June 2008


Quick update on my “Packed for the trip…” post: We had a surprise guest with us for the panel that I co-moderated at the NY Semantic Meetup that took place this week at the Linked Data Planet conference. Dan Connolly of the W3C wasn’t able to make the conference, so Tim Berners-Lee was able to sit in and participate in what turned out to be a lively discussion.

The room was larger and more formal than our usual New York Semantic Meetup settings – with eight people on the panel, yet everyone managed to fit at the table and get their opinions heard.

We carved the session into two focuses, along the lines of enterprise oriented use on the one hand, followed by the web focused utilization on the other. This is a difficult line to walk after drawing it, as we discovered during the enterprise portion that we were dancing on the other side. One interesting point raised was the possibility of adoption of the technologies being through the -as-a-service providers, where such providers would be the ones handling the heavy lifting, and the enterprises then being more easily able to leverage the capabilities – particularly for interoperability among geographically dispersed and disparate data of their own.

Later, when focused on the web side, we kept dancing back across to enterprise issues – around both impact on business models of opening up data, while at the same time creating different opportunities by changing the value formula from data to the services.

There seemed to be general agreement that it is ok to use whatever format works best for you, proprietary or compliant, when you’re dealing with inside facing data – but that outfacing should be compliant. At the same time, from a flexibility standpoint, RDF lets you go ahead withing having to know the whole of where you’re going. Still, coders would like to see the ability to do so more easily – without having to load libraries in order to do so. There also seemed a consensus that more services need to exist to take advantage of data being out there in order to encourage more enterprises to release theirs.

Rather than re-hash the details myself, I’ll point you to a good blow-by-blow of the panel. Two other write-ups are by panelist Savas Parastatidis from Microsoft, and by my co-moderator, Hank Williams. The latter contains some great commentary and insights that go beyond what was stated in the room.

As noted earlier, there is audible sentiment from within the Semantic Community that the necessary, underlying components for the Semantic Web to become a reality are in place. Discussion of the Vision is often as a destination – so to extend the trip metaphor, having the components in place means perhaps that we’ve packed for the journey.

Emergency Exit: Semantic Web (White on Green, Hi-Res)Image by semanticwebcompany via FlickrSome who are taking this journey have chosen vehicles, others are arguing about who is going to drive, who the passengers will be (never mind which one will have the back, middle seat). Once everyone is in and we’re pulling out of the driveway, you need to have an idea of the route that’ll be taken – with other questions in the back of our minds such as which stops we’ll make along the way, where we’ll fuel up, (what kind of fuel economy we’ll need), and where and in what setting we’ll actually settle when we get “there”.

So what are the routes (and vehicles, drivers…) to the Semantic Web and to enterprise adoption of semantic capabilities? These are some of the questions we’ll dig into this Tuesday night, when I’m one of the moderators on the panel at the Semantic Web NYC Meetup, at LinkedData Planet.

Additional details on the panel –
When/Where : June 17th, at the Roosevelt Hotel in New York
Panel Leader: Marco Neumann, New York Semantic Web Meetup
Moderators: Hank Williams, Founder/CEO, Kloudshare & Eric Hoffer, Second Integral
Panelists:

  • Sergey Chernyshev, CTO, Semantic Communities LLC
  • Dan Connolly, Research Scientist, W3
  • Christine Connors, Global Dir., Semantic Technology Solutions, Dow Jones
  • Taylor Cowan, Emerging Solutions Principal, Sabre Holdings, Travelocity
  • Richard Cyganiak, Reseacher, DERI and Project Leader D2RQ
  • Nic Fulton PhD, Chief Scientist, Reuters Media
  • Marc Hadfield, President and CTO, Alitora
  • Savas Parastastidis PhD, Architect, Technical Computing, Microsoft Research

Background: The basic concept of the Semantic Web is for information to be accessible and usable based on the meaning of that information as opposed to just matching text (vocabulary and synonyms) and hoping that the on the one hand, and versus complex data mappings and brittle proprietary approaches (such as directly coding together, or even APIs). In either case, the idea has been described as the standardization what can be expected in terms of the form of the information, and inclusion of the information’s context alongside the information itself.

We had quite an electrical display in the sky last night in Maplewood, NJ – along with some high winds and torrential downpour – for about fifteen minutes. Here are a few pictures I took around town:
(use the control buttons in the window below to page through the slides)

(click “view” if slide pane doesn’t appear above)

Update – Following the overnight drone of a generator, the sounds of chippers and chainsaws continue. Related stories:

SemTech08_468x60

Before posting thoughts about my panel/forum session at this year’s Semantic Technology Conference (SemTech08), I wanted to share a bit about the conference as a whole.

First, everyone seemed to walk away with a lot – regardless of their starting point. And the atmosphere provided for as much value from interaction in the hallway as from the sessions themselves. As (walking billboard for web2.0) Daniela Barbosa points out, there was a fourth-dimension of interaction going on behind (above/between…) the scenes (via Twitter and whatnot), also mentioned by Rachel Lovinger. I even heard that one attendee passed another some aspirin in the hallway based on a “tweet” about a headache.

The hidden gems

To capture some of the hidden gems of the week – the “PURLs of wisdom” so to speak (one of which was emphasis on the need for reliability and permanence of URIs – or persistent urls), I will, in the spirit of Eric Miller’s keynote, “reuse, repurpose an remix” or at least reference some of the writings of others about the week (but will most certainly fall short of MetaWeb Jamie Taylor’s gas powered blender metaphor). There were, after all, almost 150 sessions over the five days, and over 1000 attendees – so why not leverage some collective intelligence.

