The rate at which data, or content, is being produced for the Web and being generated for businesses has outpaced the rate at which conventional databases are evolving to better manage it all. It's a fact of life that we perceive on a gradual basis every day, but that we haven't yet acknowledged to be as significant or dangerous a trend as it is: Data is getting slower. Networks are getting bigger as the cloud is getting broader, and data that was already difficult to manage is becoming impossible. Content management systems today continue to be based on the types of structured database systems about one or two steps more evolved than dBASE. We've known they would be insufficient for the task, but we've put off the problem of composing a new architecture.
It's already too late for major IT companies to start that new architecture from square one; if a company has any hope of addressing this colossal, underappreciated problem, it will need to acquire the architectural project in progress. This is what Hewlett-Packard announced yesterday that it intends to do: acquire a software firm whose core product aims to supplant everything we know about databases, both the SQL kind and the Google kind. In its place would come a clustered approach whose goal is no less than to be the central repository for meaning in the world.
And in exchange for this, HP is willing to let go of the promise of Palm.
"Autonomy represents an opportunity for HP, for us to accelerate our vision to decisively and profitably lead a large and growing space, which is the enterprise information management space," said Léo Apotheker, the company's CEO, during yesterday's quarterly conference call (our thanks to Seeking Alpha for the transcript). "It also brings HP higher value business solutions that will help customers manage the explosion of information. But as we execute this deal, this will position HP as a leader in the large and growing space. It will complement our existing technology portfolio and enterprise strategy. It will provide differentiated IP for services and extensive vertical capability in key industries."
Apotheker's bet is that managing the explosion of information will bring higher revenue for his company than addressing a market interest that largely centered around consumers. HP knew that it was acquiring a consumer technology company when it announced its intent to purchase Palm just 16 short months ago. At the time, its executive vice president for the Personal Systems Group, Todd Bradley, said in order to make the Palm purchase profitable, it would have to exploit the potential revenue channels for webOS for business customers.
"While Palm currently has the Pre and Pixi smartphones, we see that as one space that right now is very consumer-oriented, and we will look at how we leverage our both retail and commercial channels to broaden the distribution of that set of products," Bradley said at the time (PDF transcript available here). "I think the tablet/slate products are such new markets, we see opportunities broadly for consumers, but at the same time, having just finished up our partner conference, enormous interest on behalf of channel partners with specific vertical deployments in things like healthcare and education. So I think you will see these products deployed in both market segments, consumer and commercial, and we will talk about timelines when we get closer to a completion."
That never happened. As recently as last Tuesday, consumer oriented publications were speculating on the outlook for webOS in embedded scenarios, such as automobiles, HDTVs and, yes, refrigerators, based on guidance provided by HP's public relations handlers. Some were wondering why the question of these opportunities was only being brought up now.
The final analysis of HP's progress with webOS, such as it was, will show that in this particular instance, scale - the thing both HP and Palm had said the acquisition was all about - worked against them. No one seemed to believe you could graft the same OS that powers a Pixi phone and a refrigerator to a heart monitor.
Which brings us back to Autonomy. As CEO Apotheker told analysts yesterday, HP intends to exploit the prospects for using Autonomy's technology as a foundation for a content management system. For now, that CMS would be a project for what, on the surface, seems an unlikely department: the Imaging and Printing Group (IPG). Autonomy describes this technology - which it calls Intelligent Data Operating Layer (IDOL) - as nothing less than a replacement for, a complete substitute for, a revolutionary disruption of, Google.
An Autonomy white paper (PDF available here) explains its iManage Universal Search product, which is based on its IDOL platform, this way:
Traditional search engines cannot understand the meaning of information. Unfortunately, this inability to understand information means that other documents that discuss the same idea (i.e., are relevant) but use different words, are often overlooked. Equally, documents with a meaning entirely different to that which the user searches for are frequently returned.
Users expect search to be simple, fast, comprehensive and secure. Universal Search is built on the IDOL platform, the leader in Meaning Based Computing, which forms a conceptual and contextual understanding of any piece of data, including text, voice and video, regardless of data type or storage location, performing advanced operations on that information in real-time.
Elsewhere in Autonomy's literature is a monkey wrench it hurls directly at Google, with hopes of messing up its gears. Here, the company attacks the value of Google's page ranking technology in the enterprise: "in many cases, the most popular information is also the most relevant. The importance or popularity of a Web page is approximated by counting the number of other pages that are linked to it, and by how frequently those pages are viewed by other users. This works quite well on the Internet but in the enterprise it is doomed to failure. Firstly, there are no native links between information in the enterprise. Secondly, if a user happens to be an expert, perhaps in the field of gallium arsenide laser diodes, there may be no one else interested in the subject, but it is still imperative that they find relevant information."
This is what HP is buying: an opportunity to disrupt Google. If IDOL is every bit the next stage of database evolution that Autonomy makes it out to be, then HP (at least in its executives' own minds) is not surrendering to Google at all, as some consumer publications this morning are suggesting. As HP perceives it, rather than cutting off Google's left arm, it's targeting the gut. Apotheker pointed out yesterday that Autonomy is in the midst of a transition of its product portfolio (which to date has targeted high-volume information generating verticals such as legal) to SaaS-based deployment. It's about one-third of the way, he said, and HP is very much interested in seeing Autonomy go the other two-thirds.
In other words, Apotheker is trying to make the scale argument work the other way: He envisions a scheme where a high-volume data engine that addresses high-end clientele scales down to a more general level, by way of the cloud. Along the way, he would be happy to provide collateral damage to Oracle, his arch-nemesis and the standard bearer for all things SQL.
"We will be looking at identifying as many synergies as we can, as quickly as possible," said the CEO. "The first obvious one is to give Autonomy access to our channels across the world, which will help them in a significant manner. And at the same time, we will also be working across our software teams and their software teams to actually create additional capabilities there."
The end result of which could be a restoration of the IPG division, whose cash cow at one time was quite literally black ink, to a growth position that's just as much engaged in services as every other division of the company. Even if HP stops making PCs (which apparently wouldn't hurt Apotheker's feelings all that much), it could be the standard bearer for a class of cloud-based technology that could break businesses' dependence upon both structured databases like Oracle, MySQL, and SQL Server, and contextual databases like Google.
But for this new and improved scale argument to work, customers will have to believe that you can graft the same vision of an all-seeing repository of all written meaning, to the same Android phone that gives you the traffic report and lets you pummel birds against the sides of pigs. Scale has failed HP once already.