OpenSource Connections: The Ghost in the Search Machine - An Elastic{ON}16 Recap

I'm trying to put my finger on why I enjoyed Elastic{ON}16 so much. Certainly the talks were top-notch. I especially enjoyed the BM25 text scoring talk — which walked through the math behind relevance scoring. I loved learning how Eventbrite builds recommender systems using Elasticsearch. The walk through of New York Times search had me spinning with ideas (which I blogged about) and Mark Harwood’s graph search demo seeded my colleagues and me with many ideas. 

But I think my favorite part was how Elastic{ON}16 enabled amazing conversations. See I just finished writing this book on search relevance. So our Elasticsearch consulting company was primed to have a ton of awesome conversations about how to improve search. We weren't disappointed. Lunch, the coffee line, walking around, and our sponsor booth seeded innumerable intersecting discussions. We got to troubleshoot with engineers from Apple, Facebook, and Roku about their challenges and to share our enthusiasm for the Elastic Stack.

This "hallway track" as it's called makes being a speaker even more inviting. You know your talk will trigger a dozen followup conversations. Personally, I don't want to speak to sound smart. I want to speak to be stumped by the audience and learn. I want to be asked hard questions — and most importantly to find the folks better at what I'm talking about than me!

This made being a "spotlight theater" speaker feel especially attractive. I enjoyed how speaking was very one-on-one and conversational. It let others share ideas that I hadn't thought of — which only created a snowball of additional ideas and conversations. The small, 50-seat "spotlight" venue enabled this back and forth, which I loved.

Being a speaker at Elastic{ON}16 is also refreshing due to the amount of prep Elastic puts into you as a speaker. They meet with you, help you refine your slides, and make sure you're comfortable. You also meet with your stage manager — thanks Auberie! So for anyone uncomfortable with speaking, this is a great first "major conference" experience where you get a tad bit of extra guidance on the road to success.

Oh, so what did I talk about? My talk The Ghost in the Search Machine highlights all the crazy and weird ways you can use Elasticsearch's inverted index to find *anything* — images, at risk students etc. 


And if you'd like to see it or meet me live, catch me at any one of these Elastic  meetups (Washington DC or Dallas), with even more to be announced!

Doug Turnbull is a search relevance consultant at OpenSource Connections. Author of Relevant Search. Doug crafts search/recommendation solutions that “get” users. To do this, Doug uses Elasticsearch, sprinkling a little natural language processing and machine learning on top for good measure.

  • We're hiring

    Work for a global, distributed team where finding someone like you is just a Zoom meeting away. Flexible work with impact? Development opportunities from the start?