3 ms·

Not sure what the EFF is trying to get at here. After the decision, it says rejections rose 31%, but now they have fallen back down. The EFF contends this is

Not sure what the EFF is trying to get at here. After the decision, it says rejections rose 31%, but now they have fallen back down. The EFF contends this is because the Patent Office is ignoring the law, but they don't offer any evidence to back up this assertion. Another likely explanation is that patent applicants are adjusting their applications to be in line with the decision, and thus the rates have gone back down. This assertion by the EFF would only make sense if patent applicants didn't know about the rule change. But applicants do know, and have adjusted their applications accordingly in the intervening time.

pnw_hazor 6y ago
"Another likely explanation is that patent applicants are adjusting their applications to be in line with the decision" This. We had/have no problem getting past Alice if the application was drafted post-Alice. (Except for one art unit that was being unreasonable -- but even they seemed to have given up. Maybe their boss retired?) Nothing changed except we included additional language in the spec and claims for no other reason than to get around Alice rejections.
- hcknwscommenter 6y ago
  Really? YOu have some magical language in the spec and claims that universally gets around Alice but (and I'm inferring here) doesn't narrow the scope of your claims? Please please please do tell me what this magical "additional language" is. I am not trying to troll. I am genuinely curious. I have upvoted your post. You can send me a private message if you like.
  - pnw_hazor 6y ago
    I call it magic language because it had nothing to with the novelty of the inventions. Note, Alice wasn't the first time this was required. We always adjust the language we use in applications as new case law comes down. Further, gibberish included in claims to defend against Alice rejections usually had nothing do with novelty because Alice was about subject matter patentability not novelty. Before the early Alice guidance came out we tried a lot of different things, much of it superficial in my opinion. Including: emphasizing that the machine did the work; showing that inputs are being transformed into something new by the machine; writing claims that show different parts of the system interacting with each other; and sometimes nonsense like "wherein each foobar is stored in a hardware memory" or "displaying the bizbaz on a hardware display." In some hard cases (mainly apps drafted before there was any Alice guidance or useful case law) we included claim limitations like "modifying performance rates based one or more sensor inputs" (e.g., CPU temperature) Later, when the first case in district court found that a patent satisfied Alice, we started including language verbatim from that case in specs, claims, and office action responses whenever possible. As more guidance memos came out, we included relevant language from those as well. Sometimes this resulted in claims that had a lot of words that were unrelated to the novelty of the inventions. By this I mean, we included limitations that would be present in every computer program/system by default. Storing things, displaying things, two or more processes or machines interacting or exchanging data, receiving inputs from sensors, transforming data from one shape to another, and so on. The specifics could vary for each patent application because art units or sometimes examiners were using different standards before reasonable guidance started coming out. Some patent attorneys focused more on arguing with examiners and appealing Alice rejections. We don't like to argue with examiners, so we typically keep amending claims until they are satisfied. And, then file continuations with broader claims that omit the amendments that didn't work. This type of practice requires high budget applications and may not be practical for everyone. But generally we always could get over Alice using these techniques.
stefan_ 6y ago
This article is just missing two pictures to tell the story. Rejection rates immediately after Alice (2014): https://imgur.com/veW7vQ7 https://imgur.com/veW7vQ7 Rejection rates immediately after new Patent Office guidance (2019!): https://imgur.com/ylqD5sx https://imgur.com/ylqD5sx Apparently patent applicants were entirely unable to to adjust their applications in line with the decision for a whole 5 years where rejection rates were consistently >30% (and rising!). Then, immediately after new guidance, they all "get it" and rejection rates drop below 20%? Remember, guidance is just a summary of applicable law and precedent. It is training material. If there is a significant change in rejection rates, it means Alice was somehow misapplied for 5 years and no one ever meaningfully contested it.
- johnnyo 6y ago
  So, the patent office puts out a document telling examiners exactly what to look for, and applicants read that document and use it to craft their applications. Examiners get what they are looking for. I’m honestly not seeing what the major issue is here.
- PeterisP 6y ago
  I definitely can imagine that after the Alice decision many patent lawyers would still intentionally try to push through applications contrary to the court decision - if there's a reasonable chance that it might get through, and you can afford it, why not try it? Any actual change in patent application behavior would come only when the Patent Office would assert that yes, they mean it, and they will deny these applications, and they did it only in 2019 (and issued the guidence along with it). So yes, your suggestion that "it means Alice was somehow misapplied for 5 years and no one ever meaningfully contested it" seems quite plausible for me, it's not easy to meaningfully contest it (you need an actual situation where such a misapplied patent gets abused and gets contested) and it takes years to do so.
  - pnw_hazor 6y ago
    One of the problems with Alice is that it provides very little useful guidance to anyone. The EFF argues that Alice invalidates all software patents. Others argue that it doesn't. In fact, the only sure thing the SCOTUS said in Alice was that they are not invalidating all software patents just the bad ones, while providing no workable standard for distinguishing the bad from the good. While the majority of Justices agreed that the Alice patent was invalid, a majority did not sign on to one opinion. There were multiple concurring opinions that agreed on the result but not the reasoning. The decision provides at best an "I'll know it when I see it" standard. To say, lawyers are deliberately or intentionally acting contrary to the decision is not accurate. Lawyers will work to draft patents that comply with Alice, and they will argue that the applications they draft comply with Alice. Unfortunately, very little in the Alice decision can be used to reason for or against any particular patent because it is so incoherent.

