a ruling.

        The burden of proving they infringe may rest on you, but only in terms of the subject material they are trying

  • by TWX ( 665546 )
    ...I've become increasingly used to insanity when it comes to the court and advancing technology. Next you'll tell me that the court will find that it's unconstitutional to collect certain kinds of data without a warrant...
    • Re: (Score:2, Interesting)

      by Anonymous Coward

      The summary omits the central issues in the case. Medtronic had previously licensed patents from Mirowski Family Ventures for some of their older devices. Medtronics claimed their new devices didn't use the same methods and therefore decided not to continue to license the patents. The Federal Circuit Court (which is notoriously patent-holder friendly) decided that since Medtronics had previously been a licensee and they were the plaintiffs in the declaratory judgement suit (they had sued pre-emptively to

  • Costly discovery? (Score:5, Insightful)

    by gstoddart ( 321705 ) on Thursday January 23, 2014 @11:54AM (#46047201) Homepage

    I can see this leading to some pretty costly discovery for companies being sued.

    Because it's going to amount to "in order for us to prove you violated our patent, we need you to hand over all of your information so we can find the proof".

    I hope there is a provision for saying "OK, but we're going to charge you $100 million for our time in getting this" -- because otherwise this just allows the patent trolls to cause the people they accuse to incur massive costs which might make settling cheaper.

    You shouldn't be able to make someone bear the cost of you suing them based on something you can't prove without them doing the work for you.

    This reminds me of the SCO lawsuit, where the most they ever found was, what, 7 lines of infringing code which SCO themselves had nicked from AT&T UNIX?

    • by Anonymous Coward on Thursday January 23, 2014 @12:05PM (#46047357)

      This is EXACTLY the problem right now: filing an infringement claim is trivial, it only requires the plaintiff to assert infringement (not detail it). That currently triggers discovery which is costly for the defendant. Defendants thus settle for less than the cost of discovery. That is the essence of legal blackmail/patent trolling.

      SUPPORT PATENT REFORM IN THE SENATE.

      The markup happens in March. There is a hearing today as well.

      Stay early discovery or require costs to be borne by the plaintiff.
      Permit fee shifting when plaintiff's lose.
      Identify the real party in interest and REQUIRE a bond - shell companies currently do most of the trolling and are judgment proof (have no assets).
      PROTECT END USERS - using an API can get you a patent infringement lawsuit (if the API is alleged to infringe), offering wireless in your cafe gets you sued (because Cisco allegedly infringed on someones patent and you are using their device).

    • All the more reason that we should have loser pays for all civil lawsuits in the US, as they do in the UK. If you sue someone, and lose, you're on the hook for their legal fees.

    • Re:Costly discovery? (Score:4, Informative)

      by Karzz1 ( 306015 ) on Thursday January 23, 2014 @12:49PM (#46047841) Homepage
      This reminds me of the SCO lawsuit, where the most they ever found was, what, 7 lines of infringing code which SCO themselves had nicked from AT&T UNIX?

      A little off-topic I know, but IBM was never found to have infringed on any code from SCO. SCO tried to *claim* some code that was already licensed under a FLOSS license was the same as their code and thus infringed on their IP. They made a big deal of this to the press forcing them to sign NDA's & "showing them the code". [theage.com.au]
      • Sorry, should have put infringing in air quotes to make that more clear that it was never really infringing in the first place.

  • Yawn.... (Score:5, Insightful)

    by American Patent Guy ( 653432 ) on Thursday January 23, 2014 @11:55AM (#46047219) Homepage

    A plaintiff always bears the burden of proof in showing that he is entitled to the relief requested from the court. For patent infringement, that means showing a patent has been infringed.

    The only reason this is in the news is because the appellate court (the CAFC) screwed it up one time, and the Supreme Court had to make a return to sanity.

    • Re:Yawn.... (Score:4, Insightful)

      by tomhath ( 637240 ) on Thursday January 23, 2014 @12:04PM (#46047341)
      RTFA. The Supreme Court ruled that the defendant has the burden of proof in this case
      • That's just because this particular case uses negative logic. Try substituting "libel/slander" for "patent infringer":

        Normal case:
        Plaintiff: "He slandered me!"
        Defendant: "No I didn't!"

        This case:
        Plaintiff: "I didn't slander him!"
        Defendant: "Yes you did!"

