My thoughts below in brackets.
----- Original Message ----
From: Henry Andrews <
hh...@cornell.edu>
To:
gcd-membersh...@googlegroups.com
Sent: Wed, April 28, 2010 1:00:52 AM
Subject: [gcd-membership] Community & Business
Hi folks,
I've been thinking about a concept that I think really gets to the core of several problems related to GCD membership recruiting, retention and morale. This first emerged as a digression in a thread with Jim Van Dore on the editors' list, and got a couple of replies agreeing with it. So I've been refining it a bit for presentation here.
The core idea is this: The GCD is both a community and a (nonprofit) business. We need to separate these roles and handle each clearly and correctly. Currently we are neither fish nor foul, but some messy amalgamation of both that satisfies practically no one. Folks looking for clarity and direction can't find it. Folks looking for warmth and discussion feel they're being marginalized. And a lot of things just aren't getting done.
The GCD can't survive without community. It's what drew many people here, and what keeps them here (and its decline, in the eyes of at least some, is what is discouraging them from continuing). But the GCD also can't survive without getting things done. We are a project with goals. If we don't keep moving towards those goals, we'll just be a bunch of people talking on the internet. That's not really a bad thing, but there are other places one can do that. The thing that makes the GCD distinctive is the data and its public presentation. If we don't continually improve both, we lose the interest and respect of users. As can be seen by how much interest we attracted when we finally, after years of stagnation, improved our server and data interface.
My proposal is that we explicitly and clearly acknowledge our dual nature, and designate that certain areas should be run as a business, and others should be run as a community. Some folks will gravitate towards one side or another, and others will feel comfortable everywhere. And some folks will, of course, hate the whole idea :-) But I think many of the problems we're seeing boil down to the tension between these roles. For example:
Decision making:
============
We don't want to be decisive because it goes against the notion of a consensus-oriented community. But all the folks who are just here to get their data in, or help others get theirs in through the approval process, are getting tremendously frustrated because this deference to consensus has resulted in unclear rules and no way to get a timely ruling on the ambiguities.
[Mike - This is probably our biggest hurdle at the moment as an organization I believe. And I know there is discussion to put some measures in place to allow rules to actually be set. And really kinda outside the scope of what we were tasked with in this committee so I'm not going to say much except "yes please, give us some way to make rulings". ]
Mailing List tone and culture:
====================
Folks on main are upset because its taken on a more business-like tone where off-topic posts are discouraged. Chat has atrophied as those same folks don't feel it's sufficient for the discussions they want to have. Or the placement of the border just frustrates them. As a side issue, the level of strife on a number of lists is neither business-like nor community-like. I'll come back to this in much more detail, but I don't want to give the impression that "business tone" means "OK to fight a lot and/or tell people to shove off".
[Mike - I can't really speak about what's going on now as this all seems to relate to things that happened after I left both lists last December. But what it seems like to me from reading the above is that a lot of the things that were discussed on the old editor list have been split and some went to policy and the rest went to Main. That's causing the people who used to be on main to become frustrated because they don't really care about that stuff. And I truely do believe that part of why Chat has atrophied is that some of us have better tools to communicate than an email chat list. A bunch of us are friends on facebook and conversations happen there that previously would have been carried out on the chat list. So some of the loss of traffic on chat may not have any sinister notion.]
Indexer / Approver interactions:
======================
We're dealing with frustrations all around on this. Both from indexers and from approvers, including some folks who are frustrated in both roles at different times. This is linked to the decision making problem, of course, but it is also a problem of a generally un-professional organization. We're used to fairly informal interactions among a small set of folks who were dedicated enough to work their way into the GCD. Which took a good bit of effort in the past, so we usually had folks willing to work through rough patches. Some of those informal interactions are very polite, and even nicer than one would expect from a professional team. But others, even with the best of intentions, fall short. And there's no clear way for an indexer to resolve problems aside from arguing with the approver. There's not even anything that tells an indexer what they can or should expect from an approver, or how they should expect approvers to react to their own
behavior.
[Mike - I might be mixing up several conversations from several sources on this. Are we talking about the actual interactions between approvers and indexers (ie: people not being nice, or being unprofessional in comments) or are we talking about indexers being given different answers to questions based upon which approver happens to be the one dealing with a given index?
If it is the first one, people not being nice or unprofessional, is this any worse with the current system than it was before with the old system? I know that we have had problems with a few indexers recently, but it seems to work itself out. I don't know of an approver being unprofessional. Any approver who crosses into incivility with an indexer should be talked to and removed if neccesary. My history with mentoring new indexers has been under both systems, and I've done a lot of them. Sometimes an indexer just doesn't work out. They do a book or two, decide it's not for them and just stop responding to emails. It happens and it's not really anybody's fault. I have found that a more laid back informal way of communicating with them seems to work better for me. I really don't think we'll ever get to the point where we have 100's of indexers.
If we are talking about indexers getting different answers to questions from different approvers, I know I've vented to some of you at times when I hear an indexer say "so and so told me X", and X is clearly wrong (like using commas instead of semi-colans) or whatnot. But new indexers are sometimes unreliable, or just didn't understand what the first editor said. I'm not sure this is really relevant to the task we were given here, nor do I think it's anywhere near our biggest problem. ]
=======================================
Separating Business & Community so both can thrive
=======================================
As noted, we need to be both a business and a community. We won't have much to build a community on if we don't get things done. And we won't have much of a way to get things done if we don't have a thriving community of volunteers, experts, and newly interested parties to drive and support us.
