bluereek

Watched an excellent presentation from Stefan Decker at last year’s International Semantic Web Conference (Karlsruhe).  Stefan Decker is the director of the DERI project in NUI, Galway.

The presentation addressed, inter alia, the lack of excitement about semantic web.  My key takeaway from his talk was that semantic web is about networking knowledge for the benefit of PEOPLE.  I think in getting caught up describing things in terms of computers being able to process information on sites thought use of various standards (not of interest to anyone except the technical community) etc we have lost the focus on the fact that what we really want is more collaboration, more productivity, more discoveries.  Semantic web initiatives have an end game of helping people make more use of the information.

Just looked at a nice 3 minute video from Binaryplex – promoting their beta product Hivemind.  I do not know what the underlying technologies are.  They do not even mention the worked ’semantics’ in their three minutes.  But what it is about is helping people to find expertise and, more importantly, experts in an organisation.  Recognising that people, for any number of reasons, may not update their profiles within an organisation the product seeks to address this gap, in the background.

Share on Facebook

New York Times announced last week that it will provide data marked up using RDF (Resource Definition Framework).

Why is this important?

This makes the data more useful.  You can now cross reference/ correlate the NY Times information with other information available on the web e.g. DBPedia (The RDF format of Wikipedia).  You can also develop applications which can access/ process/ interpret the NY Times data – because it is provided in RDF format.

Interesting development – and makes sense of the Linked Open Data initiative.  The NY Times is embracing RDF – to some extent it is giving away its data, but on the other hand its own data is far more valuable because it can easily be combined with other (RDF’d) data.

Quite a challenge to all organisations – especially those generating significant content – who are failing to have their data leveraged properly because it sits in its own silo.

Share on Facebook

Have just spent a couple of weeks on vacation – without broadband access at my fingertips.  Continued to monitor email and SMS – from my phone.  Probably online three times over the fortnight – had to make an effort.  Posted a few photos to facebook from the phone.

Real difference was not interacting with twitter and other social networks on a regular basis throughout the day.  Also – listened to the radio for news and read a few newspapers.

Just watched Kevin Kelly video/ presentation on future of the web.  KK (of Wired) sees the internet as one computer.  We use various devices to access the one computer.  ‘Things’ e.g. cars, clothes, devices which incorporate chips (e.g. RFID) are effectively part of the one computer.  And, indeed, we are in many respects sensors for this one computer – as more and more information ends up in the one computer.

This is enough to scare off a lot of people.  In the Q&A session KK fields a number of interesting questions, including what are the opt out options, is the one computer and the human race in conflict?  Interestingly seems that most people are happy to go along with what’s happening.  He has a great line ‘No personalisation without transparency’.  Effectively you have to open up, provide information about yourself, your business, whatever, if you want a personalised experience.

This morning read a posting about Gordon Bell – a Microsoft researcher who is attempting to record everything in his life digitally.

Interesting line in this from GB: ‘By using e-memory as a surrogate for meat-based memory, he argues, we free our minds to engage in more creativity, learning, and innovation (sort of like Getting Things Done without all those darn Post-its)’.

I have often thought that this is the case.  An example being that sometimes overprep for a meeting (reading all the material, anticipating the questions, etc) results in a less creative, open discussion.  Another example would be whether examinations are still bogged down in being largely tests of memory rather than tests of reasoning.

All of this relates closely to one of my own areas of primary interest – linked data and the semantic web.  Linked data requires entities to share more data – for the benefit of being able to correlate this with other shared data.  The semantic web aims to enable ‘intelligent’ processing of data by computers – ie the one computer referenced by KK.

I think KK is right.  The one computer is more and more a fact of life.  There are many benefits – and a number of threats.  While there are opt outs – and ways to escape e.g. go and live on a deserted island off the west coast of Ireland – inevitably the internet continues to be more pervasive (and invasive).

Looking forward to another few days of restricted broadband access.  And then back to life interacting with the one computer.

Share on Facebook

The recent Technology Forecast publication from PwC focused on semantic web and linked data.  Interesting series of articles – and I like the concept of dealing with ‘messy data’.  CEOs and other managers want to be able to merge internal ERP type data with external data.  Also reminds readers that the I in CIO is for information – and that CIOs need to take the lead on the generation and planning of relevant ontologies – given a clear understanding of their businesses and a working knowledge of ontologies.

On the same theme interesting piece by Linda Moulton – the line ‘enterprises must commit to having very smart people with enterprise expertise to build the ontology’  rings the same bell.

Linda Moulton believs that real progress in adoption of semantic web will be seen first within entertprises, later between enterprises and across the web more generally.  Seems to make a lot of sense and be the most likely scenario – however we may prefer the more holistic soltuion to emrge immediately.

Share on Facebook

Happiest days of my life.  Very fortunate to have attended Trinity College Dublin in the early 80’s.  Computer engineering – learning to program assembler for the Motorola 68000, learning fluid mechanics.  Friends, fun, social development, cricket, rugby, chess, campus in the city centre.  Developing wider interests.

But what opportunities there are now for everyone in terms of learning and collaboration!  No reason why an undergraduate in TCD would not be collaborating on a first year project with others located all over the world (e.g. students in other universities, people in industry).  Time and location no longer the boundaries they were in the past.

Fascinating book published on future of learning: The future of learning institutions in a digital age

The book sets out challenges for universities in terms of enabling and encouraging participatory learning.  These challenges also present fantastic opportunites for the go ahead insitutions.

Interesting to consider the content in terms of how learning and knowledge management take place in companies and organisations.  I referenced concerns of business leaders recently – much the same: missing the opportunity and failing to move to the more participatory and less hierarchical thinking.

