Intelligent Information Management

Intelligent Information Management

Disclaimer : the following is meant to be read in an engineering context without any philosophical consideration of privacy issues.

“Intelligent Information Management”

What does the above mean to you ?  When I first saw this, my immediate reaction was to question if the author meant Intelligent Information, which made me think about the definition of Information.  In my mind, Information was Intelligent Data or Data with Knowledge applied …  and I apologise as I mean that in an IT domain, as the word information has many definitions.

Of course if I consider a situation where there is unintelligent information, there are many scenarios … especially when we consider the semantic web … it is possible to have a situation where you can capture the rainfall against a pump … now that sounds like unintelligent information …

I then reflected a bit further and wondered if the author meant the Intelligent Management of Information ie Intelligent “Information Management” …now that made more sense … it did cross my mind that perhaps there was a counter conference running in a parallel universe called the unintelligent management of information … I would hope not.

Let’s consider however, if we did not apply intelligence  to information management (and let’s consider that our information is intelligent, so we do not capture rainfall against a pump) is that what the semantic web is supposed to enable ?  User freedom ?  As a system builder, I am always concerned that I can only answer yesterday’s questions well, and at best 80% of tomorrows questions not very well.  This is the power that the semantic web (and here I am considering an ISO 15926 implementation of the semantic web) delivers.  In one of my presentations I discuss the difference between wikipedia and a semantic wiki.  Wikipedia can only answer questions that someone has already answered but no relationships or knowledge can be applied against similar postings …  The example I give is that Wikipedia cannot answer the question “What are the top 10 cities by population in Asia / Europe / America” even though it has the answers to the population of these cities, it takes a semantic wiki to enable a user to ask the question and get the response from the existing data.

Here then lies the dilemma I have with Management of Information (from a semantic web perspective) and hopefully these questions will be asked and some answered at the Semantic Days conference.

Is management a restriction on users questions ?  How will we know what our users want to do tomorrow ?  Just as the web enables free thoughts and movement between “adjacent problems” and thus can enable a user to have much larger connected networks, so our semantic information can enable users to follow trains of related information.  Imagine if we had the semantic web on the bridge of the Titanic … as the manager of the information system for the Titanic, I may have concentrated on reading the facebook feeds of the passengers to ensure they were all “Liking” their trip or indeed looking at the management of the engines / staff to “add” KPI information for the Captain, even here I may be checking the web footprint of specific passengers to ensure those with the most influence were having the best experience… However, suppose the first navigator saw that there was unusual warm weather and his newsfeed had noted in previous warm weather years there was an increase of icebergs in the shipping channels.  Now he logs on and checks other ships in the area for increased iceberg activity, connects to a geo feed from a satellite and notes increased iceberg activity …

Now he speaks to his captain and indicates they should either change route, place a dual lookout up front and drop their speed by 10 knots …  Would information management enable that ?  I wonder in a corporate environment if we would have considered a tech savvy navigator asking the question about icebergs … knowing there was a geo feed from a satellite ?   How do we manage that, certainly there is significant benefit to management of NOT hitting the iceberg (and I propose there is benefit to the passengers also).    How would management have handled the slowing down of the ship, the diversion of the route ?  What if the captain had his benefits tied to the timely arrival of his ship in New York.  What if we had an auto-route plugged in based upon other ships schedules, shipping routes, fuel conservation etc … all valid in a normal operating mode which needs to be managed, but what happens when we need to cater for external influences beyond the managed domain ?

I do not ask this question to be provocative to information management, as there is a requirement and recognised discipline, I ask to ensure that the possibilities of the semantic web are not cut short before the people it is intended to benefit  (people who need to make decisions on trusted data sources) can gain the full benefit of this technology.

I agree we need Management of Information, I would expect though that this is always intelligent based upon the current domain, the question is … should the semantic web be restricted to a single domain …

I look forward to the papers being presented at Semantic Days 2012 …


, ,

3 Responses to Intelligent Information Management

  1. Robin Benjamins March 26, 2012 at 9:32 pm #

    I think the heart of what you are saying is that we all tend to design solutions, in whatever context or medium, with an expected outcome. This is great for that expected outcome but can easily fall short or fail for the unexpected or unanticipated situations. Specific to information management, by injecting or basing the solution on the principles and capabilities of the semantic web, we can meet the expect outcome and be in a much better position to deal with the unexpected and extract greater value from the unanticipated.

  2. Glen_B March 27, 2012 at 9:43 am #

    Yes that is the crux of the observation.

    However, I also think that a semantic web solution gives more power to the authorised users of the information to investigate relationships which are very difficult to consider.

    The capture of relationships via semantics enables a future proofing capability which does not require the input of the Solution Designer. Of course if there is a a benefit in “optimising” the workflow, then we can create an enhanced Solution, but this is now in the control of the user of the data rather than the creator of the solution.

  3. Robin Benjamins March 27, 2012 at 3:25 pm #

    Yes I agree and that is what I was hinting at with the “unanticipated”. There is big opportunity in that space.

Leave a Reply