pnw_hazor 6y ago

"Another likely explanation is that patent applicants are adjusting their applications to be in line with the decision" This. We had/have no problem getting past Alice if the application was drafted post-Alice. (Except for one art unit that was being unreasonable -- but even they seemed to have given up. Maybe their boss retired?) Nothing changed except we included additional language in the spec and claims for no other reason than to get around Alice rejections.

hcknwscommenter 6y ago
Really? YOu have some magical language in the spec and claims that universally gets around Alice but (and I'm inferring here) doesn't narrow the scope of your claims? Please please please do tell me what this magical "additional language" is. I am not trying to troll. I am genuinely curious. I have upvoted your post. You can send me a private message if you like.
- pnw_hazor 6y ago
  I call it magic language because it had nothing to with the novelty of the inventions. Note, Alice wasn't the first time this was required. We always adjust the language we use in applications as new case law comes down. Further, gibberish included in claims to defend against Alice rejections usually had nothing do with novelty because Alice was about subject matter patentability not novelty. Before the early Alice guidance came out we tried a lot of different things, much of it superficial in my opinion. Including: emphasizing that the machine did the work; showing that inputs are being transformed into something new by the machine; writing claims that show different parts of the system interacting with each other; and sometimes nonsense like "wherein each foobar is stored in a hardware memory" or "displaying the bizbaz on a hardware display." In some hard cases (mainly apps drafted before there was any Alice guidance or useful case law) we included claim limitations like "modifying performance rates based one or more sensor inputs" (e.g., CPU temperature) Later, when the first case in district court found that a patent satisfied Alice, we started including language verbatim from that case in specs, claims, and office action responses whenever possible. As more guidance memos came out, we included relevant language from those as well. Sometimes this resulted in claims that had a lot of words that were unrelated to the novelty of the inventions. By this I mean, we included limitations that would be present in every computer program/system by default. Storing things, displaying things, two or more processes or machines interacting or exchanging data, receiving inputs from sensors, transforming data from one shape to another, and so on. The specifics could vary for each patent application because art units or sometimes examiners were using different standards before reasonable guidance started coming out. Some patent attorneys focused more on arguing with examiners and appealing Alice rejections. We don't like to argue with examiners, so we typically keep amending claims until they are satisfied. And, then file continuations with broader claims that omit the amendments that didn't work. This type of practice requires high budget applications and may not be practical for everyone. But generally we always could get over Alice using these techniques.

stefan_ 6y ago

This article is just missing two pictures to tell the story. Rejection rates immediately after Alice (2014): https://imgur.com/veW7vQ7 https://imgur.com/veW7vQ7 Rejection rates immediately after new Patent Office guidance (2019!): https://imgur.com/ylqD5sx https://imgur.com/ylqD5sx Apparently patent applicants were entirely unable to to adjust their applications in line with the decision for a whole 5 years where rejection rates were consistently >30% (and rising!). Then, immediately after new guidance, they all "get it" and rejection rates drop below 20%? Remember, guidance is just a summary of applicable law and precedent. It is training material. If there is a significant change in rejection rates, it means Alice was somehow misapplied for 5 years and no one ever meaningfully contested it.

johnnyo 6y ago
So, the patent office puts out a document telling examiners exactly what to look for, and applicants read that document and use it to craft their applications. Examiners get what they are looking for. I’m honestly not seeing what the major issue is here.
PeterisP 6y ago
I definitely can imagine that after the Alice decision many patent lawyers would still intentionally try to push through applications contrary to the court decision - if there's a reasonable chance that it might get through, and you can afford it, why not try it? Any actual change in patent application behavior would come only when the Patent Office would assert that yes, they mean it, and they will deny these applications, and they did it only in 2019 (and issued the guidence along with it). So yes, your suggestion that "it means Alice was somehow misapplied for 5 years and no one ever meaningfully contested it" seems quite plausible for me, it's not easy to meaningfully contest it (you need an actual situation where such a misapplied patent gets abused and gets contested) and it takes years to do so.
- pnw_hazor 6y ago
  One of the problems with Alice is that it provides very little useful guidance to anyone. The EFF argues that Alice invalidates all software patents. Others argue that it doesn't. In fact, the only sure thing the SCOTUS said in Alice was that they are not invalidating all software patents just the bad ones, while providing no workable standard for distinguishing the bad from the good. While the majority of Justices agreed that the Alice patent was invalid, a majority did not sign on to one opinion. There were multiple concurring opinions that agreed on the result but not the reasoning. The decision provides at best an "I'll know it when I see it" standard. To say, lawyers are deliberately or intentionally acting contrary to the decision is not accurate. Lawyers will work to draft patents that comply with Alice, and they will argue that the applications they draft comply with Alice. Unfortunately, very little in the Alice decision can be used to reason for or against any particular patent because it is so incoherent.