Last edit

Summary: typo: part --> parts

Changed: 25c25

< A [ WikiHive] is a collection of wikis that focus on the various aspects or part of a topic.


> A [ WikiHive] is a collection of wikis that focus on the various aspects or parts of a topic.

A way to organize several wiki:

Hubs: CommunalWiki

Spokes: TopicallyFocusedWiki?, lower traffic

Like CommunityWiki and TaoRiverWiki?.

The community participates in entire HubAndSpokeWikis construct. Certain members care more about certain topics, but people meet in the hub.

Advantage: Spokes do not need to maintain a CriticalMass?. Only the entire HubAndSpokeWikis construct.

If spoke reaches CriticalMass?, splits off into own wiki. Develop own character.

This is really the same idea as having FractalWikis within one wiki, or as having PageClusters within one wiki. But a different technical implementation. Also, it may facilitate the movement of the spokes into their own independent wikis later on. Like those ideas, this is a way of shoehorning large topics into their own domain (their own RecentChanges?) so that those who don't care can ignore them; but without creating YetAnotherWiki, which has trouble attracting CriticalMass?, and makes things harder to follow for everyone.

One might say that under this proposal, the "lifecycle" of a wiki would start out as:

  1. New page on hub wiki
  2. PageCluster on hub wiki
  3. Spoke wiki
  4. Independent wiki

By the time the new wiki has reached the independent wiki stage, it has already reached CriticalMass?. Some topics (those which are narrow enough) would remain spoke wikis forever.

See also the discussion on "sub wiki" vs. "child wiki" on CommunityWikiSplitProposal.

PublicWebJune2004Map CategoryWikiProcess

See also: PublicRefineryProcess

Alternative Organizations

Network structure with focus hierarchies: WikiNodes refer visitors to sites dedicated to related topics at the same level of focus, as well as to broader categories or more focused sub-topics.

Network structure without clear hierarchies: InterWiki is a suite of very simple technological tools to implement wiki connections easily. A selected few SisterSites, for example, offer the ability to link to similarly named TwinPages.

A WikiHive is a collection of wikis that focus on the various aspects or parts of a topic.

Why are we not seeing any significant hub and spoke wikis?

(to be filled in by refactoring discussion below)


So, as a more immediate plan of action, we should create a WikiProliferation PageCluster, a WikiOnWiki PageCluster, a WikiForEducation? PageCluster, and a HowToRunAWiki? PageCluster. As these get bigger, we should ask Lion to make us TaoRiver wikis for them, and move all the content to there (or, perhaps move the educational stuff to the coedit space later). Also, I think we should move all of the CategoryWikiTechnology content to WikiFeatures, and rename WikiFeatures to WikiTechnology.

Awesome. Needs its own page.


  • Licensing, Licensing, Licensing.
  • DougHolton? should be the host of education-type wiki. He's got an education WikiHive running. Sad that he's not here with us now.
  • "Spokes" may have multiple hubs..! If two hubs take an interest in the same topic, I think this is the way to go. That shared spoke may even develop it's own community, out of the mixing of people. It would also route ideas between the two hubs.

I feel NameSpaces are a nice transition between cluster and independent wiki.

Yes, I agree with both of the previous comments.

Licensing and also WikiProcess. If the spokes are really delegating all of their "community" stuff and most of their process discussion back to the hubwiki, then the HubWiki? should have the same WikiProcess (or at least should be a good place to discuss the WikiProcess of the spoke wikis).

And, as you said, the licensing terms should be identical or at least two-way compatiable. However, the most important direction is that content should be able to be copied downstream, from the hub to the spokes. Otherwise, conversation in the hub will have a hard time "budding off" into the spokes.

A more specific conclusion is that, if some of the Taoriver wikis are really the spokes of CommunityWiki, then they should adopt compatible copyright terms (i.e. not PublicDomain).

Also, if they are really to become CommunityWiki spokes, then that should be noted on the relevant WikiNodes.

I cautiously agree.

We have, as I see it, four issues here:

  • WikiProcess alignment - disagree
  • License alignment - sort-of agree

I'm not so sure about WikiProcess being in alignment- I can easily envision a spoke that had a very different WikiProcess than the hub. In many cases, this would be ideal.