As others have noted (see links throughout), the week was quite a whirlwind – the only REST encountered likely being in references to enabling interoperability through RESTful architecture. From the experienced to the newbie alike, many had expectations of “SIOC (pronounced shock) and awe”, to riff off DERI’s Semantically Interlinked Online Communities project. (I had to use that, despite my understanding that our friends at Talis have intentions for that quip). But given the concurrent announcement(s) by Yahoo regarding semantic enablement of SearchMonkey, perhaps the experience to some was more like “SIOC the ‘Monkey”. And that may be just what the doctor ordered.

Much of the discussion in the semantic space has been about technology, standards, architecture – all necessary to solidify the resources upon which to build. And all consumable by those in the know. Mind you, this level discussion needs to continue – and as was stated numerous times during the week, all the pieces are “there” and they just need to be put together in ways for effective use – i.e. “just use what you need“, to quote the Freebase/Metaweb folks. And there’s not just one formula to do so; Carla Thompson noted Tom Tague (Thomson Reuters – Clearforest – Calais) referring to the subdomains of the space as Geekery Feifdoms (also described by Mark Johnson as “talking past each other”).

At the same time – while the technical level discussions were still present and engaging – there was a quite audible drumbeat this year, emphasizing the need simplify and to focus on real issues – in business terms – to take “this stuff” and put it to use to solve actual problems. At one point, Dave McComb of Semantic Arts (co-organizer of the conference) said, instead of the usual chicken-and-egg comparison (which I’ve used myself) – now we’re on a whole chicken farm. (By this, I assume Dave meant that it doesn’t really matter which comes first, we just need to focus now on producing eggs AND chickens – and running the farm).

Dave gave a great introduction (slides) to give some context and lay a landscape upon which everyone could layer their experiences of the week that was to follow – how we got where we are, what “is” semantics, comparing it to the relational model. Bruno Pinheiro describes his intro (as does Shamod Lacoul) along with that of W3C’s Ivan Herman, who gave a state-of-the-semantic-web intro as well. Some of his key advice came later, which Rachel Lovinger includes in her closing keynote summary. Particularly relevant to this post, she picks up on Jeff Pollock’s (Oracle) comment that “If you’re describing the value of something as being about semantic technology, you probably haven’t found the real business value yet” (see his presentation).

To put a sharper point on it, and to paint a picture of what this means from the standpoint of a business-side consumer, Tom Tague mentioned that one of his customers had said: “If you have to explain it, I don’t want it.” This underscores from the street-level that mainstream leveraging of these capabilities won’t be the result of promotion of the technologies but from the solving of problems for real people – which Nick Patience picks up in his first two bullet points.

Some are on the way

As Rachel also wrote about (linked above), Tom Ilube of Garlik is focused on how, in this online world, does one manage and protect their identity. By creating tools to enable their doing so and zeroing right in on a real problem for people, Garlik is bringing them onto the semantic web without revealing, nor needing to reveal, that fact. Leo Keller of Netbreeze showed tools for finding answers to questions which come from, but are not touted on the basis of, analyzing trends in unstructured text on the web. Tom Gruber’s talk emphasized that it is not “about” a particular application or technology, but is instead about the solution being woven into life via “the interface” (see Mark Johnson’s paragraph on Stealth Company). To me, this means tapping into the activities in which people are already engaged.

The bridge

The first step toward getting “there” is recognition of a divide (in this case, from technology to needs – solutions to problems). The past few years of this conference have been a major part of opening up and inviting business-needs-minded to immerse themselves in the possibilities. From the tone of the week, it seems this phase has begun. (I’ll write about industry verticals in a subsequent post). Now, the path from here involves reaching out from both the solution/tech side and the business side, with some assistance from the middle – to conceptualize the capabilities available and imagine them in the contexts of various business needs – and with consideration of the possible business models to support both delivery and consumption of the relevant solutions. I think Greg Boutin describes this all very well in his discussion of Powerset – a good read (in particular see his third section, entitled “Marketing semantics…”). He also places an accent on that drumbeat, quoting Tom Tague as saying “Clarify and focus on simple benefits”.

Tangent, or the other side

That would be a great place to wrap up, but at the risk of going off on a tangent, I feel the need to add one last point. I’d long ago come to see that the key value for and from semantics is in context and perspective. Seems pretty simple. But those are not just subject-focused matters. Within each individual (person, department, company…) there are multiple sets of motivation to be considered, and one of the dimensions of those motivations is time. I think Heidi Nordberg captures this when she writes of the need to differentiate between focusing on “long-term interoperability modeling goals” and short term focused addressing of “specific needs”. Perhaps there’s a way to address both together.

So we also need to go beyond targeting real problems, and and beyond simplifying the solution and message. The successful models will emerge from creating scenarios which enable the aligning of motivations – the sweet spot for opportunity.

Postscript: There was much talk at the conference of there being blue ocean, or uncontested market opportunities to be tapped by use of semantic capabilities (mind you, not for semantic capabilities themselves, but the use of them). Swim further off on this tangent and consider a purple ocean strategy that Henry Story shared over dinner with a group us the last evening – and may be applicable here.

Just as there were so many important and valuable perspectives shared at the conference, so are there scattered across the web. Apologies for not picking up on them all, but in addition to those linked above, here are some additional related posts:

http://developer-news.blogspot.com…
http://www.johnbreslin.com/blog…

http://www.jroller.com/Sandymountster…

Harold Carr shared some raw session notes, and for the conference’s own archive of wrap-ups, see: http://semantic-conference.com/news/#tripreport

(see also: related AXONomics post “Poetic Takeaway…“)

Zemanta Pixie

Clicky Web Analytics