A Proposed Model to Make It Work

Started by morel on Saturday, February 20, 2016

Participants:

Related Projects:

Showing all 14 posts
2/20/2016 at 8:52 AM

Erica and al., fabulous idea. Establishing recognition seals for high quality projects definitely fills a need at Geni. Thanks for bringing this up.

While thinking about how to implement this idea, a flood of "factors" hit me and my mind started to swirl. The task of granting "meaningful" quality seals may be overwhelming, if indeed the intent is to have seals that are "meaningful". (If we have half-baked criteria or if we apply them inconsistently, then those seals will become useless real fast.)

This task is so huge that I don't think that any group of "volunteers" can ever successfully control everything. Still, I believe we can create a decentralized model that will make the task easy to manage. As a starting point, here is the starting structure of a decentralized model.

The key components are:

1) The Special Collection Steering Committee

2) Project Class Boards

3) Project Champions

Let's say a few lines about each component.

THE SPECIAL COLLECTION STEERING COMMITTEE

Its responsibility is to define the "quality standards". Basically, the steering committee creates the meaning and value of the "quality seal". Hopefully, this group will be able to keep the number of standards low, simple and powerful. Standards are high level. This is not micro-management.

Another responsibility is to give the final approval of a seal issued to a project. It is important to understand that their "approval" is limited to only make sure that the standards are being met. (More on that below.)

Who can join the steering committee? Geni users that cherish quality and are willing to dedicate time in defining and reviewing quality standards.

PROJECT CLASS BOARDS

Truly, there are so many types of project that it will be impossible to establish specific quality criteria that apply to all projects. Quality will mean different things to different projects and different audiences.

Here we introduce the concept of "class of project". What is a "class"? Projects in a class all share the same quality criteria. For example:

... "Authoritative Directory Class" could be those projects that gather all the profiles of a specific group, like all the passengers of a ship. A quality criteria would be how complete the list is.

... "Story-Telling Class" could be those projects that provide a detailed narrative of an event, like a disaster. A quality criteria would be how well the story is written and how well profiles are integrated into the story.

Each class of project will have its own board.

What is a board made of? Perhaps a group of 20 volunteers with a quorum of 6.

Who can start a new board? A project champion who can assemble 20 volunteers.

What are the responsibilities of a board? Define specific quality criteria. Receive project submissions. Vote on whether a project is deserving of a seal or not. Forward their vote results to the Steering Committee.

PROJECT CHAMPIONS

Any Geni user who are the advocate, ambassador or evangelist of a project. Project champions make sure that their project meets the quality criteria established by the board they want to submit their project to.

HOW DOES IT ALL WORK

Here is an example of how this model could work:

Say that I am interested in people migration and I want to create a "Profile Migration Board". I find 20 serious volunteers and we define together the quality criteria of migration projects. We submit our proposal to the Steering Committee. The Steering Committee is satisfied that this new board will follow the "quality standards". The new board is officially recognized.

Next, the champion of the "Perche-Quebec" project is looking for a seal of quality. This is a people migration project, therefore, he makes sure that his project meets all the quality criteria of the "Profile Migration Board", then he submits his project to the board. The board evaluates the project. They may reject at first while giving recommendations on how to improve the quality. After the recommendations are applied, the board votes Yes. They submit their vote results to the Steering Committee. The role of the Committee IS NOT to agree or disagree with the board. The Committee is only making sure that the board has followed due dillengence, i.e. they followed the quality standards. If the Committee finds no objection, the seal is issued to the project.

WHY THIS MODEL MAY WORK?

Because each group has a simple purpose. The Steering Committee only sets standards and review their application. The boards sets quality criteria for their specific class of project, and evaluate if those criteria are applied in a project.

The burden is with the project champions, and this is where it should be. If a champion wants a quality seal, he/she is the one who has to do all the work to make sure their project is compliant.

What do you think? Is it worth exploring further?

2/20/2016 at 9:20 AM

I like it a lot.

It may seem a little beurocratic (spelling) but i particularly like the decentralization aspect: we need to reach more audiences, actively engage more users, and this could be a mechanism for it.

Keep working through the Perche-Quebec idea as example.

Reactions:

- 20 on the steering committee might be hard to acquire

- what external partnerships / sources would be used, how acquire? (in other words - outreach)

2/20/2016 at 9:51 AM

About "20 on the steering committee": The "20-with-quorum-of-6" I proposed was for the Project Boards.

