Dataspaces in Denmark: The 2008 Boye Conference
October 22, 2008
Earlier this year, the engaging Janus Boye asked me to give a talk and offer a tutorial at his content management and information access conference. The program is located here, and you will see a line up that tackles some of the most pressing issues facing organizations today. The conference is held in Arhus, Denmark. My first visit was a delight. I could walk to a restaurant and connect. Arhus may be one of the most wired and wireless savvy cities I’ve visited.
About a year ago, before Google decided I was Kentucky vermin, I discovered in the open source literature, a reference to a technology with which I was not familiar. In the last year, I have pulled this information thread. After much work, I believe I have discovered the basics of one of Google’s most interesting and least known technology initiatives.
Source: http://www.lohninger.com/helpcsuite/img/kohonen1.gif
Unlike some of the other innovations I have described in my 2005 The Google Legacy and my 2007 Google Version 2.0 reports. Those documents relied extensively on Google’s own patent documents. This most recent discovery reports information in Bell Labs’s patents, various presentations by Google researchers, and published journal articles with unusual names; for example, “Information Manifold”. The research also pointed to work at Stanford University and a professor who, I believe, has been involved to some degree with Google’s team leader. I also learned of a Google acquisition in 2006, which does not appear in the Wikipedia list of Google acquisitions. Although the deal was reported in several Web logs, no one dug into the company’s technology or its now-dark classified ad site.
In my tutorial on November 4, 2008, I will review the details of this Google research project. Some of the information is in step with broader industry developments. Other parts of the initiative push into completely new areas. Let me give you a taste of the functionality that the open source literature suggests Google can do with this transforming technology.
Imagine that you are looking at a draft of a contract. You want to know what changes were made to that document, by whom, and when. In addition, you want to know if anyone outside of your firm has received a version of the document via email or if a reference to that document appears in a published article. In short, you want to know with reasonable certainty the lineage or family tree of the document and the touches it experienced.
It’s clear from this example that the index information needed to handle these additional types of queries and content access are not easily available. If you have a system like Tacit Software’s running, you can get some of the information. Now think about having access to this type of meta metadata for an organization or for a group of people such as those involved in a financial deal.
A typical database query won’t deliver lineage and certainty. A typical database cannot provide the meta metadata. A commercial off the shelf data management system such as the one available from InfoBright can’t either. Even the United Kingdom’s monitoring system cannot deliver what the Google technology write ups suggest.
There are a limited number of slots available for this session. I won’t be providing hard copy handouts, but I will post for attendees of the tutorial a list of links to some of the source material I reference. I have discovered that a number of the documents which were available in late 2007 and early 2008 have been removed from servers hosting them. I have asked Google about this technology, but the company continues to ignore my requests.
I don’t know if Google will deploy this powerful “beyond database” technology. I do know that it is important to learn about its basics. Other firms are almost certain to move in a similar direction. Some innovations occur simultaneously in geographic distant regions. You can get additional information abut the conference here.
Stephen Arnold, October 22, 2008
Comments
-
- Subscribe to Beyond Search
Feature archive
News archive
-
