Well, here I am, near the airport, sadly not to go to some exotic, foreign land, but to take a shuttle to the Crown Plaza Hotel and visit the Power Platform World Tour.
My knowledge network currently stretches as far as Power BI, so I’m pretty excited to see the other components within the Power Platform, such as PowerApps, Power Automate, etc. And since I’ve been dubbed as the ‘BI Fairy’, let’s see how far the magic kingdom of Power Platform reaches.
First up, the magical realm of Power BI Data Flows.
Dataflows in Power BI
When it comes to unifying data, the generic way of working is creating a data warehouse that combines the data from different services. Usually attached to this is a specialist that creates reports for business users. Which is where the battle of realms starts.
A specialist creates a report for end-users, but one end-user wants another report for a quick insight but has to wait on a specialist and so the circle continues. I wouldn’t be writing about it, if I didn’t learn a new solution of course. Power BI data flows is designed to create pre-defined datasets from multiple sources that can be used in multiple reports. When one part of the flow gets refreshed all the attached components get refreshed as well. So, to put it into practical terms: a specialist can pre-define datasets. A business user can then use these datasets to create their own report within Power BI. Having a ready-to-go report for their specific needs without having to wait for a specialist. Sounds like magic to my ears.
My next stop in the Power Platform kingdom is the magic called AI.
Power Platform offers an AI builder, designed to be a low-code AI capability. It consist of a few prebuild models: business card reader, sentiment analysis, language detection, etc. And on the other side custom models: prediction, form processing, text classification, … But the one that holds the most magic, to me, is the object detection. The basic concept is loading in images containing objects and teaching the model the correct size and shape of the object and adding the correct description to it. After teaching the model, all you need to do is take a picture of an object and send it to the AI builder. The model then detects the object and tells you what kind of product it is.
This being the simplified explanation but, as you might of guessed, it got my imagination running. I wouldn’t be a fairy if I didn’t see a possible application for it: The object description could, if enhanced by a third dimension namely depth, perhaps be used in a warehouse to count the amount of products on the self, allowing a more accurate stock, reducing the human hours behind such a process.
Power Platform and beyond
Instead of mesmerizing you some more with my thoughts on the other presentations, I’ll wrap it up with my personal experience at the World Tour.
Unfortunately I don’t have the magic power to duplicate myself, so I had to select the presentations where I thought I would learn the most. Some lived up to their expectations, others not so much. Even though I might not have had as much hands on practice, as I was hoping, I did get a significant insight on what Power Platform entails. And, above all, I’ve met some new and interesting people. The exiting part to me is knowing that there is a lot more for me to learn and try. I’ve already ‘geek-messaged’ my co-workers with “I want to try that” or “Look how cool”.
I guess I can conclude that Power Platform is indeed a vast and magical kingdom.