For the Steering Committee itself, that is, the group that sets the high-level quality standards, the membership could possibly be any number. What is important is that the members are "quality lovers" and are willing to work with the other members in creating/updating high level standards of quality. An example of standards might be: "a project must be well-sourced" (however we define what that means). Another example that you brought up is "uniqueness to Geni" (once again, however it is eventually defined). These two examples illustrate standards that can apply to all projects, no matter the audience, the culture, the era, the language, the topic, etc.

Each class of project will have its own set of specific quality criteria and yes, they can have criteria about audience, culture, era, language etc. Each class will have its own board. I was suggesting 20 members with a quorum of 6 so that a board has a pool large enough to make it easy to find 6 voters in a timely fashion.

Ultimately, a board can decide on their membership by themselves (within reason). For example, some classes of project might be so specialized that finding 8 members could be the best they can. Other classes have such a wide appeal that they might need 25 members or more to have a fair representation of their audience. Once again, to make it easy and keep the bureaucracy down, let's boards decide for themselves.

Each Board can have their own project page, where they list their members, their quality criteria, and post a list of projects that earned their quality seal. They can provide guidance on how to submit projects to them.

2/20/2016 at 10:14 AM

About "outreach": great, fabulous point of discussion...

For example, we have initiated a collaboration with the historical society of Chibougamau for a projet related to the history of mining in Northern Quebec. For the project about German prisoners of wars who died in Canada, we have contacted the German Embassy in Canada, who provided valuable and unique insight of the historical background. I am sure that Geni users have thousands of stories about outreach and this is marvelous.

You are touching an important point: a quality project should demonstrate a potential for outreach, or is already collaborating with external groups.

Now how we define "outreach" is something the Steering Committee ought to do. And how outreach should be demonstrated ought to be set by each project board.

"Outreach" does most certainly contribute to a high quality project.

2/20/2016 at 10:39 AM

Would this apply to ALL Geni projects eventually? I could see that, to avoid duplication and organize effectively.

Re: Each Board can have their own project page, where they list their members, their quality criteria, and post a list of projects that earned their quality seal. They can provide guidance on how to submit projects to them.

So proposed Special Collection project nominations would route to the (appropriate) Board for the yay or nay? This solves the "who votes / who approves" question. Brilliance.

We haven't addressed "who implements" the SC - that's part of software development. I was presuming it would be added to the curator menu, but now I don't think it should be so easy. All it would take to break that is a curator routinely approving a project request.

What are the Project classes? You gave two examples above; but couldn't / wouldn't a project potentially overlap?

Let's develop a list of say six classes as a start, their project pages, calls for volunteers?

1. "Directory" (example project: "RMS Titanic")
2. "Story-Telling" (example ?)
3. "Migrations" (example project: "Perche-Quebec")
4. "Royal Ancestry" (example project: "Colonial American immigrants with Plantagenet Ancestry")
...

2/20/2016 at 11:00 AM

This sounds great! All i want to see is maybe an Ancient history, a colonial, a family surname, and a plantation project end up on this list.

2/20/2016 at 3:15 PM

Can you define "colonial" better ? I.E., an umbrella category for all colonists, and colonial related special projects ?

I was thinking about Descent from Antiquity - would that cover ancient history? (probably not)

How would a family surname project be a special collection though?

2/21/2016 at 7:26 AM

About "Would this apply to ALL Geni projects eventually?": It makes sense. As you say, we avoid unnecessary duplication.

The decentralization model proposed here is called "segregation of duties". It means that two key functions are the responsibility of two different groups: 1) global quality standards to the SC, 2) specific quality criteria (compliant with the global standards) to the boards.

We can see several advantages with this model:

A) No one group (or person) has too much power. This is something essential in a collaborative environment.

B) The boards are independent from each other, which means that they will be free to define quality criteria that are truly meaningful for their respective class of project (always as long as those criteria respect the global standards).

C) The "yay or nay" decision rests with the boards, who are best qualify to determine if a project is truly compliant with that board's quality criteria. Because there will be many boards, the workload of giving support to champions and evaluating projects will be spread widely, therefore, not imposing an undue burden on a small group of volunteers.