I'm imagining:

A wiki buds off a spoke. With time, that spoke becomes a spoke of other wiki as well- maybe a "Shared Spoke," or something. With time it even becomes it's own community, maybe. But it has a sort of- a stricter process, perhaps, for organizing what is made by the various hubs, and shared in the shared space.

In this case, it's good that there is a difference in processes.

As for licenses: Yes, definitely- the spoke wiki must be able to receive from the hub. It's not as important that the hub have a license that can read from the spoke.

This seems like not so much a problem. In the hub, people can just compile modifications to the spoke's page, or a whole-sale replacement. Maybe they can transclude, (and clearly label,) in a non-modifiable way, the spoke's page, for easy reference. I don't know.

But if there is a WikiLegal:SlopeOfCopyLeftRestrictions?, it seems to ramp up, pretty clearly, from the Hub to the Spoke.

Of course, you know me. I'd just make it PD across the board. :)

Heh. Maybe we need the concept of "containment zones" within wiki. "Danger Will Robinson, Danger! You're in a GFDL'ed territory!"

As for- "if some TaoRiver wiki are spokes of CW, they should adopt compatable copyright terms":

Yes, I agree. Either that, or we need to get agreement every time we ship over. Quite right.

Two things-

  1. It's not a perfect spoke, because many of these have a community of their own. Small, yes. Only remotely independent.

Shoot. Gotta go. Between giving a little, and nothing at all, I'm going to choose giving a little.

Yes, I agree that the WikiProcesses don't have to be indentical. Certainly, the whole point of some of the spokes could be to have a tigher or a looser process than the hub.

Just compatible, in the sense of shared community values.

As a minimum, people who are schooled in only one of the processes can't be considered rude if they go over to the other wiki from time to time. For example, if one has UseRealNames and the other doesn't, there'll be problems; the one with UseRealNames will get annoyed everytime a nym comes over from the other side, and will eventually get tired and break off the relationship so that they don't have to have UseRealNames battles with all the visitors.

Anyway, as I envisioned it, most of the meta-process, touchy-feely "community", "community ideals", and "what kind of a community do we want to have" discussions would happen on the hub wiki. Because the idea would be that the whole construct is one community. If the spoke wiki processes didn't share the same community values then they wouldn't be one community.

So, for example, some of the Taoriver wikis could indeed become spokes of a CommunityWiki hub, even though they have slightly different processes. The point is that they could share the same community identity.

By contrast, (although I don't know much about WardsWiki), I don't think that MeatBall could be a WardsWiki spoke. They have totally different community values and a different group identity. MeatBall would not be comfortable with some of the ways that people act on WardsWiki, and vice versa.

That's very interesting. I hadn't thought of it that way.

Heh- the British phrase, "The colonies," for talking about the United States, comes to mind. ;) I've heard people think it's offensive, but I personally think it's hilarious.

I haven't thought of WikiFeatures as the colonies for CommunityWiki, though there's clearly a tie.

I see people in WikiFeatures who I've never seen over here before. I saw someone using a nym too, but their comment seemed to be fair enough. A lot of people drop off notes on their pet projects, or a little bit about what they know. I don't think many people are actually linking into WikiFeatures, beyond from here, and am a little disappointed about that.


I don't even know if the word "community" applies proper to WikiFeatures.

I think of it more as a warehouse, or like some stone tablets out in a grassy field somewhere in Africa. People come and, you know- update the tablets. We AssumeGoodFaith, and I haven't seen any vandalism. It's proved useful: People have dropped off all sorts of information and ideas that i've never thought of before, and it's directed traffic and increased awareness (I think.)

But I wouldn't use the word "community" to describe it.

I think I'd personally prefer if it weren't a "colony" of CommunityWiki. We just happen to live close by, and use it a lot.

As for what this means for HubAndSpokeWikis- I suppose there's just a lot of different kinds of spokes.

Some spokes have communities, some spokes are probably just more like repositories.

Or am I reading this all wrong? My mind is by no means solid here. I'm floundering just as much as anyone else is, here.

Well, the current wisdom seems to be that a wiki can't function properly in the long-term without a community, i.e. that the stone tablet scenario can't work.

Personally, I think the issue is still undecided. Maybe wikis can work without real "communities" attached.

What I was thinking with HubAndSpokeWikis was that, if we assume that every wiki requires a community, perhaps we can economize on the "overhead" (of discussing community values, arguing over procedure, etc) by having multiple small wikis share one community, anchored by a CommunalWiki.

