In Search of Search: Separating the "What" from the "Why"

"Let's take an example of a common search platform," Devin continued. "In order to search, let's just say, a SharePoint repository, I usually need to stand up a server that talks to SharePoint, then create its own index, then makes that index viewable to its own proprietary interface. That index, the server that the index sits on and the transmission of the data to the index and then back out of it is three times the infrastructure I need if I'm just talking directly to SharePoint. And we're not going to get rid of SharePoint anytime soon; you know as well as I do... it's a weed."

I tried to steer the conversation back to the viability of "enterprise search" versus the application of search to individual business functions. "It would make my life a heck of a lot easier if I could wave a wand and make enterprise search possible," he said. "There are companies out there trying to create that red button that turns on enterprise search, and spending gobs of money trying to do it."

But the fact remains that enterprise search is more of a state of mind than a reality. There is no such thing as enterprise search; there is merely search applied to solve various business problems. Which makes enterprise search more of a strategy than a technology. It's about controlling those instances of search so that the integrations that need to occur can happen. And those that don't need to occur, you aren't wasting money trying to force them. "That's our analysis, too," agreed Devin. "The products that purport to be enterprise search can be bright and shiny, but they are a very expensive undertaking."

In their defense, I said, there are not many search vendors who would try to tell you that a boil-the-ocean approach is wise. At least not anymore. What they WILL tell you, though, is that you should attack a specific business problem, solve that and move to the next one. In that way they hope to virally grow within an organization to the point that they become the de facto corporate standard. Voila! Enterprise search! And it's not really an ineffective way to approach it. Department heads get together either at formal meetings or informal company gatherings and water coolers and talk about their successes. They then try to replicate that success, one department and one function at a time. "That's right," agreed Devin. "And if a product has become a corporate standard, that's generally enough business justification to invest in it. But how many different solutions to different business problems does an organization need? If a certain search product works in one example, it might not work in another. So we get many solutions working in organizations, and the ‘hot offerings' in the market keep coming."

Who Knows What?

I was curious about the overall knowledge in the legal community regarding search technology. Are litigators and judges up to speed? "At the highest levels, such as some of the federal district courts, they are," Devin insisted. "However, there is so much noise in the marketspace. For example: What does litigation hold actually mean? What does preservation actually mean? There are many well-spoken, well-read judges who disagree with one another and are issuing conflicting case law about some very, very important activities in the e-discovery lifecycle. That's the biggest problem we have right now," he said.

"After you get below that federal level, the education is low. It's the wild, wild west, quite literally. For instance, when we use the term ‘forensics,' the immediate connotation is that we capture every one and zero on a hard disk. That's obviously not necessary, but that's been the impression for so many years that it's ingrained. Targeted search in a forensic effort is maybe only eight to 10 years old. The utter lack of education—and the lack of standards across the board-is the biggest problem we face."

He shared a personal example. "I looked at some business documents the other day on my wife's iPad. Two days later I went back, and lo and behold, those documents had been downloaded onto the iPad. I thought I had merely viewed them in a Web interface, and that was that. I never hit a button that said ‘download.' But the network isn't where it needs to be, especially for wireless devices, so they can't buffer these documents in the cloud somewhere and allow simple viewing. They download them to the physical mobile device. I bet you there are many IT managers who don't know that. They may fear that, but many aren't sure."

Education is obviously the key, and there isn't much of it, according to Devin. We'll try to remedy that in this KMWorld White Paper and those to come in the future. But in the meantime, it's instructive to think not about the "what" of technology, but the "why" of technology. It's a theme we stress when we undertake one of these documents, and we try to impress on the other participants. Search is probably one of the greatest examples of a technology in "search" (pun intended) of a solution. Devin has found one, and the other participants in the paper would quite likely agree. See if you do.  

KMWorld Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues