@Camilla:

I'm thinking of creating just 12 core groups. One for each issue area. Group admins would ideally be WiserEarth Editors who are knowledgeable about their group's issue area. They will have to spent time regularly to update the group's content (especially the main bulletin board), facilitate connection and interaction, and perhaps do some outreach to those beyond WiserEarth. How to assign who gets to be a group admin will probably be decided by the editors community, democratically, somehow (we'll need to figure this out). How does this sound?

Yes of course, 12 groups won't be enough. Members can always create new groups as they see the need for them. Other than that, what Angus suggested can also be a solution. To have a new group module displaying keywords related to the group's issue area. Each keyword links to a categorized search result page which serve to direct members to site-wide content and activities (just like the ones in the mockup page). Another solutions would simply be an automatic link up of related (existing) groups based on a mapping of their current Areas of Focus to the new Earth-Charter-based Issue Areas. This of course will mean display in the Related Groups module/tile.

Bottom line is that these 12 groups will serve as starting points for new WE community members, and spaces where they can always go back to from time to time, or on a regular basis, to contribute, interact and collaborate with people of similar interest/work.

@Angus:

Good to know that the idea works for you. Yes, I've done a mapping of the current 379 AoF to the 12 Earth-Charter-based issue areas (see Using the Earth Charter as AoF taxonomy). Of course, extra pair of eyes and minds would help to ensure accuracy.

I like the idea of a new group module/tile consisting of keywords/tags based on that mapping which links to content and activities across the site.

@Roger:

Thanks for the good idea. I like it. The list of orgs, people, groups, etc. does seem unnecessarily redundant. However, since they're still necessary to describe the breadth of WE's community and content related to an issue area, how about if we then move these links to the popping out explanation for the info icon?

Always loved the genius of google's simplicity.

P.S.

I'm also in conversation with Dominic (DominicStucker) of Earth Charter International, which suggested some the following changes to the wording (already implemented in the mockup):

• Human Rights and Development --> Human Rights and Well-Being

• Consolidating Democracy --> Strengthening Democracy

• Education for Justice and Sustainability --> Education for Justice, Sustainability and Peace

• Respect for Wildlife --> Respect for the Community of Life

• Culture of Peace --> Culture of Nonviolence and Peace

He also suggested that the link to the popping-under full Earth Charter text be consolidated. So, there's only one link now, which I think looks better than having three.

Dear all,

On Earth-Charter-based AoF

We (me and Camilla) have just finished mocking up an Earth-Charter-based Areas of Focus landing page. Would love to get your feedback: http://www.wiserearth.org/article/ab58cf44ad3f1db351dfbfa3ce25b13d.

Here's the full Earth Charter for reference.

And some notes on the mockup:

• Areas of Focus will probably be renamed to Issue Areas as shown in the mockup.

• The short labels for each principle needs reviewing to best reflect Earth Charter's main principles.

• Each short label have an 'info' icon. When a user hover their mouse on it, it will pop a small window showing the original EC principle and some keywords associated with it (taken from EC sub-principles and beyond). We'll need to work on the content of the pop ups.

• Below it is an explanation of how WE's Issue Areas are based on Earth Charter.

• At the bottom, viewers can read the entire charter divided by sections. Clicking on the section title will pop-down the entire EC section under it (much like the info icon in the current AoF listing page).

• On the right are keywords (based on current Areas of Focus and user generated ones) which will link to a search result page for each keyword, displaying all content tagged with that keyword by category (groups, people, organizations, resources, events, etc.).
  
  It will also link to existing wikipages on the keywords (current AoF portal), or encourage creation of a new one. In there should also exist a link to the DBpedia page for that keyword. We can also think of a technical fix to facilitate syncing between the wiki and the DBpedia page.

On portal for each issue area

Question: Do you think it's a good idea to use groups, instead of modified wikipages, for future portals?

Wikipages seems a bit limited, and we'll need a space for discussions anyway, having a separate one in the public forum seems unintuitive and impractical. Comment section simply is too limiting. So, groups seem like the way to go.

If the answer is yes, this will mean that WiserEarth will have 12 core groups around the Issue Areas.

Potential:

1. Provide destinations that are populated, active and interactive enough for each community member and visitor to take advantage from and contribute to.


2. A showcase of how WE is unifying and integrating the movement it wishes to serve.


3. Will probably grow into some of the largest online groups around these issue areas anywhere on the internet (depends on how WE total userbase grows).


4. Facilitate formation of new (sub)groups as the core groups becomes too general for specific conversations and collaborations.
5. Solve a problem plagueing online communites and social networks, as observed by Todd (TDStong) in 1000 groups of 2 (essentially: many groups with few members and little activity).

