WiserEarth Group Support

Help and advice for WiserEarth groups, discussions about groups and examples of successful groups.

GROUP DETAILS

Created: Feb 20, 2008

Updated: Nov 19, 2009

Membership: Open

Public

 
Created: May 04, 2009
Updated: May 26, 2009
Viewed: 416 times
Page Status: active
  •  
Not Yet Rated

Sharing Engine

Edit this Page
Bullet Point Summary:

The Sharing Engine...

• Is a TN sub-project delivering a website and set of web tools that will assist TN participants to publish and find information. Starting with news, events and moving to projects, issues, skills and other relevant items of interest.

• Will harvest information from TN participant websites and aggregate these information feeds.

• Will make these information feeds available for republishing by TN participant websites and any one else who might wish to take the feeds.

• Will encourage TN participants to describe their content in machine readable ways to enable their content to be included in "deep" searches - where multiple and specific criteria can be used to build detailed searches

• Will provide a search interface that will enable multiple and specific criteria to be used. This is very different from the one line text field you find in Google. Criteria could be based on date, location, language and much more specific details depending on the object being searched for.

Using the Sharing Engine:

Different parties will have different ways to interact with the Sharing Engine...

• Publishers will register their original content to the Sharing Engine. This will be done in a number of ways: registering individual pieces of content; referring to a "feed" of their content. These feed could be complex queries that could filter information captured by the Sharing Engine or shared by the publisher.

• Community websites will be able to take feeds from the Sharing Engine and have these filtered too. For example the Sharing Engine will hold all events in the TN. However, only those events in 50 miles of the community are of interest. Hence, this is a filter that will be able to be implemented. And this filter can only be implemented if the original event was published with the exact location coordinates in the first place.

Implementing the Sharing Engine:

• Initially, we will build a reference implentation using Drupal as this CMS is advanced in the area of RSS aggregation and syndication. RSS will form the basis for this project but we will then move on to Atom and RDF - both of which are being built into Drupal.

• The aim here is to produce a working system very quickly and cost effectively. So, wherever possible we will reuse code and modules already in existence. We will also contribute any work back to the Drupal community for wider reuse. This is good Internet community relations.

• Much of the work here will need to be done by the TN support group that will market, promote, educate and support TN websites implementing the tools that will enable them to interface with the Sharing Engine. Thankfully, because we are using standards, such as RSS, there already exist many implementations and plugins for all blogging systems and more advanced content management systems.

[? • Daniel Harris will be attending the Drupal conference in Paris in September and will put forward the Sharing Engine concept for a presentation on approval by participants and the board. This would effectively give TN exposure to thousands of developers in the Drupal community which is nice. ?]

Scope:

This summary provides tactics and stragey for assisting TN members to communicate with each other - to share information and make it easier to find relevant information about others.

Overview:

Currently we have a real problem with information overload. There is too much stuff to sort through to find the information that we really want. But the problem of too much information will not go away. In fact the problem is going to get worse.

Looked at from another angle, the problem is more that we don't yet have the most effective tools to manage, filter and search all this information. However, recent technologies in the field of information management will help us out of the quagmire and allow us to specify more accurately what information we want to receive.

One of the aims of the whole web project, that should be highlighted, is to assist in promoting the effectiveness of the whole Transition project; to showcase success stories using qualifiable metrics in order to show usefulness and effectiveness to tackle current problems and so access more private and government funding. More generically we need to build scenarios and usecases for different user types interacting with the TNL website and the wider web project.

The web project should enable Transition users to share and find useful information that they can apply to their own neighbourhood and/or project. The type of information that should be shared are: projects, events, news and issues.

TNL's Role:

If one views all the TN related websites as virtical then TNL should seek to promote the horizontal. In a sense foster communication between all the different sources of information "out there".

Sharing Engine:

From centralised to fully distributed (de-centralised).

Initially, the Sharing Engine will be built using a centralised model. This will enable more rapid prototype building and testing in the field as the complexities will be less than a fully distributed (de-centralised) sharing system.

However, a fully distributed (de-centralised) information sharing system is the ultimate goal. Where all Transition websites around the world have the ability to share information on projects, events, news and issues no matter what platform/framework/software their website is running on.

Sharing information becomes a whole lot easier when we make our information "semantic" and bring meaning to our information in a machine readable way. When the information is machine readable it enables real people to use filtering and search tools to assist in finding very specific content.

Searching:

Bringing structure to the information we publish enables us to make more intellegent searches.