Share on Facebook

No doubt about it – linked data seems to be where it ia and will be for some time.  CEOs traditionally have one eye on the external and one eye on the internal – relying on COOs, CFOs etc to drive the inside efficiently while they figure out the positioning, the alliances, the competitive advantage.  Most of the CIO work has been focused on providing the systems to enable the COO, CFO, etc run the organisation efficiently.

The CEO needs linked data.  She needs to be able to compare and contrast using external data, preferably in conjunction with internal data.  That’s all about linked data – in spite of the data being held in completely different structures. (Enter semantics, tags, RDF, etc).

Have spent a lot of time in the last 18 months working with companies figuring out their document management strategies for the first time – down to detailed taxonomies.  Now, in the context of linked data and semantic web am looking at ontologies.  Without doubt thinking through the ontology questions forces people to figure out processes, relationships and different types of structures.  Seems to me now that the linked data and the document management imperatives are not separate – rather they require a more holistic approach to the analysis and design.

I would expect linked data, content management, mashups, collaboration – all to become part of the same thinking and solutions.  Indeed wikis and blogs are part of this – as they challenge the traditional role of wordprocessing type documents – through providing greater interactivity/ collaboration.  Even developments such as wiki enabled web based training soltions (cf DERI and pergamon) are further examples of this merging of tools, environment and solutions.

So you may come at this from a collaboration and document management perspective or from a BI and data useability/ refereneabililty perspective.  But much of the thinking and problem solving are common, interlinked and overlap.

Share on Facebook

Linked data poses some interesting questions for us as individuals and in our organisations. Traditionally we have held that information is power – and therefore have guarded our information. Much of the time this has included guarding our data. To make linked data work we are looking to encourage much greater publications and sharing of data.

I was recently looking to complete some research on behalf of a client. It required me to review financial and operational information (including annual reports) for approx. 50 global companies. After some google work I ended up looking through a whole series of pdf files – available on the internet – and compiling my own analysis. And I have no easy way of updating this analysis.

However if this data were published in a different format – to facilitate its being read and analysed by computer applications – then this analysis would have been available to me instantaneoulsy. And updating the analysis would be trivial.

Back to safeguarding our data and/or information. If you have invented coca cola and have the secret formula then you want to keep this secret. Your concern is that you have a fantastic product and you want to maintain your competitive advantage.

If you are a soft drinks manufacturer you probably also know what percentage of nine year old males, living in any particular catchment area, drink your product more than once per month. This data is also of value to you – perhaps in terms of planning marketing campaigns, advertising initiatives, pricing plans. But perhaps you would be willing to share this information – in order to be able to correlate it with information that other groups may have about habits of youth in a particular neighbourhood.

Is there any real point in people the world over wasting time effectively completing the same analyses – in private companies, government bodies, voluntary organisations? Is it not an incredible waste of time? In fact, in an era when we have finally begun to concern ourselves with energy waste can we not recognise that this duplication (many times over) of effort is a major societal waste of time?

We have driven some of the sharing of data through initiatives re disclosure – to protect shareholders, citizens, etc. However we have not yet got to the point where we are rewarding companies and organisations for making more of their data available in useable formats. There are major potential savings and benefits if we can change the mindset.

I do not underestimate the challenge faced. Our business training and experience has been to develop and maintain competitive advantage through having greater insights, knowledge, etc. What the semantic web is suggesting is that to succeed we should be much more generous with our own data – in order to gain access to far richer and deeper data, while at the same time serving the common good. We now need to see real sample models for people and companies adapting and succeeding with this approach.

Share on Facebook

I thought 3 made a brilliant job of promoting the Irish Open – using networks including facebook.  However I think the opportunity to play a virtual round at Bethpage for the US open beats this.

This is an example of the web adding significantly to the the user experience ie to the TV golf spectator or the golf newspaper article reader.  Anyone who takes the time to play 18 virtual holes at Bethpage (not having playing the real course) will have a greater understanding of the challenge to be faced by Tiger Woods, Padraig Harrington, etc.

And then there’s the traffic, the advertising potential, etc.

Good luck to all of you who try.  I hope you have more skill at virtual golf than this blogger.

Share on Facebook

Spent the morning at a workshop run by DERI (Digital Enterprise Research Institute) at Enterprise Ireland.  If we spent more time focusing on what we can achieve through the likes of DERI and the Irish Learning Alliance (ILA) we might begin to dig ourselves out of our current difficulties.

Excellent presentations by Johnny Parkes, Bill McDaniel, Liam Moran and Mark Leyden.

Web 3.0 – in terms of getting at the data across the web – has great potential.  Poses interesting challenges/ questions for organisations traditionally obsessed with confindentiality of their data.  However for those who understand and resolve the connundrum (sharing their data) web 3.0 offers the potential of much greater insights and decision making.

Share on Facebook

I had commented some time ago on the differences/ advances between web 1.0, web 2.0 and web 3.0.

Earlier this month Greg Boutin published an excellent series of three postings in which he has looked at web 3,0, semantinc web and data linking in consdierable detail – in an attempt to ensure that we have a common understanding (ie get the semantics right).  His postings alos cross reference to some of his own previous postings and an excellent TED talk by Kevin Kelly.

Kelly talks about the interent becoming the ‘One’ computer – and therefore topics such as ubiquity, transparency, personalisation and globalisation.  It is a fascinating look into the future – and challenges us all to think about how we would interact with this large ‘organism’ or system.

Boutin is also very much up for the challenge and the opportunity.  However he is relatively cautious in his assessment of the situation in his third posting.  While not doubting Berners Lee, he does seem to suggest that the hype may be a little ahead of the reality – as evidenced by the lack of commercial applications exploiting linked data (there being a shortage of linked data).

Share on Facebook