Celebrating JMP Champions: Hans De Roos, Octoplus Information Solutions

[#1]

Hans De Roos, Manager, Octoplus Information Systems

 

Hans taking part in one of JMP's Analytically Speaking sessions.

What is your favorite feature in JMP?

Before I was a JMP user, I was a SAS user. In fact, I was likely the first SAS user in South Africa, perhaps even the first in Africa. I was very impressed with SAS and could immediately replace some of my own developed statistical programs (at that stage, developed in FORTRAN, PL1 and APL – IBM Matrix language).

Around 2003 I became aware of JMP, probably JMP 4. The thing I liked the most about JMP was the ability to open a statistical program without knowing exactly which statistics you wanted to run on the data – the exploratory part of JMP. Most other programs required a pre-defined idea about the type of statistics you wanted to run.

What was your first job (ever)?

After matriculating in the Netherlands, I got a job as an assistant in the surveyor’s office of The Snowy Mountains Authority in NSW in Australia. This was a major adjustment for me as I moved from a relatively protected environment to a company that, at least at my level, consisted mainly of immigrants from Eastern European countries. I learned a lot – including how to play chess and poker and how to drink beer.

What is your proudest professional moment?

The proudest moment still happens today when I give a demo on JMP – the “wow” moments, when you show some of the features of JMP, like the link between your graph, your data and other analysis based on interactive selections. In addition, my association with a product like JMP and the developers – it is almost part of my family.

How did you get interested in the field of statistics? Is it something you always knew you wanted to do, or were you inspired by someone/something?

I started studying statistics years ago. At the time, studying statistics was not that popular. I studied mathematics and statistics for my B-degree and concentrated on statistics when I was furthering my studies, mainly since our lecturers were so inspiring.

What do you like most about the work you do?

When developing JMP applications, you meet people from different disciplines. You can learn from them, and at the same time, you share your knowledge with them.

How are you currently using JMP?

In addition to marketing JMP in Africa, we provide training to JMP users. Currently, we are developing a number of add-ins for the mining industry.  

What is the first project you worked on using JMP?

My first project was self-initiated. I developed a stock exchange application that I never really used, but it is still on my bucket list to finish, especially now with the current data suppliers available.  

Is there anything you would like to say to JMP development or John Sall?

First, I would say that I’m proud to be part of JMP software. For me to give advice for the future would be a bit arrogant. I would never be able to create a product like JMP. The success of JMP is very much dependent on the way JMP is developed. John is not only the CEO of JMP, but at the same time, a developer.

How do you see the field of data science progressing in the next 20 to 50 years?

I do not even know what will happen in the short term. One of the big problems in statistics is that some statistics do not mean anything if the underlying assumptions are not met. Also, how to interpret some of the results is still a challenge. I think we will see statistical methods being applied by people with limited statistical backgrounds. I expect that future software will reach out to solve some of those challenges. 

Artificial intelligence is also playing a critical role. I can see a scenario where you buy data along with your software, data that is customized and relevant to your industry. The data will have “certified” statistical properties to support particular assumptions and statistical techniques. Training will be available to augment the data with our own gathered data, based on some design. The AI will adapt itself (by updating the algorithms it is based on) to the new added data. Data will be generated for almost any discipline – industrial, social, economic, medical – as well as new disciples not necessarily given much attention today, like music and emotions. How could statistics support a community that has no jobs as defined in today’s terms? For people whose work is taken over by robots or for people who have no access to resources, how can we create a sustainable community to foster the self-worth of its inhabitants?

Biotechnology, including genetic analysis, will become a major focus for new technology. As seen by the Apple watch and others, the prevention of sicknesses can dramatically be increased by knowing more about the body, and in particular, your unique body. Small monitors will be permanently placed on you. It will also not be limited to physical performance but also to the mind’s performance – emotions, stress, etc. Your generic expressions will be stored and monitored. Medicine will be more focused on the person taking the prescription, meaning less trial and error.

What advice would you give a beginning JMP user?

Try to maintain a discipline in your framing and approach. As the tools become better and faster, there is no substitute for sound scientific methods.