If we go with this, we'll probably need some technical modification to these 'special' groups:

1. Auto populating of each entity module everytime a new entity is created, or the AoF (Issue Area) section of an entity is edited. So, probably no 'Add' feature anymore. Just 'Create', 'Manage' and 'See All'. Not sure whether people should automatically be members of these groups upon registration (based on their chosen Issue Areas).

2. May need ability for admins to 'lock' bulletin boards, i.e. make editable only by them. This is the equivalent of current 'locked' AoF portal pages.

3. May need a 'display randomly' feature for members module to facilitate serendipity discovery and connection (and avoid the group to center on most active members only). Much like the one we have in network section of entities. We should make this random view default and allow easy switching to 'most active' view.

4. May need to limit group messaging only to admins to prevent abuse.

5. May need multiple discussions module to facilitate categorization and avoid overload of the module.

6. May need to modify group header further to accommodate Earth-Charter specific needs. But may use as is too. Depends on how things develop.

How does this look and sound?
Bowo

@Angus: I think I agree with you.

Overall, I get the feeling that the hurdle to contribute content in WE is too great at times. Filling in all the details of an entity's info, selecting from a big list of AoF and writing the About section. If we simplify the AoF as you suggested, and present only the most important field for entity info upon creation, I think this will make it much easier for community members to contribute content. We can encourage further editing upon clicking the Save button. I know this is another discussion topic but something I see as relevant and important to mention here.

Back to simplifying AoF. Actually the first category "Respect and Care for the Community of Life", and the 4 principles under it, acts more as a base / foundation / overarching purpose for the other three categories and 12 principles:

While the 12 principles are the actual categorization of issues / actions. So, it may be best to just use the latter as AoF. 12 AoF is definitely simple and highly manageable.

However, I don't quite agree that AoF will loose much of their relevance when we take a recommendation system whole-heartedly. I see the key purposes of AoF and the portal pages as a place for learning, an entry point for search, and a place of serendipity encounters. Having the AoF structured along Earth Charter will help a lot toward this end.

It would of course be very useful to the community if we can implement a good recommendation system.

@ Rehan and Bowo: I worry about multiple level taxonomies and competing taxonomies which doesn't solve our complexity issue with AOFs, or the globalization issue in my opinion. Tagging any level of n in the taxonomy would be extremely confusing. Also, I think we should avoid bringing the Solutions 'taxonomies' too much into this conversation. They are really very specific and I doubt that too many of them will be created.

So I return to creating some mix of simplicity with the Earth Charter and allowing complexity with a tag cloud and DBPedia. The top level of the Earth Charter just simply speaks to me, probably because it came out of such a collaborative process:

I. RESPECT AND CARE FOR THE COMMUNITY OF LIFE
II. ECOLOGICAL INTEGRITY

III. SOCIAL AND ECONOMIC JUSTICE

IV. DEMOCRACY, NONVIOLENCE, AND PEACE

(and the 16 sub principles that fit under these four)

I could easily see someone looking at these four categories and saying - "my work/mission fits in one or more of these boxes", and the same for the 16 sub-principles. If a taxonomy is too long (like ours) it may simply serve to balkanize information and make it hard to find - I often say to myself AOF A and AOF B are so close together that I should tag with both or someone is not going to find my content. This is simply not helpful. If we categorize existing records to the Earth Charters 16 sub-principles (a small enough number) - based on a mapping from existing AOFs - we create the simplicity. If we then take the existing AOFs and turn them into tags we generate the tag clound associated with the record. Then we just need to make sure that tags are easily linkable / searchable / and somewhat confined.

What is the problem with this approach? Your thoughts and comments please?

Wow, this is probably the first virtual brainstorm I've taken part of :-)

It might be useful to return to the original question, which is actually 3 different problems:

• AOFs are too complicated for users
• AOFs are not conducive to "globalizing" the site
• AOFs are not very scientific

I think we all agree that these are fundamental issues with the current AOF system, and I think each point deserves its own debate.

AOFs are too complicated for users

I feel like a major stake in the resolution of this issue resides with the manner in which the user navigates and interacts with the AOF (categorization) system, especially if we are attempting to expand it to an n-level hierarchy. Any chance we could get IDEO to draft us some prototypes ;-)

AOFs are not conducive to "globalizing" the site

The initial point here was that the current AOFs are too cumbersome to translate. Personally I think the core issue is much deeper than language translation, and is closer to cultural translation. It might be a large error on our part to pre-suppose how the rest of the world should categorize content. That's not to say the rest of the world cannot use whatever system we devise, but it would be ideal for them to categorize content in a manner which directly transcends from their values (and language). This is one of the "3rd rails" of international development / aid (especially when involving technology), but I thought it might be worth bringing it up.

