Chien and Wu: Decoding Patentable Subject Matter

6 thoughts on “Chien and Wu: Decoding Patentable Subject Matter”

1. 5

   I think their proper title should be “large-corporate Advocates Chien and Wu.”
2. 4

   As to the title of “Decoding Patentable Subject Matter,”

   Let’s not get sloppy.

   You have patent-eligible subject matter concerns under the law, and then you have patentable aspects of law.

   The two are NOT interchangeable.
3. 3

   >>Colleen Chien and Jiun-Ying Wu

   Both of these people need to give a history going back 10 years of where all their money has come from. Their work is not academic, but paid advocacy without ethics.
4. 2

   (Resubmitting with some HTML formatting changes for readability. Wish list for 2019: a PatO comments section that supports editing.)

   Very interesting. Initial thoughts:

   To make office action data much more accessible, the OCE used artificial intelligence methods to extract information from office actions and code each one based on the extent and type of rejection. The initial
   release of the dataset provided insight into 4.4 million office actions mailed from 2008 through mid-July 2017 for 2.2 million unique patent applications.

   The resulting file was 1.31 GB, too large to be opened and processed by standard spreadsheet software, but through Google’s BigQuery cloud software, it is now possible to access the dataset from a standard laptop…

   …or just, I don’t know, dump it into a database? I’ve got a sqlite database of every patent application published by the PTO, with a full-text index of the specification, claims, etc. – it’s 750gb and it responds almost instantly to any query. Pretty basic stuff.

   Within impacted classes of TC3600 (“36BM”), the 101 rejection rate grew from 25% to 81% in the month after
   Alice, and remained above 75%… Among medical diagnostic (“MedDx”) applications, the 101 rejection rate grew from 7% to 32% in the month after Mayo and continued to climb to a high of 64% and to 78% among final office actions just prior to abandonment. … However outside of impacted areas, the footprint of 101 remained small, appearing in under 15% of all office actions.

   Well, there’s also been an uptick in TCs 21/24/26/28.

   What’s more – there needs to be some more decoding done here, because not all § 101 rejections are alike.

   Pre-Alice, I encountered a steady stream of Nuijten rejections with minor quibbles about CRM language. What’s more – it’s common practice to leave that as an open issue until substantive grounds of rejection are addressed, for two reasons: (1) to conserve prosecution history estoppel, and (2) because examiners become more flexible, and open to a broader range of options, when the only obstacles between the application and an allowance are minor formalities.

   Post-Alice, the § 101 rejections are more earnest, and examiners are digging in their heels – including where they really can’t, but where the deferential nature of the PTO’s (former?) tolerance for examiners’ § 101 decisions was emboldening. Thanks to Berkheimer and Director Iancu’s perspective, the extreme deference may get reined in.

   The upshot is that lumping all § 101 rejections together in these aggregate metrics hides an important distinction – and the magnitude of the § 101 uptick.

   Since the release of office action data, we have looked for evidence that the two-step test had transformed patent prosecution as the headlines would suggest. We did not find it, because, as the PTO report notes, a relatively small share of office actions – 11% – actually contain 101 rejections.

   When patents are grouped by WIPO CPC code, rather than PTO AU code, a sustained increase in 101 rejections following Alice can be discerned in four out of the five major technology sectors (except mechanical engineering) but in no month or any sector does the prevalence of 101 rise above 20%.

   Wait, what? It’s awfully difficult to reconcile these two statements. The conclusions dramatically understate the significance of the impact.

   My conclusion, looking at the same numbers is quite different: Alice and Mayo have created a new grounds of rejection – one that did not formally exist under Bilski and earlier cases – that is affecting a significant number of applications across 80% of the tech centers, and that is dramatically increasing rejections in certain areas.

   The authors’ conclusion from their own data is so implausible that, in my book, it damages the authors’ credibility.
   1. 2.1

      DS: The authors’ conclusion from their own data is so implausible

      What conclusion are you referring to? This one?

      Overall, 101 has not had a huge impact on patent prosecution.

      That’s not “implausible”. That’s just a fact. Nobody is stating that the Supreme Court’s clarifying 101 jurisprudence hasn’t had a big impact on cr @ppy logic and correlation patents, or even on half-baked and incredibly broad nutraceutical / pharma claims that read on natural products. Of course that’s been the case. But overall patent prosecution has not been changed much. The bigger impact has been on litigation which is also not surprising because it was litigators who recognized the problems and guided the Supreme Court to its solutions. I grant you that the solutions are not ideal but that’s because in Alice especially the defense’s hands were tied and they weren’t able to address the bigger issue which is causing all the alleged “confusion” for people like you.
5. 1

   Very interesting. Initial thoughts:

   To make office action data much more accessible, the OCE used artificial intelligence methods to extract information from office actions and code each one based on the extent and type of rejection. The initial
   release of the dataset provided insight into 4.4 million office actions mailed from 2008 through mid-July 2017 for 2.2 million unique patent applications.

   The resulting file was 1.31 GB, too large to be opened and processed by standard spreadsheet software, but through Google’s BigQuery cloud software, it is now possible to access the dataset from a standard laptop…

   …or just, I don’t know, dump it into a database? I’ve got a sqlite database of every patent application published by the PTO, with a full-text index of the specification, claims, etc. – it’s 750gb and it responds almost instantly to any query. Pretty basic stuff.

   Well, there’s also been an uptick in TCs 21/24/26/28.

   What’s more – there needs to be some more decoding done here, because not all § 101 rejections are alike.

   Pre-Alice, I encountered a steady stream of Nuijten rejections with minor quibbles about CRM language. What’s more – it’s common practice to leave that as an open issue until substantive grounds of rejection are addressed, for two reasons: (1) to conserve prosecution history estoppel, and (2) because examiners become more flexible, and open to a broader range of options, when the only obstacles between the application and an allowance are minor formalities.

   Post-Alice, the § 101 rejections are more earnest, and examiners are digging in their heels – including where they really can’t, but where the deferential nature of the PTO’s (former?) tolerance for examiners’ § 101 decisions was emboldening. Thanks to Berkheimer and Director Iancu’s perspective, the extreme deference may get reined in.

   Since the release of office action data, we have looked for evidence that the two-step test had transformed patent prosecution as the headlines would suggest. We did not find it, because, as the PTO report notes, a relatively small share of office actions – 11% – actually contain 101 rejections.

   Appendix A of the same document backs up that conclusion, listing 21 art units in the 2100-2900s that demonstrate rejection rates over 50%. (It’s not clear from this data how many of these are § 101, etc. … or, frankly, why the appendix is even included, as it is neither explained nor referenced in the body of the article…?!)

   When patents are grouped by WIPO CPC code, rather than PTO AU code, a sustained increase in 101 rejections following Alice can be discerned in four out of the five major technology sectors (except mechanical engineering) but in no month or any sector does the prevalence of 101 rise above 20%.

   Wait, what? It’s awfully difficult to reconcile these two statements. The conclusions dramatically understate the significance of the impact.

   My conclusion, looking at the same numbers is quite different: Alice and Mayo have created a new grounds of rejection – one that did not formally exist under Bilski and earlier cases – that is affecting a significant number of applications across 80% of the tech centers, and that is dramatically increasing rejections in certain areas.

   The authors’ conclusion from their own data is so implausible that, in my book, it damages the authors’ credibility.

Comments are closed.