Essentially, the spokes would delegate the topic "this community" (or "" for Python fans) to the CommunalWiki.

(I'm not sure if that's a good idea, since it hasn't been tried much. I'm just clarifying what I mean here).

But then- what happens when the spoke starts building a community of it's own?

Hm. Regardless.

I really like this page, and I believe it's the founding blueprints for a theory of wiki budding.

I feel like dissecting it into smaller pages now. Thinking about how to do it.

In that case, eventually it leaves its parent(s), and becomes no longer a spoke, but rather an independent wiki!

But it will be glad that in the beginning, before it had CriticalMass?, it economized on ProcessOverhead by sharing its community and process with the hub wiki.

(yeah, go ahead and split it if you want to)

(Grafting discussion that started on WikiDebateBase)

I've looked at Visual wiki as a ScratchWiki for … just collecting stuff.

And it has become a ScratchWiki for …. just collecting stuff.

I have been feeling around for the purpose. It now strikes me that perhaps it should be a repository for WikiAsYouTeach?.

Which brings us back to HubAndSpokeWikis.

But it hasn't been working. Why?

Why do we never take content out of CommunityWiki, and store it out elsewhere?

"somewhere where you have a set of well polished concepts or arguments."

Perhaps it could be named something like a "Foo Reference Book", or "Foo Reference Material", or a "Foo Dictionary", or "Some Foo Axioms".

I agree entirely that I want to present a link to someone in a friendly way.

  • More like "You're interested in X? I found this page about X very interesting, not like some other pages that just confused me.".
  • Less like "Here's a link to a bunch of people that are on my side, not on your side. Post some comments. It'll be fun watching the flame war."

Why do we never take content out of CommunityWiki, and store it out elsewhere?

I think it's just psychology lagging technology again. People have this perception that once something is written, it's set in stone. You can't take a few chapters out of this book, a few out of that book, some nice illustrations from some other book, and bind your own custom book. All you can do is add a few comments in the margins. Once you send a letter to someone, you can't un-send it. Once you've posted something to Usenet News, it's there forever, immutable. All you can do is add follow-up posts.

Re-arranging material is not yet something people are familiar with – especially "other people's material" – except for a few musicians, media editors, and programmers.

But I hope that people will become comfortable with improving documents (both on the small scale of spelling, the larger scale of organization within one wiki, and the large scale of moving chunks of text from one wiki to another).

I don't see this as tech lagging psychology, or the other way around. I'm even confused about the "again" ("…it's just psychology lagging technology again.") That is, I don't see where the first instance was.

Are we talking about the same problem? It sounds to me like you are talking about "why aren't people using wiki;" WikiAdoption.

We're quite comfortable and familiar with using wiki, and re-arranging pages, I feel. We have refactoring problems, but I think we've explored that pretty well, within the domain of a single wiki.

What's interesting here is that we've agreed to the HubAndSpokeWikis model, and we've agreed that we think this is the way to go. And yet we don't do it.

So, there's probably something concrete and useful that we can learn here.

I'm thinking perhaps:

I am coming to believe that the defining feature of the online casual organization is the software. (I similarly believe that the defining feature of the quality of an education is the materials or the software.) We can call this the "hard cybernetic theory of human society:" We are organized as much as our tools permit us to be organized.

I can imagine us "pulling" and "pushing" pages to other wiki- MobileContent.

I can also imagine us taking a pages RSS feed, and including it in our own RecentChanges.

The key thing seems to be working it into our SharedAwarenessSystem, our working space.

Maybe David is referring to just that: We've had to learn to rearrange other people's words, other people's pages, and we need to learn to rearrange other people's sites. Same thing, different scale.

As for myself, here's why I think we're not seeing it too often:

  1. Our scope is rather broad, so topic is not a real reason to move off-site. Personally I'm not interested in the MultilingualExperiment, so I would invite people to move these pages to a different site. On the other hand, MattisManzel is not alone. In fact, his position is not even a clear minority position – many people seem to enjoy his cluster.
  2. We are a small group and share a lot of interest. Thus I practically always appreciate your posts, or David's posts, for example. There's never a reason to move any of our pages off-site, because I interpret a wiki as a tool for a small group of friends – and you're my friends in the sense that I care about what you care about, and I like to read what you write (and hopefully the other way arround).

Thus, the conditions for budding off wikis are not there.

