Jupyter — It has its purpose but are we using it beyond its purpose — Quick thoughts
As good as the Jupyter development environment is, it is being used beyond its purpose and it is creating Python developers who become so dependent on real-time response.
Python is a general purpose language and it can elegantly accomplish object-oriented concepts and different design patterns including dependency injection that for example Java Spring shines in. You can also easily organize your code into modules using Python modules and Python eggs.
I understand why the real-time nature of Jupyter notebooks is useful when doing some database queries using Pandas or PySpark, but at some point that flat-structured code (aka script) needs to be modularized (turned into code) and used outside the notebooks in non-UI automated fashion.
This is why in the enterprise world of Java and .NET, Python is associated with “quick script” applications. When it comes to building microservices, it would be good to see Python on the equal playground with Java, C# .NET and NodeJS.
Yes, Python is dominating the world of machine learning and DevOps, but what about enterprises using it to build their backend microservices. What if the intermediate step to that destination was the DevOps building capabilities in the form of RESTful services to enable application software engineers in the process of building product value.
Thank you for reading this article. Keep geeking out !
Almir Mustafic