        In either case, it seems overwhelmingly logical and obvious that the burden of proof is on the person claiming they were slandered, irrespective of whether they're the plaintiff or defendant.
      • Re:Yawn.... (Score:5, Informative)

        by Uninvited Guest ( 237316 ) on Thursday January 23, 2014 @12:22PM (#46047537)
        Correct. The long version: The plaintiff in a patent case is usually the patent-holder, who is seeking damages for infringement. In those cases, the patent-holder-plaintiff already had the burden of proof. In this case, the (potentially infringing) plaintiff is seeking protection from patent infringement lawsuits by suing the patent holder, requiring the (patent holder) defendant to prove that the patents are valid and/or that the plaintiff infringes the patents. Normally (and the appeals court found), the plaintiff would have the burden of proof. According to the appellate ruling, the plaintiff (potential infringer, seeking protection) would have to prove that they were not infringing, or prove that the patents were invalid. The Supreme Court reaffirmed the lower court ruling: The patent holder, whether plaintiff or defendant, must prove that the other party (plaintiff or defendant) infringed the patents, and that the patents are valid.
      • by alen ( 225700 )

        yes, they said mirowski has to prove their patents were violated

      • No, they ruled the patent owner always has the burden of proof. It so happens, that's the defendant. It is the "always" that was the important part of the ruling.

        Until fairly recently, if you licensed a patent, you gave up the right to contest it. That changed, but there was still some grey.

        In this case, the Plaintiff had licensed for one product but wanted a judgement concerning a newer product. The Defendant asserted the burden of proof was on the Plaintiff since they had licensed it and now wanted to

    • A plaintiff always bears the burden of proof

      If that were true, would SCOTUS have been ruling on it?

      That they were ruling on this to me suggests it may not have been quite so clear.

      Because otherwise it wouldn't need to be decided by SCOTUS.

      • by thaylin ( 555395 )
        Because no one ever tries to challenge facts, ever.... Also the court did not say it was perfectly clear previously, just that this is how it is, so deal with it.
    • It's hard to imagine the CAFC screwing up anything just one time.

    • Yep, I don't get this article at all. This is not a sign of reform or relief. It is business as usually in patent cases.
  • This ruling on its own might only help slightly. But aside from that, I think this kind of 9-0 ruling (!) coming from the U.S. Supreme Court is encouraging for other patent-related cases that the Court will soon be hearing.
  • DOS Patent Trolls? (Score:5, Interesting)

    by Luciano Moretti ( 2887109 ) on Thursday January 23, 2014 @11:56AM (#46047229)

    How hard would it be for OpenSource Projects and Small Companies to file for a declarative judgement lawsuit for every software patent held by trolls? I'm assuming that filing fees would get cost prohibitive quickly, but would we be able to DOS attack the patent trolls and the courts they use to prove the point?

    • Fairly hard. One of the requirements for asking for a declaratory judgement is that you have to either have been sued or have a reasonable fear of being sued by the patent-holder. If the patent-holder hasn't actually sued anyone yet, it's easy for them to v get DJ actions dismissed. Even if they're sued others but not sued you, traditionally it's been easy for them to argue that you don't have any reason to believe you'll be sued yet (not that that is not a promise not to sue you in the future). In this par

      • by Grond ( 15515 )

        One of the requirements for asking for a declaratory judgement is that you have to either have been sued or have a reasonable fear of being sued by the patent-holder.

        That's not the standard for declaratory judgment jurisdiction in patent cases and hasn't been since the 2007 MedImmune case, in which the Supreme Court rejected the Federal Circuit's "reasonable apprehension of suit" test. The MedImmune standard is "whether the facts alleged, under all the circumstances, show that there is a substantial controversy, between parties having adverse legal interests, of sufficient immediacy and reality to warrant the issuance of a declaratory judgment." MedImmune v. Genentech

  • Haha wow (Score:2, Interesting)

    by Anonymous Coward

    Cue a million companies queuing up to sue for declaratory judgement in order to force companies to show their cards instead of vague "something you do violates some part of one of our patents pay 20% of your income or be sued" threats.

  • In my opinion the next step after that, should be to abolish software patents.

    • In my opinion the next step after that, should be to abolish software patents.

      Legal says "No - and would you like a job in the mail room better?".

    • All patents must be abolished. They only encourage rent-seeking behavior.
  • I get it for patent trolls but if it's a legit ripoff of a patent of mine now I have to play P.I. and break into their office and interview people and steal papers and hide recording devices? Because otherwise good luck finding proof from a tightly locked down company giving you the runaround. They won't hand over so much as a blueprint without a court order.
    • by thaylin ( 555395 )
      Well you could just not claim that they are infringing unless you know for a fact they are infringing, that is why this verdict came down. If you dont make the claim they cannot file for a judgement.
  • This is almost as bad as listening to lawyers talk about software.

A committee takes root and grows, it flowers, wilts and dies, scattering the seed from which other committees will bloom. -- Parkinson

Working...