D) Champions will be able to ask a board for support. I think this is very important. Champions (and collaborators) are working hard on their projects. If they need guidance or assistance on how to make their project with better quality, it will be great to get support from a board. This is collaboration at its best.

E) Another advantage is recourse. If a project champion is not happy with a "nay" decision, he/she can ask the SC to verify that the board did apply due process correctly. The SC may ask the board to reconsider a decision if it looks like that board may have miss something.

F) Project champion will be able to choose which board to submit his/her project to. If a project is strong in two or more classes of project, it will be nice for that champion to have a choice.

All that said, to answer your question: I think we can make this model work simply and easily to all quality projects.

Private User
2/21/2016 at 7:41 AM

morel I like your ideas.

2/21/2016 at 8:14 AM

About how to implement the "Steering Committee". Here are a few ideas.

A) The SC needs a project page. I suggest we embed a SC section within the "Geni Special Collections" project. At the moment, I don't see why we should create a separate project for the SC. (If we feel a separate project is needed in the future, it will be easy enough to do).

https://www.geni.com/projects/Geni-Special-Collections/32210

B) We need a membership. I think the larger the number the better. More minds can contribute to developing great global quality standards. The wider the variety of background the better. I think that anyone can ask to be a member. If someone is truly a "quality lover" and is committed to collaborate constructively and in good faith, and is willing to invest some of his/her time and experience, I see no reason why that person couldn't be a member of the SC.

C) The SC has two voting responsibilities: 1) approval for the recognition of a new project class board; 2) review each project seal approval (by a board) for making sure that due process was applied correctly. To make voting easy and not cause undue burden on volunteers, I recommend we agree on a quorum.

Say for example that we have 35 members/volunteers on the SC. We could agree on a quorum of 12. This means that only 12 favorable votes are needed. (If more votes are cast, then this is fine too.) We can of course deliberate among members during the voting period. I think there is nothing wrong with having those deliberations public. I don't think they need to be confidential.

Say a member is on vacations. Thanks to the quorum, not being able to vote while away would not stop the process. This will relieve pressure on the membership.

Software-wise, it would be great to have a "voting page" where we can click on Yes or No or Abstain.

D) The SC determines and establishes the global quality standards. Perhaps it is too early to say, but I have a feeling that there is no need to have "a lot of" standards. Those already suggested (e.g. unique to Geni etc.) are a great start. I think what is important is not having a lot of standards, but having just a few powerful one... that are very well defined. The better the standards are defined, with plenty of examples and guidance, the easier it will be for boards to draw meaningful quality criteria from them.

I suggest that those standards be posted in a section in the "Geni Special Collections" project.

E) Finally, the SC needs to post a simple process for how a group of Geni users can create and submit a new project class board. The SC also needs to describe a clear "due process" for yay or nay by boards. The new boards that will form over time will all commit to comply with the due process. Other than that, each board will be fairly independent (the way it should be in a decentralized model).

I think that those steps (from A to E above) can get us starting on a solid foot relatively quickly. Over time, we just need to be open to add more rules, guidance or clarifications when needed.

Once we have the global standards, the due process for project boards and the software part for voting, I think the Steering Committee will be fairly low maintenance. Once stable, I don't see a lot of changes needed. I don't think it will take too much time from the committee members.

2/21/2016 at 9:23 AM

About "What are the Project Classes?": I think this is the fun part, where the Geni collective can get very creative and inspiring.

Perhaps over time there will be several dozens of classes. I think the number is not what is important. What will really make a project class board successful is the willingness of their members to help project champions in developing projects of the highest quality.

Quality is often in the eye of the beholder. This is why we need to give boards a lot of freedom in defining specific quality criteria for their class of project. Of course, all quality criteria must be respectful of the Steering Committee's global quality standards.

I think that anyone can try to start and sponsor a board. It is them who have the burden to demonstrate to the Steering Committee that their board is relevant. For example, if there exists already a "Migration Class" with its own quality criteria, and someone submits a proposal for "North-South Migration Class", then they better have solid reason why this new class is needed. One would think that the quality criteria for migration projects should apply as well, whichever the "direction" of the migration. But maybe there is truly a need for different quality criteria. Ultimately, it is the sponsor who has to make a compelling case.

