Data Science

The world of data science, from scratch, is flooded with open source machine learning software such as PyTorch, TensorFlow, Python and many more. Most of them think that open source is the most widely used tool, but it is not the case.

Can you guess which tool is widely used?  It's Excel. 

Referring to Excel spreadsheet, the CEO of Anaconda, Peter Wang says that, “It is the most successful programming system in the history of homo sapiens”, in one of the interviews. As it is very easy to run, it can be high-yielding too. Due to its ease in data science and business analytics, Wang thinks that Excel succeeded without an open source. But he also believes that Python will attain success only with the help of open-source in data science.

More of a process

The software application has always been seen as a product for which the customer pays, but any product can never quench the thirst of the customers, so it is better to call software a process that keeps evolving in the big data science sector. So, Wang emphasis's saying that open source provides an opportunity to refine and better the software as a service. 

Open-source software inspires more and more people to participate in this process to attain success. Wang estimates that 90-95% of the users are left behind in this process of refining software. This is because the left out people believe that others are going to deliver the value of the software on their behalf.  In contrast to it, Wang says that the open-source operating software for data science has become so fruitful with the addition of users who in turn became makers in the world of data science. 

According to him, one of the biggest reasons for Python’s success is that it made things easier for even an average person, grabbing an opportunity to choose data science even though most people aren't scripting Python programs.   

Way ahead 

Open sources are vital for the new developments to take place in data science and machine learning in the future. Wang refers to Python as a remix culture of learning and teaching things. The Python world is full of codes of machine language. The coders lay the foundation for upcoming builders to make changes on it, says Wang. By just limiting the user layer and user-facing API around the process is not the right thing to contribute to data science. The developments in data science can pave a way for others to get motivated to participate and explore.

Excel better for innovations?

Wang thinks it's a confident yes. He thinks that when people start working with others in a team the results are always better. Excel has the ability to change data science forever that can broadly define things more clearly.