As for why existing "spoke" wikis have not fared too well (other taoriver sites, for example), is I think totally not related to the concept of hub and spoke wikis. The reasons they fared worse than they could have are:

  1. WikiSpam started at about that time and these wikis were not prepared. This scared some people away (eg. me)
  2. Some of the projects didn't catch much attention with some people on this site, because their scope was very technical. Since the group of users on this site is already quite small, the shared group of users was even smaller. So small, in fact, that you might start to believe that those wikis were not even arranged as spokes around a hub.
  3. Out biggest "spoke" I felt was the WikiFeatures wiki. Unfortunately it migrated half-way to Oddwiki and then the old wiki was not shut down. Now we have two wikis, and this gives the impression of them being unmaintained, decreasing trust in them.

Anyway, that's my gut feeling about the situation. In short: Hub and Spoke Organization of Wikis continues to be the state of the art of managing topic and communities on a wiki basis. :)

Maybe what would make more sense is a sense of "submission" and "collection."

Suppose a group started collecting some particular sort of page. We would submit them copies of our page. If they did something really cool with our pages, and it was a topic we weren't actively working with, we could then redirect our pages to theirs, either wholesale (#redirect), or with a link "this is really being developed over there…" I can imagine this happening.

Another way it could happen if there were some interest that was interesting to (say) half of CW, but not the other half. That would spoke off as well.

Another type of interaction would be: We pull a page from another wiki, work on it here, and then send it back to the other wiki. We may, we could imagine by some RSS related technology, continue to observe changes to the page that are made on the other wiki, as well.

For example, let's suppose CommunityWiki took an interest in particular pages on Wikipedia: We actively help develop those pages. We merge the RSS feeds for those pages into our own. We maintain our own version of the page, (ignore licensing considerations, for the sake of this story) and it's perhaps similar but different to the Wikipedia version. We work on both.

Is this hub & spoke? I would say, "No, this is a different kind of model." Maybe loose connections, SharedDevelopment?, or something.

If I may rephrase what you were saying, Alex:

The reason we're not budding off, is because we're all interested in the same things. The condition for budding is when the cares are different. But they're not: We're all caring about the same things. So we do not meet the condition for budding.

If this is correct, then I believe I am feeling around for something different: The ability to segregate data in a way that cooperates with other communities. In that respect, I would think it should be possible, but we do not do it.

(Side note : Lion, you wrote "For example, let's suppose CommunityWiki took an interest in particular pages on CW …" - I assumed you meant Wikipedia, not CW and changed it. Though I feel uneasy about changing what other people write.)

(Oh, right- that's exactly what I meant. Thank you!)

Hm, good point. In that case my answer would be that I don't work in too many other wiki communities: I read MeatballWiki, EmacsWiki, Oddmuse:RecentChanges, my own, and a few blogs… There's practically no opportunity for the sort of cooperation you're describing, Lion. I guess I'm just not the kind of person to act as "go-between" for extended periods of time – I might lurk, but when I leap I'm in or out not half-way in between.

I totally agree.

It's like:

  • Who would we cooperate with?
  • What would we make?

We've got like- there's Meatball, right over there, across the corner. And then there's- you know, there's CraoWiki.

Now, we do cooperate at times. MachineCodeBlocks is (intermitantly) worked on by Mutante from S23, FlorianFesti? from MoinMoin-land, and then- there's Bayle and I here. So, that's InterCommunityCooperation, right?

But it's hardly HubAndSpokeWikis. We're not talking about working together on a focus-area death-star-of-wiki bona fida spoke. We're just talking about a spec. Being built on a single page. Maybe it'll grow to two.

So, it seems to me, that- HubAndSpokeWikis does not work, with the possible exceptions of:

  • getting a page from Wikipedia, hamering on it together, putting it back
  • occasional involvement ina WikiFeatures page, or something

"Dabbling" in outside processes.

Maybe we should call this other thing: PageImportAndExport?. Or PageContribution?. Or something like that.

But I feel like there's something I'm missing…

ShiftingPageStewardship? ? (well, something like that)

Maybe it will start being more meaningful when there is a better and wider understanding of WikiNode and InterCommunityCooperation are more widespread ideas ? - when a lot of communities get interested in working together, rather than just having individuals working together ?

