This site uses different types of cookies, including analytics and functional cookies (its own and from other sites). To change your cookie settings or find out more, click here. If you continue browsing our website, you accept these cookies.
Born to solve began with a few videos in the Alteryx Academy, as a way to understand how someone thinks through a problem in Alteryx, and comes to a solution. We went live at Inspire Nashville, where I got to solve a new-to-me challenge in front of about 300 of my closest data friends.
We were so excited to present this at Analyticon in New Orleans. And then the world changed. But with the Herculean efforts of @TuvyL, we were given an online platform.
Born to Solve, ADAPT edition, features one of our community’s top contributors, @danilang. Year after year, Dan blows us away with his contributions to the community, whether answering questions (over 500 accepted solutions!!), solving weekly challenges, getting stars, or having great conversations, Dan is a fountain of knowledge.
Just like every other Born to Solve event, we began with a problem that Dan hadn’t seen before. This one called on New Orleans, the original location for Analyticon this year.
Given the location of a hotel, and the locations of entertainment venues in New Orleans, find the three closest to the hotel (excluding adult establishments to keep things G-rated, and no admission because I'm spending way too much on the New Orleans food scene) and solve for how many miles it will take to go from the hotel to the first venue, then from venue to venue, and finally back to the hotel.
So here’s the challenge:
We all had a glimpse into Dan’s solving prowess as he talked us through his thought process. Catch the replay here
Another solution, my first approach:
Take a look at our preparation of the venue data. Same approach, just a different order of the tools. Neither order is faster than the other.
And yes, I crossed the streams. (For those missing out on the reference, please take the time to watch the original Ghostbusters. You can thank Dan and me later for the laughs).
After the Find Nearest tools, our approaches diverge. Both approaches take only .9 seconds to run. So is one solution better than the other? Nope! Both work, and both work quickly. We think and solve differently and that's a great thing.
But what if the spatial category is such an unknown to you that you’re put off from solving? Not to fear, this challenge can be solved with Favorites tools only. Take a look at the wonderful spatial formulae available in the Formula tool, and you may find you prefer to solve with those.
The point here is that there is no single correct solution. There isn't even a best solution here as all take about the same time to run. Three different approaches, all get to the same answer. That’s the beauty of Alteryx. It supports your individual thought and solving processes.
Let’s see together how many ways we can get through this. Give this challenge a go, and please share your solution in the comments!
You, too, are born to solve.
PS - I've attached the solutions to this post. The original workflow, unsolved, is entitled NewOrleans.yxmd
Born to Solve was lot of fun to participate in. Many thanks to @estherb47 for hosting and all the hard, behind-the-scenes work from @LeahK, @JoeM and the entire AFG team for hosting the ADAPT program
The images in the previous post only show my final, pretty solution. Things were a lot messier during the session. After using the Find Nearest tool, I was left with the task adding up the 4 distances and thing went off the rails at that point. About 45 minutes in, my workflow looked like someone had thrown a plate of spaghetti at the screen. But this just shows that solving isn't always easy and along the way we can all make mistakes and have to backtrack. It's all part of the process.
Thanks to all the attendees for their comments and encouragement during the train wreck. You people were great. One comment in particular stuck in my head. This person said something similar to "Create a bounding rectangle and calculate the length of the perimeter" Unfortunately, I don't remember the person's name, (and I also can't access the chat window anymore) but I think It had at least one "V" in it. In any case, here's an implementation of that idea.
The 1000s of tools that I was throwing at the canvas were elegantly replaced with a Poly Build tool to create the convex hull and a Spatial Info tool to calculate length of the perimeter.
Here's an image of the final path from the New Orleans Marriott to the three bars and the final stumble back home.
Of course, this is the distance as the crow flies and involves some serious parcour to climb the various buildings in the way. A more accurate path would follow the streets, requiring significantly less athletic ability. But this, as my professors used to say, is left as an exercise for the student.