This week, I was absent from class due to attending the NDSI conference in Boston, so this is going to be a short reflection on the group project. Apologies for not being able to reflect on the in-class discussion of open source within industry.

Our main contributions to scikit-learn have not changed in the past week, i.e., making automatic validation for public functions. However, last week is the week in which the least number of our pull request were accepted. The maintainer who usually reviewed my pull requests did not appear a single time in the past week, and Jiawei and Romee got only a few of their minor pull requests merged.

Due to these concerns, we began to search more urgently for new issues to work on. As I have mentioned in my previous post, BayesSearchCV is a possible new feature to move into scikit-learn. However, except from one response saying that it might be more suitable to move to some smaller projects such as scikit-learn-extra, there haven’t been any more voices from the maintainer team. Though the end of the semester is already approaching, and that maintainers haven’t made a final decision, we may decide to work on BayesSearchCV next week. Even if is likely to be unaccepted, we agreed that making an attempt is better than merely waiting and doing nothing.

One thing I have been thinking about since long ago is to directly reach out do some maintainers via email. However, considering that scikit-learn is a loose team that work in spare time (as documented in the contribution guide), urging them may come off as impolite. Also since our group is not in a desparate situation where we have made no pull requests, doing so is unnecessary. Regardless, though I do not know what my teammates think about contributing to scikit-learn, I myself find it interesting and I may continue to make contributions even after the class ends. Waiting is not that bad then, in this sense.

Open Source Within Industry

Though I did not attend the in-class discussion, I still want to say a few words about open source within industry, especially since open source has become more and more widely adopted concept within various industries in recent years. The idea of freely accessible and modifiable software and technology has led to a significant shift in the way businesses operate. Open source solutions not only tend to be more cost-effective, but can also be customzied to meet the specific needs.

For instance in the healthcare industry, we have the OpenEMR platform designed for maintaining patient health records, which is widely used around the world. OpenEMR iis highly customizable, allowing healthcare providers to tailor the software to meet specific needs. Its wide range of features include patient scheduling, medical billing, electronic prescribing, and so on. It is also designed work with other healthcare software systems and share patient data securely. I think this is a good example of the power of open source in the industry.

Summary

In this blog post, I have mainly talked about the diffulties in our group project, and my attitude on dealing with this problem. I have also briefly mentioned open source within industry, and apologies if that part is completely off topic from the discussions in class. Finally, thank you very much for your meticulous reading and hope you enjoy it.