How do you create self organizing clouds?

Home Forums Technology How do you create self organizing clouds?

This topic contains 11 replies, has 7 voices, and was last updated by  Steve Ressler 5 years, 9 months ago.

  • Author
    Posts
  • #153904

    Jeff Chao
    Participant

    I’ve noticed that one of the draw-backs of this whole social media thing (or the Internet in general) is the proliferation of content/data without much organization besides search and keywords. There’s no taxonomy or global context. Of course, there’s digg, mashable, stumbleupon, etc. but they’re really just hit lists based on popularity or traffic volume. After some period of time (usually very short), content just falls off the radar and effectively becomes irrelevant regardless of merit/value.

    When it comes to something like government transparency and what sites like Govloop are trying to do, how do you give valuable content longevity? How do you establish “authority for content” via a cloud that yields a solid foundation of knowledge instead of the constant proliferation of more and more content?

    There must be a way.

  • #153926

    Steve Ressler
    Keymaster

    Would love ideas.

    We’ve been working on taxonomy structure with tags and meta-tags but that’s not perfect. Our other approach was condensing best of govloop into items like guides – https://www.govloop.com/page/15-commandments-for-government-agencies-on-twitter

    One site that does a pretty solid job is Quora which has a variety of formats to push best answers to questions

  • #153924

    Jeff Chao
    Participant

    For content organization, there has been some interesting nerdy work going on in the form of the “semanic web” for searches and such but knowledge organization is a tougher nut to crack because of context. I’ve been thinking of an idea to organize knowledge by process streams but not sure how that would work in a UI. In my head, it looks kind of like a cross between a scrabble board and timetube.

  • #153922

    David Phillips
    Participant

    Jeff, Very interesting questions but I would suggest that the technology exist but not the policies. Storage and archiving based on priority has existed for several years, However it has centered around how current the data is and how long you want to keep it. Currency may indicate some degree of relevancy but not your “authority for content” that you would demand. The trick is determining an algorythm that defines relevancy. Currency might be an element. Intelligence agencies have been struggling with this since Sun Zu. It is particularly pertinent now because we have so much data…

  • #153920

    Jeff Chao
    Participant

    Hey David–I agree that defining relevancy is the key. From a technology perspective, I suppose the search engine or whatever functional implementation would need to essentially “understand” the content which seems like a unlikely prospect (although I’m sure IBM Watson fans would disagree). But I’m thinking there has to be a way to get social media to do it where the intelligence still resides with the user but some mechanism exists that enables the content to be self organizing.

    Maybe it’s a heisenbergian problem.

  • #153918

    David Phillips
    Participant

    Hi Jeff, Maybe but only if you are dealing with quantum mechanics… I was trying to be funny but it begs the point that relevancy is tied to the subject matter and the individual’s interest. If the subject was engineering a civil engineer might find information on fluid mechanics more relevant that quantum mechanics. I agree that the answer lies with tools that allow the user to dynamically request, analize, and store in a fast and iterative manner. Just a thought….

  • #153916

    Steve Ressler
    Keymaster

    Yeah we’ve been looking into Open Calais which does some pretty interesting work on semantic web and meta-tagging. Pretty cool project – you can run your data through it and it helps organize and tag it

  • #153914

    Owen Ambur
    Participant

    Content should be organized around (strategically “aligned” with) the longer-term goals and near-term objectives that agencies are charged by law to accomplish. In addition, as directed by the eGov Act, it should also be discoverable based upon the stakeholder groups it supports.

    Subsections 202(b)(4) & (5) and 207(d) of the eGov Act require agencies to:

    • Work together to link their performance goals to key groups, including citizens, businesses, and other governments, as well as internal Federal Government operations; and
    • Adopt open standards (e.g., StratML) enabling the organization and categorization of Government information in a way that is searchable electronically and interoperably across agencies.

    Section 10 of the GPRA Modernization Act (GRPAMA) requires agencies to publish their strategic and performance plans and reports in machine-readable format, like the Strategy Markup Language (StratML) standard (ANSI/AIIM 21:2009 & 22:2011).

    The purposes of the StratML standard are outlined at http://xml.gov/stratml/index.htm#DefinitionPurposes In addition to complying with section 10 of GPRAMA, use of the StratML standard — particularly the <Stakeholder> element — will enable agencies to comply with the requirement of the eGov Act to link their performance objectives to key groups.

    All of the existing “social” media services are immature for business-quality usage. However, by marrying them with an open data standard like StratML, we can create a powerful web of intentions, stakeholders, and results.

    All that agencies need to do is comply with the law, i.e., GPRAMA and the eGov Act.

  • #153912

    John Able
    Participant

    The answer may be a mashup of Google and Match.com — a powerful search engine that uses an algorith based on a comprehensive understanding of our individual content preferences.

    Google, of course, does this now by remembering all our previous searches (and the content of our gmail and Google Docs), but Google has a profit-motive, so it selects sites that are most likely to encourage click-throughs to advertisers. This kind of ulterior motive gets annoying and no longer really locates the content we’d most like to find. In fact, Google now knows so much about us that it’s getting creepy. But if our preferences profile was securely separated from Google’s prying eyes, the incredible power of a Google search married to our profile of perfect content match preferences might actually find what we’re looking for.

  • #153910

    Jeff Chao
    Participant

    xml.gov fell off my radar so thanks for the reminder about the work being done here. I’d be interested in seeing Mark Logic’s StratML search service prototype or other applications that have leveraged content organized in this way.

  • #153908

    Bill Brantley
    Participant

    @GovLoop – Have you looked at OpenCalais – http://www.opencalais.com/

  • #153906

    Karen Jane Smith
    Participant

    You have raised an amazingly complex question. I guess it all comes down to the fact that people think differently: what matters to me might not matter to you, the language/terms I use, might not be terms that you use, the importance I see in information, and the links I make in it, might not be the same for you. So where to from here? I think that we must fall back on old fashioned cataloguing by general subject matter and author. It is up to the reader to make the links – but I do think we need to provide assistance to people on how to search and scan information for their particular purposes.

You must be logged in to reply to this topic.