Because, isn't that a bit what HubAndSpokeWiki? is about ? Inter-wiki action (reproducion ? transfering responsability ?) - not just awareness, as with WikiNodes ? At least, that's how I see things around here - OnTheInternet, we have a lot of people interacting, creating stuff, fitting in … but each community is a bit of a world of it's own, there aren't really mechanisms built out of communities - all is built out of people. If those communities are about, say, developing software, yay, it works ! Because, gettign software to interoperate is easier. It works for wikipedia, too. But it doesn't work for places like this - I can't fit CommunityWiki or even C2 wiki in any bigger picture.

HuAndSmokeWikis? calls up an image of something of which a given wiki is just a part - in a way so that it's WikiProcess is a component in the general mechanism.

At least, that's how I see things, but I feel I've drifted off on a tangent, too … this is too big-picture-we-shall-conquer-the-worldish ^-^

Maybe that's the point: Cooperation is always about people. Communities are just tags, they don't cooperate. They don't even exist in the real world. They are a name for a group of people. And it's people that cooperate.

Taking this back to my original idea, HubAndSpokeWikis are a way of splitting an existing wiki when you're reaching the limits of size or growth. Since we're not seeing any of that, there has been no need of wiki splitting and building a hub and spoke constellation to keep the community alive.

I don't know; That seems a bit overly literal to me, and I'm not sure what we explain by it: Companies cooperate. Governments cooperate, software cooperates, it seems that there are all these groups cooperating with one another. I'm not eager to drop the idea of aggregate cooperation.

Keichi is a member of Nekomi Tech Motor Club. But one day, he's walking by the rival Four Wheeler's Club. He sees Toshiuki hard at work on his go-cart. Out of kindness, he decides to help. This is personal cooperation.

Later, there's some big conference. The leaders of the Nekomi Tech Motor Club and the Four Wheeler's Club get together, sign some papers, and the members of the respective communities work together to host the conference together. This is InterCommunityCooperation.

It's true that at every step, it's only people talking with each other. But in the participants minds, the interaction is as so:

  Jim  \       /  Janet
  John -+-----+-  Jill
  Josh / A   B \  Janice

Jim, John, and Josh see themselves as members of group A. Janet, Jill, and Janice see themselves as members of group B. The respective members justify their mutual interactions on the basis that Group A and Group B have made a decision to work together. They may have never interacted before, and they may never interact again later.

In John's mind, the reason he's working with Janice is:

    --> the Idea of membership with Group A
      --> the Idea of affiliation with Group B
        --> interaction with Janice

(Actualy, this might be a neat page to have: WorkingWithOtherGroupsByAlliance?, or something like that. We want to stress that the reason people are interacting with each other is because they believe that their groups are interacting with each other. The fiction is intentionally made real.)

If John does not believe in membership with Group A, then he will cease to work with Janice. However, even if John disagrees with the notion of Group A affiliation with Group B, he may still interact with Janice, because he is a member of Group A.

Perhaps we should simply call the new page: GroupsAreReal?. (The unvoiced part of the title: "Social." "Social Groups are Real.")

Social machinery is real. We can't just say: "Oh, it's just people interacting."

We know it's real, because we run up against it, every single day. There are many times where some social machinery isn't working right, and, everyone would like it to work right, and perhaps even everyone agrees, spoken with individually, on the exact specifics of how to make it work right. Incredibly difficult and rare, but let's say it were the case, right? Even if everyone agrees on exactly how to make it work right, if the social machinery is not set up right (nobody has the time to call the meeting to address the issue, collect the trusted vote, and perform the change,) then the change still will not go through! It's because of social machinery and things like GroupsAreReal? and what not.

I think "ActorNetworkTheory?," or something like it, describes this kind of stuff. I've been reading about it on the side.

I wonder.

Sometimes a project with clients will involve people I like and people I do not like. Sure, I will collaborate with them because our respective companies have decided that we should. But there is a world of difference collaborating with the people I like collaborating with and and those I don't like collaborating with.

So yes, our respective organisations could decide that the project pages on our respective wikis should be organized in a "hub and spoke" config, and maybe we would do it because we are told so.

But if we could decide for ourselves – what would happen? We would all collaborate with the people we like, and then we can put all our pages on a wiki for our group, and that's it. We don't need hub and spoke because we don't care about the people we don't like.

