Contribute > ideas > #pdxtags

Category: 
Community

Portland is a quirky collection of distinct neighborhoods cultivated in the best of books, food carts, microbrews, coffee, music, and outdoor recreation. Portland is a friendly and civic-minded community, striving for civic creativity, civic awareness and involvement, in addition to economic and environmental sustainability and active participation in making Portland a better place to live and work.

What if we attempt to harness all of the above favorable attributes by pulling together so many disparate information sources and services into a single community-driven interface or dashboard? A dashboard that could be configured to individual users tastes, and be optimized for the type of user (i.e., resident, business owner, government, etc).

What's missing is some way to tie everything together, to organize the information. An ideal solution would make use of existing methods of those already actively participating and embrace hyper local content down to the neighborhood block level.

What we already have (please suggest more):

  — Geo location based Tweets (point or neighborhood based)
  — Collection of other messaging platforms (forums, smartphones)
  — Media outlets (newspapers, blogs, hyper local sites, etc)
  — Commercial Aggregators (Google Local, Yelp, Craigslist, etc)
  — CivicApps, PortlandMaps, PublicAlerts, PortlandOnline, PDX Reporter (iPhone app)

Proposed solution (again, please refine/comment):

Create a community driven database of (hash) tags that can be added to content and messaging. Allow tags to be tied to user created geographic overlays (boundaries on a map). And, most importantly, encourage whole communities to use the tags whenever creating content for those they hope to target locally or extend their reach to others.

To achieve this, the following types of tools would need to be developed to facilitate/drive adoption:

  — Tools to harvest tagged content from many information sources (Google, Twitter, Facebook, Craigslist, Flickr, local blogs, newspapers, online communities, etc).
  — Tools to identify and accommodate the inclusion of existing tags, make it so tools could be embedded in local websites (via API).
  — Tools to bind tags to particular geographies.
  — Tools to notify property and business owners for a given geography tag, to promote a tag for use (advertise an event, organize a neighborhood watch, etc).
  — Tools to demonstrate and drive client-side adoption; consumer-side receipt of producer-generated tags.

Critical Success Factors:

  — Will require broad participation and adoption by the community; concepts like Twitter and Facebook work because everyone participates, not because they're hard or breakthrough technologies.
  — Must work with existing web ecosystem and not duplicate efforts or compete in anyway with private sector.
  — Work proactively with developers to participate and empower them in creating an ecosystem with these tools.
  — Moderation performed by citizens to combat spam, abuse, and profanity.

Discussion
maxogden :: May 7, 2010 - 3:59pm

this is more or less what I am shooting for with www.pdxapi.com. the ideas you have listed are all great examples of client implementations that could be built on top of a nice centralized web data store. i'd love to tackle these problems in an in-person discussion with all interested parties. perhaps at the upcoming civicapps meetup, or sooner?

rnixon :: May 10, 2010 - 1:02pm

Thanks Max. I've arranged for a meeting on Thursday, May 27th at 6pm for an open discussion of ideas and datasets with data providers. Sort of a precursor event to the upcoming CiviCode Day at Open Source Bridge. Please see http://civicideas.eventbrite.com for more information.

aeschright :: May 18, 2010 - 1:05pm

Two other local projects that might overlap with this: Portland Wiki (http://portlandwiki.org/PortlandWiki) and Tagal.us (http://tagal.us/). I'm skeptical about the "broad community adoption" aspect of this. Cataloging things is entertaining to only a certain section of the population--the wider majority generally use hashtags because it's something fun (see #doingitwrong or #electricblanket). For a system like this to succeed, it will have to allow for a smaller number of users doing most of the work.

artwells :: May 20, 2010 - 10:57pm

Feels like this will require a big firehose of information and powerful filters, or a lot of volunteer editing in the end. It seems like it would be tricky to keep the earlier stages from sputtering out, getting buried in noise, or going dark. It could start out as a simple definition list (e.g. http://wthashtag.com/Main_Page ) with some commonly, currently used, tags (e.g. #pdxtst ) and expanded cautiously to raise awareness of the common list. After that it could use some key players to bless suggestion or a reputation network (like Janrain/Aboutus's Jyte only with a purpose?). Once things get going more automated systems could be structured around discovering live trends realtime filter geo-coded postings from flickr, buzz, twitter (also using those people who are part of existing Portland lists). It would always be messy, I imagine, but it would be very, very useful.

Snacky :: July 5, 2010 - 5:39pm

Sounds interesting, but the description is somewhat abstract for me to fully understand. An example would be cool. E.g., what would a set of inputs and outputs look like? What's a Use Case that it'd support?

hilljohng :: October 21, 2010 - 7:08am

I'd agree with those who want to see an example of how it might be used by the general public, and that only a select few in the metro area are going to devote much time to tagging content. That said, the word firehose caught my eye. Reminded me of Nozzl Media and their efforts to create tools to scrape data into a usable stream. Perhaps if one harnessed that power and combined it with Open Calais to auto-tag the content, you'd have something at least somewhat usable before you even asked the user community to help out. I've always thought it would be cool if someone built a news web site where the readers could tag content, though that could also quickly get out of hand. One would probably have to create a filtering system so users could control the stream, ie, check this box to only get content tagged by those you "follow" and/or who "follow" you, or check this box to show me content tagged by the media outlet's staff, etc. Is anyone going to take this idea and other Civic Apps ideas to Bar Camp this weekend for discussion? It would be awesome to hear what people had to say.