AOFs are not very scientific

I see the root of this problem coming from the fact that our current system is so large and complex, maintaining it and ensuring its logical validation has become an insurmountable task.

The solution I propose to these 3 problem areas is essentially a web 2.0 approach categorizing our content.

@Arash, I understand your weariness of multiple taxonomy tagging from a user experience POV, see below.

Solution: user's have the option of subscribing (and creating their own) taxonomies. By default all user's are subscribed to AOF or Earth Charter taxonomies (whatever we decide on). If a user visits an entity record that is not in one of their subscribed taxonomies, then the record's tags are hidden (although there should be an option to see all public taxonomies the entity belongs to). If the user is viewing an entity that is in one or more of their taxonomies, then for each taxonomy the tags are listed.

The majority of this technical functionality exists within the upcoming Solutions release. We would just have to modify and open the functionality to extend it to all entity types.

Hi All,

Sorry to jump into this conversation so late, but I'm happy to see that there's been a healthy amount of brainstorming on the subject.

From a technical standpoint, I'm interested in defining the requirements for implementing this suggestion on the site. Bowo did a good job of describing the steps to implementation in an earlier post:

1. Map the WiserEarth AOFs to the individual elements in the Earth Charter taxonomy
2. Develop a one-time, automated process to migrate all of the existing AOF tags on individual entity records to the corresponding Earth Charter element, using the mapping in step 1
3. Redesign the AOF user interface component so that it can accommodate an n-level taxonomy
4. Modify the existing AOF detail pages so that we can repurpose as  much of the Earth Charter definition information as possible, and also provide links to the existing AOF portal pages so that we don't lose the effort that was put into putting that content together
5. Provide auto-complete functionality on the keyword fields so that users are guided towards using DBpedia keywords wherever possible

Does that accurately summarize the requirements? If so, then on a scale of 1 to 10, it's proabably a 5, as far as technical complexity.

I'm a bit wary of using multiple taxonomies to tag records because I think that it introduces a lot of complexity from a user experience standpoint. Unless we expect our community of editors to take on the task of curating the tags on all of the WiserEarth content, we need to make sure that the task of adding or editing records is very simple for the end user.

One thing to be weary of during this conversation is the quintessential "which box does this go in" problem. It is in our nature to strive to put everything into neat little separate boxes, but I need only to point to the platypus as an example that one system of classification is not the ultimate answer.

As Angus noted below, the upcoming release of Solutions allows for (eventually) all entities to be classified by multiple, user-defined taxonomies.

The end result would be that an entity can be "tagged" by WiserEarth AoFs, Earth Charter Keywords, DBPedia categories, etc.

I took the opportunity of an annoucement forwarded by Thomas Bandholtz on the Linking Open Data forum at http://lists.w3.org/Archives/Public/public-lod/2008Oct/0006.html

to push the message I had proposed here a while ago.

http://lists.w3.org/Archives/Public/public-lod/2008Oct/0012.html

Concerning Thomas' message. The eEnvironment Terminology Workshop to be held in Prague in March 2009. See http://www.e-envi2009.org/?workshops

Seems the description is quite close to our preoccupations here ...

"Environmental terminology and its semantics constitute a major building block of SEIS (Shared
Environmental Information System) and SISE (Single
Information Space for the Environment in Europe) as important instruments for discovery, understanding, and integration of any kind of accessible information. They have already taken a long way starting with early subject heading systems of the libraries, moving on to multilingual thesauri such as GEMET, some of them evolving towards more expressive ontologies.

This workshop will present several domain-specific and interdisciplinary examples and discuss common design issues such as terminology structure models, cross-referencing, symmetric vs. asymmetric multilingualism, identity and reference, publishing terminology in the Web, and linking environmental data to such published terminology."

Thanks David for your comment - indeed, the inconistencies definitely would be addressed I think under this initiative.

Thanks Clayton for your insights. Perhaps Bernard could respond to why a newly created taxonomy is not an idea solution for WiserEarth in the long term.

This suggestion proposes that the existing 379 categories would be combined and reorganized (and if duplication exists, some may be eliminated) under the Earth Charter's 61 categories. However, the specificity of the many categories would be maintained using DBpedia labels.

As Bernard mentioned below, the suggestion would be for multilingual Earth Charters to then be hooked to a single set of URIs. Including links to related concepts in DBpedia. This would allow the specificity of the existing Areas of Focus to be maintained (without duplication or giving too much emphasis on certain categories) while providing a simpler taxonomy for users/orgs/groups to work with.

DBpedia

URI