Governance for Schemas and Standards:

For the Sharing Engine to work participating TN websites will need to adopt standard practices and ways on publishing certain metadata information. This is not as scary or as authoritarian as it sounds. For example all websites currently adhere to HTML web standards otherwise they will not be viewable by web browsers. So, in order for their data to be fed into the Sharing Engine the metadata must be in a specific agreed format.

In order to kick off the Sharing Engine trial a set of initial Schemas and Standards shall be created by the core group (with help from the wider community) for TN participating websites to adopt. However, it is clear that a formal process will need to be introduced that will enable collaborative modifying of these Schemas and Standards on a frequent basis. This will enable more information terms to be introduced or for current information terms to be modified. Keep the Schemas and Standards fluid will enable the TN to be kept up to date.

Note that constantly updating the Schemas and Standards will not cause expected confusion between the Sharing Engine and websites that are not using the most current Schemas and Standards because all modifications of the Schemas and Standards will be tagged with a version number. All versions of the Schemas and Standards will be compatible with each other with use of version conversion tables.

Effort must be put into marketing the Sharing Engine concept to existing TN participant websites. Adopting standards and specific publishing practices never happens by itself in the early stages of a service. 

Graceful Degrading:

A term that is used in computer interface design is Graceful Degrading meaning that if a device or system does not have the capability or power to carry out an action or display an object then it will falls back to a less flashy more basic interface that will still provide the functionality but perhaps be less slick. This concept is extremely important in the case of the Sharing Engine where we have to deal with so many different and dispersed website systems with wildly varying capabilities. So, we need to specify and document the ideal scenario for sharing information with an ideal website system and then, for each system TN participants are using, define how the functionality will be replicated taking into account the quirks, deficiencies or advanced technologies of that system.

In the first instance this will mean the use of RSS feeds. RSS feeds are pretty standard on most content management systems.

Interfacing to the Sharing Engine:

All TN participant websites will need to register on the Sharing Engine. But there will be a number of varying ways for the Sharing Engine to obtain the metadata from the TN websites. More detail about how systems will interface to the Sharing Engine:

• HTML - In the case of a website being crafted purely from HTML and with no database capabilities, it should still be possible to register the content on the Sharing Engine. In the worst case this will be done by filling out a form on the Sharing Engine to act as a signpost to "point back" to the content from the Sharing Engine. Next best will be a standard XML piece of code inserted into the HTML content that when harvested by the Sharing Engine will act as the signpost information.
• Ning
• WiserEarth. (See WiserEarth API Developers).
• Joomla
• Drupal
• Plone
• Wordpress

We could provide plugins for these frameworks and online platforms. These plugins would enable schemas to be shared and received.

We are sharing information about:

• Projects
• Events
• News
• Journeys - to share travel
• Skills - to share and trade
• Issues - problems discovered - could this be used for or trialed with a technical issue queue ticket system in the first instance?
• Solutions
• Pictures
• Video
• Products and Services offered and wanted

Share a link:

A popular tool amongst social networking sites is to enable their members to share a link to their personal feed or blog or profile. In the same way we could produce (or adopt) web tools that would enable anyone to "share a link" to an event, news item, project or any other defined TN object. There would also be the option to include a rating or attending tag.

Reuse:

In all cases the aim is not to reinvent the wheel but to reuse tools, applications, standards and modules already in existence. 

Geotagging:

Where possible all shared object metadata should include information about where the content is related too.

Filter positioning:

A very flexible system for obtaining information from the Sharing Engine will be provided. TN participant websites will be able to take full feeds and then process (filter) these feeds at their own site. Or people will be able to create their own filters on the Sharing Engine via a simple but powerful graphical query builder (much like those found on Apple Mac systems).

Hosting:

There are cases for TNL to keep this work managed in-house and for contracting management out. There could be issues about a third party holding private data about TN members. Surely TN members would feel safer to have their details reside with TNL and not a third party. Also there is the issue of authenticity of content. For example it has been know for Google to censor their content. If the Sharing Engine was hosted by a third party could there be potential conflicts of interest?

Phased Rollout:

With such a potentially far reaching project it's important to get a working system up and running very quickly. Hence, the recommendation is not to spend months specifying the Sharing Engine in theoretical detail but to rollout a series of working milestones. There are many projects and tools available for us to build this system. But we also need to have TN participants working with us as we create the Sharing Engine.

