How Search Can Improve Content Discovery On- & Offline

You’re probably pretty happy with Google search today, right? It’s incredibly fast, extremely reliable and almost always delivers the desired results. What more could you ask for?

I think the problem with today’s search solutions is that we’ve limited them to what’s online. If the content has a web address and it’s been crawled by the major engines it’s properly analyzed and presented in search results.

But what about everything else? Once again, Evernote is a terrific example of what could be.

I’m a huge Evernote fan and I’ve configured it so that all my notes are exposed and retrievable in a Google search. Alongside the standard web, news, maps, images, etc., search results categories, Google also shows a frame with Evernote’s Web Clipper results. Simply put, a single Google search produces results from the web as well as my Evernote archive. Simple, yet powerful.

Why does it have to stop with the web and Evernote? Why can’t one search be configured to retrieve results from all my content streams?

Let’s start with the documents on my computer and in the cloud. They’re mostly Office applications, so a search needs to understand the structure of Word, Excel, and Powerpoint documents. I’m not talking about simply searching file names; this search functionality needs to know whether the phrase is buried in the document itself.

Don’t forget about Outlook and all the other email applications. Search needs to sift through everything in my inbox, folders, and attachments.

How about all the digital books, newspapers, and magazines I read or scan every week? My search tool needs to capture, index, and report back on all that activity as well. I sometimes rate articles and books I read, so the search algorithm needs to understand those rankings and include them in its algorithm, pushing higher-rated results towards the top.

Let’s also not forget about websites I’ve visited. This search tool should understand which sites I frequently visit and which pages I’ve spent more time on, reflecting the fact that I’m reading rather than scanning. This too is critical information for the search algorithm.

Next, it needs to understand my social graph and factor that into the search results. I’m much more active on Twitter than Facebook, for example, so what are the most recent relevant tweets that belong in my search results?

I realize this starts to clutter the results page. That’s why it all has to be configurable by the user. Clicking on/off checkboxes in a list should allow me to show or hide the various sources in search results.

I’m able to search each of these sources individually today, of course, but there’s no uber-search tool allowing me to consolidate and search across all sources with one query.

Finally, and here’s where it gets even more interesting, I want the ability to curate and share my search results. Today you can do this by sharing the url from the results page; for example, here’s a Google search for my employer, Olive Software. That’s a start, but now I want to insert links to other sources, including all the ones noted above (e.g., documents, emails, ebooks, etc.).

Yes, there are countless sharing, opt-in, privacy, and copyright issues to navigate before this vision becomes a reality. But imagine how powerful the results will be when these capabilities become standard features in every search engine.