Loading ...
Sorry, an error occurred while loading the content.

Re: [TaxoCoP] The taxonomy honeymoon is over...

Expand Messages
  • Madi Weland Solomon
    Hi Stephanie. I ve kept my 15 year relationship with structured data fresh by moving to ontologies, big data, and data analytics.  Last year I found a new
    Message 1 of 40 , May 16, 2012
    • 0 Attachment
      Hi Stephanie.

      I've kept my 15 year relationship with structured data fresh by moving to ontologies, big data, and data analytics.  Last year I found a new interest in privacy and public policy and completed two certification courses from the International Association for Privacy Professionals (IAPP).  The topic is so new it makes me feel young.  

      I suppose my taxonomy honeymoon is over too,  but the commitment remains strong. I've stopped expecting all my needs to be met by any one thing and have learned to augment my working life with side interests. Intellectual Property, schema.org, OER, the ukulele.   

      There aren't many of you Stephanie!  On behalf of your taxonomy, I'm pleading for you to stay. Let's take the relationship to a new level!

      Madi Weland Solomon
      Dir. Content Standards
      Pearson plc

      From: stephaniealemieux <lemieux.stephanie@...>
      To: TaxoCoP@yahoogroups.com
      Sent: Wednesday, May 16, 2012 2:23 PM
      Subject: [TaxoCoP] The taxonomy honeymoon is over...

      Hi everyone,

      I guess what they say about marriage is kind of true about careers - cause I've got the 7-year itch when it comes to taxonomy! It's not that I've fallen out of love with taxonomy, but after a number of years doing the same projects over again, it's starting to feel stale.

      Does anyone have any suggestions in terms of related domains to branch out into that are new & exciting?
      - business intelligence?
      - design?
      - data security?

      Need to find a way to spice things up, lest I get stuck too far in the doldrums.

      Thanks for any suggestions!

    • Patrick Lambe
      Brilliant contribution Marilyn, many thanks! From the presentation it s clear that Google is playing in taxonomy territory, without ever using the word. In my
      Message 40 of 40 , Jul 16 9:05 PM
      • 0 Attachment
        Brilliant contribution Marilyn, many thanks!

        From the presentation it's clear that Google is playing in taxonomy territory, without ever using the word. In my book I define taxonomy as having three functions, each of which underpins one of the "design patterns" that Shawn speaks about:

        Taxonomies have a semantic function by providing a controlled vocabulary - in Google these are the "Discovery patterns" allowing them to match terms to entities, extract and disambiguate entities and reconcile terms to a single ID

        Taxonomies have a classification function to group related things together - in Google these are "Interaction patterns" that allow uses to filter (explicit use of facet language here) and navigate from one entity to other related entities via recommendations

        Taxonomies have a mapping function to give an overview of a knowledge domain - in Google these are "Presentation patterns" for displaying concepts, particularly for providing mashups (which get very interesting with facets) and visualizations such as network maps, trees, timelines, graphs etc.


        Patrick Lambe
        Tel: +65 62210383

        On Jul 14, 2012, at 1:47 PM, Marilyn wrote:


        Interestingly, Shawn spoke at Google's big conference recently, and the video has been posted. While he doesn't use the words taxonomy or schema initially, I think you'll see that Freebase is in fact the semantic basis for Google's Knowledge Graph.

        Here's the link to his presentation, entitled Google I/O 2012 - Knowledge-Based Application Design Patterns:

