Happy World Poetry Day! To celebrate, I’d like to share how leaders in the SEO field are applying the dynamics of poetry to search engine AI.
Bill Slawski, a guru in the SEO community and founder of SEO by the Sea, lends insight into how this future is beginning to play out:
“Creating and understanding poetry requires an ability to grasp concepts and relationships between words that might elude computers, but are often easier for humans to handle and understand. A first step toward that level of understanding is understanding the context of a query, an answer to a question that takes into consideration not only all of the words that might be contained within that query, but also past experiences – past questions that might be similar in nature that a computer has hints to from previous searches within its memory – the query sessions performed by others, and the answers that searchers seemed satisfied with when performing those searches.”
Slawski’s comment made me realize how we write poetry is also how we write for the Web. Web content often needs to be short and concise with keywords that grab your audience the same way as in poetry.
When Google hired noted futurist and leader in AI development Ray Kurzweil as their director of engineering in 2012, he already had co-invented patented programs that read and write poetry.
As Slawski pointed out on his blog discussing Kurzweil’s poetry patents, “Part of what he is working on is to get computers to talk like humans. If a computer can create poetry, it can tell what a page is ‘about.’”
In an interview with Keith Kleiner, founder of Singularity Hub, Kurzweil talked about his work at Google, saying “My project is to actually get the computers, Google computers, to understand natural language, not just do search and answer questions based on links and words, but actually understand the semantic content. That’s feasible now.”
He also explained why Google was his choice as the best place to develop these more intuitive AI. “The resources I would need were uniquely at Google,” Kurzweil said. “For example, the knowledge graph which has 700 million different concepts and billions of relationships between those concepts. That’s something you need to do what I’m doing at Google which is understanding natural language and that’s not something you can create on your own.”
While Google is studying how the poetry of natural language applies to search algorithms, Google’s search has already inspired a poetry all its own. I came across a fun project called Google Poetics that shares screenshots of Google’s recommended searches with thought-provoking results. I was surprised how well they work as poems. Here are some of my favorites shared on the site:
The future of SEO is looking rather poetic, bridging the gap between the abstract, ethereal human mind and the literal, clinical data storage of a computer. By teaching search engines to read between the lines, Google is developing a sense of the deeper meaning behind our queries.
Katie Manry is a freelance writer and editor in Tampa Bay, FL. Connect with Katie on Google+.