Basically I'm claiming that organisations of volunteers will not "decide to collaborate with each other" unless there are people in the two organisations that like each other and feel some MutualInspiration. The people that are not interested will not cooperate with each other unless pressured to do so, I claim.

I think that the group's decision-making process goes a way towards determining interest.

If I understand your argument right, you are saying: InterCommunityCooperation may well just be individual choice in cooperation, because it's individual interest in the end that determines who interacts with who. (So, we may as well just dispense with the illusion of the group.)

I'm arguing the other way: I think that the group is a thing in people's heads, and that if two groups, which are non-human entities, make a decision to interact together, then the people will interact. It's still based on individual interest. But, individual interest is guided by the group.

The mechanism may be emotion: When you join a group, you sign over some emotional sway over yourself. Your emotions then sway your thoughts to be inclined to follow suit with the group.

This is not absolute: If someone really really doesn't want to participate in an activity, then you are right: Regardless of what the group chooses, they wont' cooperate.

The reason why I think it's okay to talk about the group as an entity, is because it is mechanical. I could argue that the Linux kernel doesn't exist, because there is just a collection of machine code instructions. "There's no kernel, only instructions."

But there very clearly is a kernel there. This is probably because things are organized in such a way and shape that we call it a kernel.

I suspect the same is true for humans arranged into groups. There are near-mechanical laws that govern the group.

For instance, this wiki itself is a mechanical grouping mechanism. Because of the way it shapes our vision and interactions, it causes me to view the things that take place here as part of "my group." It causes a series of thoughts to take place in my mind that reinforce the abstraction. If someone asks, "Do you belong to the CommunityWiki," my mind will automatically, respond with "yes." Like automatic interpretation of text.

I've been reading Bruno Latour, who talks a lot about these ideas. I've been primed by reading Cybernetics and watching cybernetic movies (T3, Matrix, etc.,.)

Ok, I accept that there may be cases where joining a group will affect you in such ways that you will share some of the group's goals even if they were not originally yours. After all, the group will do information processing, and based on the new input you get from mailing lists and meetings, you will change your mind on some points, be convinced of some things, start caring about particular issues and bang! suddenly "the group made you do something" – such as collaborate with somebody else. We also agree that this is not necessarily the case in all individual to group relationships and for all issues.

What does this tell us, if we return to the original question you asked: Why are we not seeing any significant hub and spoke wikis?

  • Let us assume that Community Wiki and Crao Wiki "want" to cooperate as groups. If groups are strong motivators, then there are several explanations.
    • The groups don't really want to cooperate. It's an illusion created by some vocal members such as ChristopheDucamp.
    • There are technical or social limitations that make cooperation difficult. As for our side, I think our technical integration of Crao Wiki is pretty amazing. We have NearLinks, we have NearSearch, Crao Wiki is a SisterSite. Similarly, we have an amazing translator and vocal enthusiast like ChristopheDucamp helping us. That's why I think this option is out.
  • Let us assume that some vocal members of Community Wiki and Crao Wiki want the two groups to cooperate. Assume that "groups" don't act as motivators and don't automatically mean shared interests in particular topics.
    • The topics don't really lend themselves to cooperation. There are no individuals on both sides interested in a discussion with the other side. I don't think this is true. We share some topics with Crao, we have the technical means to cross the gap, and yet we don't with the exception of a few vocal members like ChristopheDucamp. The only remaining explanation is the language barrier.
    • There really aren't many people that want to cooperate. Basically the same individuals trying to rally us are the ones interested in cooperation, and they're failing to motivate others. That's what I think is happening. We need to identify more interested individuals and help them, or identify likely candidates and motivate them. One by one, on an individual basis (assuming that group decisions don't count, remember).
  • Any explanations not depending on either assumptions:
    • The language barrier is too high and ChristopheDucamp's heroic translation efforts are in vain.
    • The social barriers are too high and it is hard for wiki users from one community to effectively know their way arround in the other. No good starting points, no filtered RecentChanges, undecipherable CommunityExpectations. This seems also quite plausible, I fear. Eventhough we have filtered RecentChanges, the user interface is arcane, surprising, and non-obvious for somebody who has only ten minutes to invest in figuring out whether it is worth more time to get the hang of it.


When I look over CraoWiki, it seems to me that CommunityWiki has a lot of presence: because ChristopheDucamp's heroic translation effort has succeeded!