        Best, Marilyn

        --- In TaxoCoP@yahoogroups.com, "Marilyn" <langfeldesigns@...> wrote:
        > Hi, I'm new both to taxonomy as an interest and here on the list...
        > I believe that Freebase, the product of Metaweb, the company that was purchased by Google in 2010, is very likely the basis for the Knowledge Graph. Freebase is much smaller than the Knowledge Graph though, and is uses CC licensing, so Google is most likely keeping most of its data separate from Freebase. The structure of Freebase was patented by Metaweb, so Google could purchase Metaweb and its team, let them continue to develop Freebase with its user community, and at the same time make its own internal, proprietary version with its huge datastores. This is my guess to the answer to the question posed.
        > Freebase page on Wikipedia: http://en.wikipedia.org/wiki/Freebase
        > According to the Wikipedia article cited above:
        > "On March 3, 2007, Metaweb publicly announced Freebase, described by the company as "an open shared database of the world's knowledge," and "a massive, collaboratively-edited database of cross-linked data." Often understood as database model using Wikipedia-turned-database or entity-relationship model, Freebase provides an interface that allows non-programmers to fill in structured, or 'meta-data', of general information, and to categorize or connect data items in meaningful, or 'semantic' ways.
        > "Described by Tim O'Reilly upon their launch, "Freebase is the bridge between the bottom up vision of Web 2.0 collective intelligence and the more structured world of the semantic web.[3]
        > [3] "Freebase Will Prove Addictive". O'Reilly Radar. 2007-03-08. Retrieved 2007-03-09.
        > "Freebase contains data harvested from sources such as Wikipedia, ChefMoz, NNDB, and MusicBrainz, as well as individually contributed data from its users. The structured data is licensed under the Creative Commons Attribution License, [3] and a JSON based HTTP API is provided to programmers for developing applications on any platform to utilize the Freebase data. The source code for the Metaweb application itself is proprietary.
        > "Freebase runs on a database infrastructure created in-house by Metaweb that utilizes a graph model. This means that instead of using tables and keys to define data structures, Freebase defines its data structure as a set of nodes and a set of links that establish relationships between the nodes. Because its data structure is non-hierarchical, Freebase can model much more complex relationships between individual elements than a conventional database, and is open for users to enter new objects and relationships into the underlying graph. Queries to the database are made in "Metaweb Query Language" (MQL)."
        > and
        > "[U]nlike the W3C approach to the semantic web, which starts with controlled ontologies, Metaweb adopts a folksonomy approach, in which people can add new categories (much like tags), in a messy sprawl of potentially overlapping assertions.[3]"
        > Shawn Simister is the Google product manager for Freebase. Freebase now has a Google+ page as well as the mailing list, so it is possible to contact him directly. I'm not sure he can answer everything we'd like to know (proprietary Google info), but you never know until you ask.
        > Best,
        > Marilyn Langfeld
        > --- In TaxoCoP@yahoogroups.com, Patrick Lambe <plambe@> wrote:
        > >
        > > Hi Claude
        > >
        > > I think formally any construct of entities and relationships of this
        > > kind of complexity and scale is an ontology, however it is arrived at.
        > > The language used in the blog post suggests this. However, we are
        > > admittedly speculating here - maybe we should ask Google themselves?
        > > Anyone have inroads or contacts there?
        > >
        > > P
        > >
        > > Patrick Lambe
        > > Partner
        > > Tel: +65 62210383
        > >
        > >
        > >
        > >
        > >
        > > website: www.straitsknowledge.com
        > > weblog: www.greenchameleon.com
        > > book: www.organisingknowledge.com
        > >
        > > Have you seen our KM Planning Toolkit?
        > >
        > > http://store.straitsknowledge.com
        > >
        > >
        > >
        > >
        > >
        > >
        > >
        > >
        > > On May 22, 2012, at 12:00 AM, Claude wrote:
        > >
        > > > --- In TaxoCoP@yahoogroups.com, Patrick Lambe <plambe@> wrote:
        > > > > I think the Knowledge Graph is probably working off an ontology in
        > > > the
        > > > > background - I suspect "knowledge graph" is the product of intense
        > > > > marcoms meetings about how they could avoid the ontology-taxonomy
        > > > words.
        > > >
        > > > With all due respect, and unless Patrick has internal information
        > > > about how GKG is working, I would suspect the opposite: that it is
        > > > based on associations constructed over time from user queries,
        > > > without an underlying ontology.
        > > >
        > > > In which case, this work would be more directly inspited by Henry
        > > > Lieberman's "Common Sense" work at the MIT Media Lab, and all the
        > > > resulting "concept net" body of knowledge:
        > > > http://csc.media.mit.edu/
        > > >
        > > > This is the basis for the Media Lab spinoff Luminoso, which I
        > > > mentioned here before. The big advantage of a capability that is NOT
        > > > based on a taxonomy or ontology is that you can discover things you
        > > > didn't know in advance might be there. The big drawback is that you
        > > > can't start small: you need a huge corpus of information to build
        > > > the knowledge graph in the first place. That's where Google has an
        > > > advantage: if it perform some analysis of user queries (and which
        > > > result page they clicked on after seeing the results of a query),
        > > > they get billions of semantic connections every day.
        > > >
        > > >
        > >

      Your message has been successfully submitted and would be delivered to recipients shortly.