This will be a staged process started in news, then events, leading to projects and other objects more specific to the TN community. We'll also be in constant discussion with other organisations such as CAT and the Soil Association.

Implementation:

The Sharing Engine should be a built in such a way that it can be integrated into existing websites and also distributed as a standalone install.

Framework:

In order to rapidly demonstrate the potential of the Sharing Engine I recommend that we implement the first iterations in Drupal. For example Drupal has news aggregation practically built in. Take a look at the Transition Town New Zealand site for example:

http://www.transitiontowns.org.nz/aggregator/sources

Milestones and Estimated Costs:

1.
2.
3.

Future Development:

The service described here by the Sharing Engine is highly centralised. This is necessary in order to kick off the project and make it easy for us to manage the roll-out process. It will also enable us to make swift changes to interfaces and functionality. However, centralised systems are rarely desirable as systems mature and technologies evolve. So, the Sharing Engine should be seen as a reference implementation for a node of a fully distributed information sharing system. The advantages of fully distributed systems are resilience and redundancy and information authenticity and security.


Comments (1 - 10 of 10)

Login to Post a Comment.
Sm_avatar

 

 

 

Here is example of one combined custom search box. Not sure if it will come in here, if not, here it is on ning site. It combines 100s of sites.

http://transitioncalifornia.ning.com/page/search-1

 

 

Use for research & networking


Onsite searches @ WISEREARTH.ORG & IDEALIST. ORG
get more refined geographical results.
WiserEarth Proxity Map
generates lists within any size region. Highly recommended.




Return to Transition California Home Page

 

Here is my feed and search experimental pages, do not miss tool tab. I also setup a Google Reader tool and Google Marker tool in my browser to collect the feeds and links, then they are automatically added to a custom search or to my reader, where i can then add it to a pre-existing custom feed that can be generated from any site using the widgetry. i have have tried the best i can to seed some sites with these tools.

 

Yes, I personally picked links and feeds that I believed were Transition friendly, also, i would judge whether to include sites that were linked within linked site for searches. All ning transition sites include the sites that are listed within it.

 

http://transitionearth.ning.com/

 

more feeder experiments here using the grou.ps platform

 

http://grou.ps/hopedance

Sm_avatar

Hi Phil,

 

Thanks for the flag. Have changed the wording to your second suggestion.

 

Keep it coming.

 

Cheers Daniel

Sm_avatar
Hi Daniel, sorry if misunderstanding, but

"Will make these information feeds available for republishing by TN participant websites."
looks a little odd to me, in the context of

  • 'sharing',
  • "...we do not need to be restricted to TN website information but could also ingest feeds from other sources and make those too available for syndication"


Isn't 'sharing' normally seen as a two way process? Is there some reason why you don't (it doesn't) say

"Will make these information feeds available for republishing." or
"Will make these information feeds available for republishing by TN participant websites and any one else who might wish to take the feeds." ?

Sm_avatar
Yes, I like this SHARE LINK WITH TRANSITION NETWORK idea. One could share a news item or an event or a project or an issue. So, this means that it doesn't have to just be the news, event or project "owner" that can post a "signpost" feed/link item but also third parties can share links too. And there'll probably be ratings in there too at some point. Good idea. Thanks.

Hey, I hope you can take this the right way, but how about putting the good ideas at the top of your post? Like a news article? Heading, summary and then detail and story. Would bring clarity and structure to our process.

Love!
Sm_avatar

I have been agonizing and preparing for two surgeries I have procrastinated over a year. It should not be anything too serious. I am also near end of settling my parents estate with two greedy psychopathic siblings. So currently I am just useless.

I received post from Andrius on transitionuk.ning.com, and asked him to come here to contribute some of his ideas. Perhaps he will have some more considerations to confuse us! I am just grasping most of the time, and hoping that my past mistakes might not be revisited, and some of my better current understandings conveyed.

I would believe that there are standard markers or coding for digesting the imports and exports of information. Google is full of all sorts of instruction on how to properly code and mark up data. Currently I have been playing with google reader, and someday hope to become fully competent in editing feeds for selective presentations. but I know there are other programs that can assist in editing and presenting feeds. maybe some are free like google's? and, of course, I am totally impressed with wiserearth's google map interface.

i sort of see a TN CENTRAL functioning mostly as a support tool for technology transfer and information and advice, a place too where technology support people might show up too. for instance, i used to run program called DRAGONFLY, and continually relied on their main website to get addons and other relevant info pertaining to getting my website functioning or developed they way I wanted to.

