CS squared and B cubed
I completely agree with you that the “devil is in the data”. In this case, the availability of training data is the key element to success. Thus, I believe John Deere is much more set up for success than any other start-ups because of its unique access to create training data through its machines. This is a prime example of training data being a barrier to entry. Deere’s machines are already used by farmers; there is no additional work to do for adoption.
In other cases, start-ups can get access to data through signing exclusive contracts from data providers, or through creating the data themselves (e.g., deploying field teams or cameras/drones to create picture data). However, for agriculture, you need data across large spans of land, and Deere is uniquely positioned to take advantage of it.
However, a question remains on whether Deere should monetize this simply by selling the rights to the data, and focus on their core competency of machines, or whether they should develop new capabilities for machine learning.
I completely agree with your argument that the human element needs to be considered along with the data in making critical decisions. We have learned in class that machine learning is imperfect, and especially prone to bias. I actually think that companies like Netflix should also employ some traditional customer research methods such as surveys and focus groups to complement the data.
I am also intrigued by your question around how can Netflix capitalize on new trends, or “innovate” on its content, instead of creating more and more similar shows based on existing viewership data. I think two concepts can help here: 1) prototyping, and 2) open innovation. Netflix can figure out a prototype equivalent for content, for example, an animated trailer that illustrates the concept of the show. The goal is to have relatively cheap way of building the prototype, and allow customer reactions to test its commercial value. Another idea is that Netflix can leverage open innovation to allow its creative views to submit content ideas for the rest of the community to vote on.
I see quality standards and FAA control as key challenges to overcome. As we learned in class, with new disruptive technologies, we tend to hold 0% defect or failure rate as our mental standard, when rationally the quality standard should just be higher than what we have today. If an airplane accident happens, and it was revealed that many of its parts were manufactured through 3D printing, that could hinder the overall progress significantly even if it wasn’t at fault.
Secondly, due to the size and complexity of airplanes, I don’t see a full plane being produced through additive manufacturing in the near future. The value is much more likely from parts, service parts, etc.
Similar to many of the comments above, I’m skeptical about mass production through additive manufacturing in a meaningful timeline. I also agree that the key value-add of 3D printing is on customization. 3D printing enables flexible production processes and customizable products. An interesting implication that hasn’t been mentioned is the 3D printing’s impact on the supply chain dynamics. With customization, Nike products are likely to move to a “made-to-order” mode, which is more fit for Direct-to-Consumer model, which may remove the wholesaler and retailer players in the supply chain. Manufacturing plant management will also look different, as the type of inventory would change, and overall inventory levels will likely drop.
Because of the biases identified in your article, I’m equally skeptical about how meaningful the self-reported patient data is. I’m more worried about the self-reporting bias than population bias. Based on the screenshots you shared, it seems that the platform is collecting demographics data, which can be coupled with the patients data. If handled appropriately, the population bias may be mitigated. However, for self-reporting bias, it is harder to address. One way can be to develop machine learning algorithms to detect false data, e.g., patients who are reporting symptoms that are highly unlikely to happen together. Another way is to integrate medical records data, or partner with physicians to review and approve the reported data on the platform for the patients they see.
As for the value PLM creates, it should focus on the patient community. Thinking more about monetization, its patients community can also be ripe ground for identifying clinical trial targets.
I’m very impressed by HackerOne’s business model, as it essentially created a legal platform and a market for hacker activities. Your point about the education space as a next step for HackerOne is particularly interesting. On one hand, it increases the supply of the platform contributors, and further builds the credibility of the platform’s expertise. The partnership with universities is a brilliant idea. I can even picture a similar model to Kaggle for machine learning problems, where students may work on real company vulnerabilities for their coursework, which increases HackerOne’s supply of researchers, and at the same time trains the students. However, I’d like to see more control systems from HackerOne to prevent malicious actors to abuse the hacking knowledge they learned. For example, if a hacker found a particularly lucrative vulnerability, what’s to stop the hacker from profiting from that vs. getting paid by the company? Should the hackers violate the rules, are they registered in a way that can be identifiable?