Only: We don't get the benefit of it. {;D}=

I don't know, maybe it's just me, but looking through CraoWiki, I see a lot of references to CommunityWiki, and I see that a lot of ideas that we develop over here are used over there.

Only problem? The cool stuff they are doing doesn't get used over here!

And they are doing cool stuff- it doesn't take much looking through to see what's going on.

I'm really impressed by what I see going on over there. I just wish it were easier to understand.

But back to InterCommunityCooperation and HubAndSpokeWikis:

I think the two main obstacles are:

  • Translation.
  • No clear goals.

I suspect that if we could magically solve translation, though, that we would see a lot of InterCommunityCooperation.

Specifically, I think we would see:

  • Shared technology efforts. (The feature sets of our wiki would become more similar.)
  • More interlinking.
  • Perhaps even comingling of our regular RecentChanges pages.

But HubAndSpokeWikis is something else entirely; I think that would require a specific: "We want a shared compendium of knowledge about foo, ready for shipping as a collection."

Without clear goals like that, I don't see how it can happen.

Though, you know: I realized while organizing CategoryVoting, that we don't really have much. MeatballWiki and WardsWiki have far more. I can imagine us saying: "We talk in those wiki to talk about those ideas." Or, rather: "Whatever we say about those things, we ship back over to those wiki, should we have something to add to what they've already said." Or: "We will help refine those pages."

Reviewing some of the thoughts on this page made me think that the issue is not just how pages are shared among wikis, but rather how we share information, in general; something that we all do a lot of outside (as well as inside) wikis. To illustrate;

  • In my set of working associates, there are a number of people whom I recognize as 'experts' in some subject, just as they recognize me as having a greater expertise in some other subject areas. Since we have all known one another for a relatively long period of time, we have found is that there is a natural tendency for each of us to bring information that we know one of the others will appreciate, to the attention of that individual. This happens as frequently with Email, or a conversation as it does within a wiki – after all, its simply a matter of what communications media is most convenient to the initial pair. Another factor that comes into play, however, is the recognition that other, mutual acquaintances may also wish to have access to the same information, at another time. If this is deemed likely, then we tend to post the material, rather than using a media that is more directed and less enduring, such as an email or a telephone call.

These considerations, and related factors such as convenience, or the wish for a certain type of collaborative feed, seem to be much stronger forces than the technical ones that are specific to a wiki and its user community.

To rephrase it in my own words for my own understanding:

We have experts who recognize each other. They take notes for each other, and pass them to each other. (Messages.) If they think that they have interest to more than just one person, they'll put them in a document somewhere. (Documents.) DocumentsVsMessages.

The initial connection is based on what is most convenient to the original pair. There is a tension between the convenience of the original convenient interaction, and the later posting.

Can you explain more what you mean by: "These considerations, and related factors such as convenience, or the wish for a certain type of collaborative feed, seem to be much stronger forces than the technical ones that are specific to a wiki and its user community." ..?

It says, "stronger forces than," so you're comparing two things. The first is, "convenience," and I'm not sure what else. (There were a lot of considerations that came before.) It's stronger than the technical forces that are specific to a particular wiki, and it's users.

Correct me if I'm misunderstanding, but what I'm interpreting is: "There's a lot of conversation that is very convenient to have, in e-mail, or face-to-face, or by telephone, or whatever. It's very convenient of us to have these conversations. I like wiki and it's interesting and useful, but it doesn't connect very well with all these more convenient forms of conversation."

If I am in fact understanding right, then: Yes. I strongly agree.

And we need to, and can do, something about it.

This is what I was calling here the OneBigSoup project. The OneBigSoup page, as it stands, is a bit out of date- but, the basic idea: Mushing together our various mediums- remains the same.

If we are e-mailing someone, for example, we should be able to say, "This e-mail is publicly visible." It should have a ConnectionNode?- sort of a miniature wiki, connected next to it, so that people can link it into a larger discussion. That is, should someone happen to stumble across this e-mail you sent, they should be able to add a link that says, "oh, that's sort of like this thing over here that you might want to see," so everyone can see this thing in greater context.

If we are having a voice conversation, we should be able to automatically speech-to-text it, and index it, live. If we said the word "collaboration," then anyone who was searching for live conversations about "collaboration" should be able to find the conversation we are having, and listen in on it in real-time. (Provided we didn't mind our conversation being public, that is.)