Also, by experiencing many other websites (running Dragonfly) I was able to figure out what worked well and what did not. there would be forums where people could find technical assistance for various problems. but in the end, it was all about Dragonfly, and eveyrthing conforming to their program platform, or them conforming to API, PHP, open source, Google, etc. FYI, after Dragonfly, Ning, Grou.ps, mixxt.com - joomla, plone, drupal, all look like proprietary impossible-to-steer programs, usually require $60 an hour technicians to program and run.

and as for Google, after going through all their new widgetry a few times just over the last 8 months, and examining programs online, it is remarkable what could be around the corner. or maybe the key or a key will be in the calais semantic stuff that upperholme just supplied a link too.

but for visual sake, check out what google does for the Real Estate market, the real estate programs/maps, if rewritten or tweaked, could make an interesting tool for Transtion Network Central. I think  yahoo also has real estate programs too, but I would be more inclined to stick with the more popular google database rss mapping capabilities.

 

Consider too, the ways in which one can SHARE with Facebook and other sites. In my browser, i have now SHARE ON FACEBOOK and GOOGLE MARKER, with a click, it automatic shares or transfers a page to facebook for me, or adds a marker to my custom search engines which i use at my websites. or, i automatically add a feed to my RSS news

Sm_avatar

Hi Sandi,

 

I'm reading what you write but feel the need for some expansion on your ideas. I'd appreciate getting more clarity from you specifically in the area of what you are recommending for me to do (rather than not do) if you get my drift.

 

I really understand the point of recommending for us not to go alone. That's why we would base the Sharing Engine on industry standards for news, events and media descriptions. You've prompted me to consider that we do not need to be restricted to TN website information but could also ingest feeds from other sources and make those too available for syndication.

 

For projects and skills, where the objects in question are more specific to TN, we may need to spin our own schema. However, the encapsulation/encoding method will still be industry standard be it RSS, Atom, RDF or microformat. The aim here is not to isolate TN from the rest of the world but to assist TN to interface with and share with (receiving and giving) the rest of the world.

 

Also, it's about encouraging TN websites to participate in semanticising their data. The Sharing Engine is a reference model that we hope that TN websites will adopt. This will only assist in the TN message spreading still more rapidly - as people are then able to use the data in mashups and ways that we haven't even considered yet. Cool...

 

Please keep your comments coming they are an inspiration to me.

 

Cheers Daniel

Sm_avatar

thanks for sharing this. Sounds exactly like all the acrobatics and issues I faced in 1995 when I first tried to setup a mutlipurpose website and directory that could function on many platforms and levels, that are now, of course, defunct or "degraded."

 

A long time ago, i took our online HTML database catalog and macro inserted the necessary markers to make it easily searchable with google per entry. Today i need to do a complete overhaul of the site. But is still sort of works.

 

I think Google will continue to be the future for organizing and retreiving information. advanced search feature allow for

 

keyword site: macronet.org


on html records


Perhaps grading systems too, can help preselect suitable material automatically. Perhaps some of the systems you believed were not complicated or sophisticated enough, such as Ning, Google, Wiserearth, etc. might eventually develop the overriding capacities to accomplish the tasks you need. I would not focus so much on backward compatibility that much to start. Yahoo has grading systems, Ning does not.

 

While we were busy here hashing over the non future of a TN core site, Transition sites doubled their population on Ning and Idealist.org joined the Obama activist/volunteer online project. Is trying to be too independent hijacking bigger and better opportunities to work with larger or emerging platforms and initiatives?

 

Sandi

Sm_avatar

Take a look at http://opencalais.com/ for an example of a third party hosted service that takes a good bit of the pain away. I think there are other approaches that may be equally/more applicable.

 

Regards

Graham

Sm_avatar
With the soon to be launched WiserEarth API something like this could be created - I'm imagining that a bunch of TT org listings, an event, a resource etc is sharable through modules/widgets that work on the main TT platforms in use to today - Ning, Google, Yahoo, Joomla, Drupal, Plone etc.
Sm_avatar

>Sharing information becomes a whole lot easier when we make our information "semantic" and bring meaning to our information in a machine readable way.

 

That may be so, but isn't it a big challenge to get large numbers of people to set up info in this way? Any ideas how you tackle that challenge? Are there any actual, current examples of people creating this sort of info re sustainability / transition town stuff etc that you can point to that we can look at?

1 to 10 of 10 Comments