So, how do we split this? Let's look first at our mailing lists. They should break down as follows:
Business mailing lists:
=================
gcd-policy
gcd-tech
gcd-board
gcd-editor
Community mailing lists:
==================
gcd-main
gcd-chat
What does this mean? Business mailing lists should each have an extremely clear focus, defined by the board. They should have conditions under which discussions should be started, and deliverables that each discussion should produce. They should have rules that keep discussions moving towards resolution. Rules of conduct should be strictly enforced, as violations disrupt business. Habitual violators should be removed fairly quickly. More on what rules should be involved later. In general, business mailing lists should be driven by the business needs, and populated by folks who are working towards those needs (possibly in different ways, with disagreements, but still trying).
[Mike - Agree about the business mailing lists 100%]
Community mailing lists, on the other hand, should be driven by the interests of the community. Discussions should be open-ended. Rules of conduct should be enforced, as violations disrupt community, but there can be more leeway if the community members are not distressed by the situation. While each community mailing list should have a focus, it should be fairly broad, and the edges need not be strictly defined. Any community member may request that a thread that's drifted well off-topic be moved to another list, but there are no strict rules or consequences unless there's really an egregious offense (like discussions of religion or politics for their own sake, as opposed to in comics, taking over gcd-main).
[Mike - I agree in theory, though in practice it may be more problematic than that. As I remember from way back in the day, we used to just have Main and all email went thru there. Then due to some people carrying on long discussions that were not comic related (ie: the infamous "Ghost Riders in the Sky" thread) Chat was created. Then along came the Editor list, which was used for the discussion of "how will we index this in the GCD" type discussions.
While it was never officially it's "charter", Main and Chat became the places where anybody could hang out and discuss topics. If it was clearly not about comics it went to Chat. Any other topic that was comic related really could show up on either list, though most of the time it would start on Main. But a thread on the Chat list that started about some random non-comic topic would usually wander into some comic related discussion at some point. Almost never (I can't remember a single instance anyway) would one of these topics that wandered back to something comic related be asked to move to Main since it didn't belong on Chat anymore.
Maybe it's time to merge Main and Chat back together into one list and see how it goes?]
Looking past the mailing lists, there's the notions of what privileges and responsibilities go with either being a member of the GCD community (whether a voting member or not) vs taking part in advancing the GCD as a business. We need to strive for a certain level of professionalism in our business dealings (data collection and decision-making), while actively and intentionally carving out space for an informal community free from those sorts of restrictions.
Community:
Privilege: You get to be on the mailing list(s) and participate in discussions, and index or contribute error reports.
Responsibility: You are responsible for being nice to other people.
That's about all there is to the community side. You get to be there, and you're required to be polite. Otherwise, just do whatever :-)
[Mike - Community privlige/responsibility - A agree 100%]
Business:
Privileges include:
** Contributing to policy discussions
** Making policy decisions (through voting, consensus, whatever)
** Approvals
Responsibilities (some of which are also privileges) include:
** Representing the GCD professionally in communications with indexers, error reporters, community members, etc.
** Approving/rejecting indexes
** Working with error reporters to get their data into the system
** Keeping up with discussion topics that need to reach decisions
** Staying focused in decision discussions and producing decisions in a timely manner
[Mike - I have some minor quibbles with some of the above, but not enough to bring up at this point]
Not everyone who is involved on the business side needs to do everything. But the general ideas are the same with each role.
On the business side, there's a lot more to being part of the GCD. And some of it will likely not sit well with everyone. But we need to be consistent and professional in how we interact with our users (indexers and community members). We need to prioritize clear, documented, and *timely* decision-making over endless debate and deadlock in the name of consensus. And also over going out of our way to give everyone their say in the discussion. Producing policy decisions is work, and if you want that level of influence, you have to be willing to do the work to keep up with it.
I expect this will be a controversial position. But we're stagnating. The decisions we need to make are not life or death. Some mechanisms can be put in place for appealing decisions to the board, and/or reviewing decisions based on new input, or evidence that a policy is not working out. But continuous, courteous, and clear motion forward is key. Not only will it directly further our business aims, it will make the site more comfortable for all members of our community, particularly those who want the business side of things to *stay out of their way*. The less they find themselves in arguments with approvers, the happier we'll all be.
[Mike - I also firmly believe that ALL Business lists should be open for public view. Then there is not perception of a "secret club" over on the business side. Honestly I think every list in the GCD should be publicly viewable. I also know that is doomed to failure on the Community list, for various reasons I never understood clearly but many over there believe is neccesary.]
The topic of how approvers should fit into the business is a large one, and one that Lionel's started to approach on the board and editor list. But it goes along with the business theme. Rather than leaving all approvers to their own devices, we need some guidlines and examples of how to handle difficult situations. We need to publish a dispute resolution system so that indexers and approvers have an option other than just shouting at each other- approvers can go to the editors' list, but indexers have no easy recourse (many of them are not on any list, and don't want to be, but they're still an important part of our community and we want to keep them happy if at all possible).
[Very well thought out Henry, and while I may have issues with some minor things, overall I think you've nailed a pretty good start. I have the gut feeling that there are other topics that we should be dealing with also in this committee but at the moment they escape me.]
Mike Nielsen