If you're browsing the web, you should be able to see the other people browsing with you. (Again, provided that they aren't in "voyeur" mode: that they don't mind being visible.)

There's all kinds of stuff, that we can use to connect ideas with each other. We shouldn't have to choose between convenient messaging, and full posting.

When these technologies go online, I believe, we should see the InternetConcentration like we have never seen it before. I think, when this happens, we'll start to organize ourselves better. We'll start grouping ourselves and coordinating our groups with each other. I believe this, because we are doing it already: We can see it in the complicated structure of wikipedia, we can see it in the Meatball-Crao-S23-CommunityWiki shape, we can see it in the KDE, GNOME, Apache organizations. My theory is that as we get richer communications technology, we will organize ourselves in much more complicated and effective ways.

Expanding an bit on "These considerations, and related factors such as convenience, …"; I meant the non-technical factors that affect my selection of media.

For example;


  • ('A' = I wish to communicate (via a message) with one other person)
    • and that person is not within range of my voice,
    • THEN - I am likely to use a telephone - since I talk faster than I type.
  • ('B' = I wish to give a message to two or more than two people)
    • and cannot convene a meeting.
    • THEN - I am likely to send an Email to the group.
  • ('C' = I know the same message will be needed later)
    • and I know all of the recipeints can search a central location
    • THEN - I am likely to post there.

This pseudo-code is used simply to introduce some examples of a few of the the criteria that I may consider before selecting a communications method and its related technology. The main point I was trying to make is that these types of considerations come into play well before I consider any of the technology issues related to a using a wiki to communicate, much less a particular wiki's features.

I thought this observation might be helpful because so much of the preceding content was so focused on the wiki technologies themselves that it is possible to assume that these as 'Necessary And Sufficient' conditions, as opposed to being 'Necessary, But Not Sufficient' conditions for a wiki's success.

Now, I quite like the OneBigSoup extensions to this point that you've added. I believe that the evolutionary directions you anticipate will be explored successfully. In fact, I am already aware of several voice recognition projects that allow me to automatically digitize the voice mail that people leave for me, and file and index it, as one example. It will be interesting to explore other related ideas that extrapolate how wikis fit into what is generally refered to as the digital revolution.

The observation is very helpful.

We need to think about how to integrate the collection and organization capabilities of wiki with our actual workflow.

I know that there are some wiki where you can CC: an e-mail to a wiki page. I know Bayle's talking about integrating SocialBookmarking with wiki. These two things (e-mail, bookmarklets for SocialBookmarking) are convenient, whereas (like what you were saying-) writing wiki posts isn't. (Unless you're messaging someone on the wiki!)

I'm going to be thinking about it for a while, and probably writing some software.

alex, i can tell you, why you're not seeing hub and spoke wikis . because this is a private wiki, isn`t it ?

  • du sagst es selber . ganz klar und deutlich . aber private wikis sind ein widerspruch in sich und sollten verboten werden (will sagen: sie sollten nicht wikis genannt werden, sondern privat.seiten) . ungl√ľcklicherweise war auch das erste wiki privat (aber nicht weil es ein wiki war, sondern weil ward cunningham auch nur ein mensch ist und fehler macht) . doch wie er sich r√§uspert und wie er spuckt, das habt ihr ihm trefflich abgeguckt (schiller) .

i'm not interrested in your private parts .

Conversation about InterCommunityCooperation moved to InterCommunityCooperation.


Hub-and-spoke-wiki observed. ;D

Technorati-Developers-wiki:MicroFormats delegated to the microformats wiki.

"this page has been contributed to the wiki. you can see the new version here."


Define external redirect: HubAndSpokeWiki CriticalMass TaoRiverWiki ActorNetworkTheory WikiForEducation PageImportAndExport HowToRunAWiki SocialAffordances ConnectionNode FlorianFesti SharedDevelopment DougHolton WorkingWithOtherGroupsByAlliance WikiAsYouTeach SlopeOfCopyLeftRestrictions HubWiki ShiftingPageStewardship HuAndSmokeWikis PageContribution TopicallyFocusedWiki GroupsAreReal

EditNearLinks: TaoRiver WikiTechnology CraoWiki WardsWiki WikiNodes EmacsWiki MeatballWiki SisterSite YetAnotherWiki FractalWiki WikiOnWiki PageClusters MoinMoin MeatBall TwinPages CommunityExpectations