Therefore, what are the project classes? We can start by proposing some that seem like obvious candidates, as follows:

1) "Directory Class": Projects that contain a comprehensive list of profiles within a clearly define scope. The list must not be trivial. Sources to justify each profile in the list are provided. A high quality project will show a critical analysis of the reliability of the sources, identify discrepancies between sources, and provide explanation of why some profiles may not have been included in the list. An hypothetical example, was "John Doe" a passenger of the Titanic? Source ABC says yes. Source DEF says no. The project would expand on the rationale of both sources and make an educated decision of whether John Doe is included in the list or not. Such analysis should contain an angle original and unique to Geni. Outreach to related authoritative organizations or experts have also been sought after.

2) "Story-Telling Class": Projects that provide a compelling narrative of an event. Events in scope of this class are intense human experiences of virtually any kind, such as a disaster, a discovery, a sporting tournament, a journey. An event should have a clear start and end date. A high quality project will fully integrate all profiles involved with the event, directly and indirectly. For example, an hypothetical project on "The Manhattan Project" would features not only the scientists and researchers, but possibly also their family, politicians, military leaders, peace advocacy groups. The story should offer findings and opinions that are original and unique to Geni. Outreach to related authoritative organizations or experts have also been sought after.

The brief descriptions above a just that, a brief description. Their respective board need to define clear and measurable quality criteria. For example, "comprehensive list", how do we measure that? 90% of potential profiles? 95%? 99%? 100%? Another example, "compelling narrative", who says if a narrative is compelling? a book critic? a professor in literature?

I think this is why it is important to give freedom and independence to the boards. The "story-telling" board will probably have members specialists in literature, communication, writing (in the language the story is told) etc. I am sure that together they will be able to define powerful quality criteria specific to their class of project.

It will be fascinating to see what the boards and their members will come up with. We will witness great creativity I am sure.

What may be other examples of Project Classes? I think Geni users will come up with great ideas, but if I have to give a few more possible classes:

- Hypothesis Class: Projects that present a genealogy-related hypothesis and test that hypothesis throughout the project, with documented method, state of the art, tests conducted, results, findings, conclusion.

- Community Class: Projects that seek to "rebuild" a community that existed during a time period (or still exists today), with their influential community members, demographics, history, culture, notable events and milestones, challenges, successes and tragedies.

- Collaboration Class: Projects that successfully gather a large number of Geni users and external contributors of a wide range of background, expertise, culture, geographical locations. The members of these collaboration projects are dynamic, very active, do frequently discover new approaches to genealogy research and are highly influential.

- Profession Class: Projects that present a complete history of a profession or trade, with descriptions of the different skills and knowledge. The key profiles of a profession projects are highly detailed and their contribution to the profession well established. The project expand on the impact of that profession on genealogy i.e. on the lives of those profiles who benefited--or were victim--of the people who exercised that profession.

Those are just ideas, but I hope we get the gist. I think that the Geni user community will come up with fascinating project class candidates.

2/21/2016 at 11:03 AM

You're right, the project classes are the fun part. :)

I am now seeing this a little differently (in a good way).

This isn't "how to make Geni Special Collections work" (although GSC's would be a "class" within).

Rather - this is a model for ALL projects. A project to structure projects.

And it's pretty brilliant.

So far I do not see a need to request any Geni software development resources, always a plus, as then we are dependent on their schedules & priorities.

Yet as a volunteer initiative by Geni - enthusiastics, the initiative will serve to increase Geni project & profile quality & professionalism, which should result in reputation enhancement & therefore, increased membership. I can't imagine they will object to anything making Geni an easier sell. :)

How are projects identified that participate?

I was thinking

- logos added to project of sign off / acceptance by the board (or boards) submitted to
- a project page index listing the projects under their scope
- an addition to the project title with the board initial in front so it can be searched for. I.E., DIR - RMS Titanic

I suggest as a next step spinning off this topic into a "related" project:

Project to Structure Projects

With a summary of findings on the front matter, and details (perhaps) copied from this discussion.

Private User
2/23/2016 at 7:06 AM

Little correction for the front page of the project it was actually the Easter Rising project that was selected by the National Library of England.http://www.geni.com/projects/1916-Easter-Rising-in-Ireland/17385

The RMS Titanic has been accepted by them too but after I submitted it to them later.

2/26/2016 at 10:46 AM

Thanks Private User - done.

Showing all 14 posts

Create a free account or login to participate in this discussion