WebAnnotation is the idea that the whole web should be open to edit by those reading the web sites. People could then leave notes for themselves and each other, about the site specifically or about the content of the site (and those notes would then show up other places where the content shows up.)

There are several ways to do WebAnnotation:


The OverlayWeb makes the entire world wide web editable.

The OverlayWeb is a second web that twins the world wide web. When surfing the OverlayWeb, you aren’t limited to the original author’s version of any webpage. The owner’s version of the webpage is just one branch in the revision history tree for the page. You, the person browsing the page, decide which branch or branches to focus your attention on.

There are ‘’many’’ OverlayWeb’’’’’’s in existance right now, and likely to be many more in the future.

See also: MeatballWiki:WebAnnotation.


TedErnst: This site just came up in a google search: http://www.shadows.com/shadows.aspx?url=http://tedernst.blogspot.com%2F2005%2F04%2Fbloglines-down.html

Seems they want to have people rate websites. Do they have a page for every page on the web? Hmm.

TODO: summarize & write-up; include others as well,

== Phrase Annotation ==

todo: merge/integrate ConnectionPoint into this page

See also the example for PhraseAnnotation? on the ConnectionPoint page.

== The One Global wiki vs. lots of independent wiki ==

Benefits of lots of independent wiki (“WikiLandia”):

  • People want their own space.
  • People don’t trust others who are unaccountable to them.
  • People want to move in directions that others don’t want to go.
  • People want to tinker with the underlying technology.
  • Feature implementation gatekeepers.
  • Some people don’t like still other people.
  • People don’t necessarily trust other people.
  • People want to be with their own clique, but not other people’s cliques.
  • Some people are interested in one thing, other people are more interested in other things – different wiki have different ideas of what is OnAndOffTopic.
  • Policing problems.
  • RecentChanges problems. (see CategoryRecentChanges)

Benefits of one big all-inclusive wiki:

  • easier to learn one WikiSyntax than a dozen or so
  • easier linking
  • OnceAndOnlyOnce – a filter against a certain kind of spam only needs to be written once.

Benefits of an overlay web:

  • no longer frustrated when one comes across outdated information – can correct it right there, and/or add a link to a better source of information – even if it’s not hosted on a wiki.

= Discussion =



There are many ideas for how to create a single unified GlobalWiki.

I favor a single unified (distributed of course) page database that may be accessed through various portals. Each portal provides a particular view onto the space of the GlobalWiki. Even websites that don’t explicitly participate are included in the GlobalWiki by virtue of an overlay (similar to http://www.wikalong.org/).

new: 2005-09-18 12:45 UTCLionKimbro:

This is, actually, believe it or not, a very common idea. Check out MeatBall:InterWiki, and then Wiki:InterWiki – ancient pages!

So, what’s going on here? Why don’t we have it yet? Why isn’t CommunityWiki working on this task?

First, I’d point out the obvious: WikiPedia. WikiPedia is pretty close to the one big global wiki. After WikiPedia, there are the WikiCities, which can be for any topic. But putting that aside.

We have come to these ideas:

  • We shouldn’t look at just wiki.
  • The dream is organized society.

Let me explain for a moment.

‘’“We shouldn’t look at just wiki.”’’

We originally worked on InterWiki, but then realized: “Oh, wait, these other mediums are useful too.” WhatCommunicationSoftwareToUse, MediumAttributes.

We now think that we should work on getting wiki to work together with all the other mediums, and get all the other mediums to work with one another: OneBigSoup.

‘’“The dream is organized society.”’’

(See also: DreamsOfWiki)

Why have we all dreamed of OneBigWiki, the DistributedWiki, the MeatBall:InterWiki? I think the motivation is actually that we dream of an organized, integrated society.

Here are reasons why we think we prefer WikiLandia to one GlobalWiki:

‘’(reasons to prefer WikiLandia to one GlobalWiki moved up)’’

etc., etc., etc.,.

This may initially sound ColdBlanket disappointing, but I think it’s actually very liberating, because it points us in the right direction: What is it, actually, that we really want?

I suspect that you share the vision of the HiveMind.

This is primarily a social vision. No doubt, technology has just begun dramatically increasing our collective powers. But it’s going to be ‘’much more’’ than just “here’s one distributed wiki, all people should use it.”

Establishing a global Internet community is going to take some work. This has been one of our major works (SuperProject’’’’’’s, specifically: OnlineOrganizing?, OnlineTools?) since realizing that OneBigWiki isn’t going to cut it.

Here are some of our major thoughts on the topic, to date:

  • FracturedDemocracyBuilding – The problem. Hoards of groups that would like to rise up and do amazing things, but don’t even, ‘’can’t’’ even, communicate with one another. Not with the way we model things now, at least.
  • InterCommunityCooperation – Solving the problem. We can should must will figure out how to solve the FracturedDemocracyBuilding problem.
  • WikiNode’’’’’’s – The WikiNode system has actually succeeded. Wiki Nodes are hardly ubiquitous, only a tiny fraction of wiki actually implement them. But they have fulfilled their purpose: They have dramatically increased visibility between related wiki, they have increased InterCommunityCooperation. They have caused people to see new possibilities in online organizing. The nodes have grown beyond our own local community here. It works. However, it is still a baby. It just needs time to grow.
  • MachineCodeBlocks – The WikiNode system is just a beginning. The next step is to automate identification and connection. The MachineCodeBlock? is a general system for attaching machine-readable MetaData to ‘’any’’ page of ‘’any’’ wiki. (And beyond just wiki.) The first major application of MachineCodeBlocks will likely be to label WikiNodes. When that happens, we can use GraphViz and a spider to create a rudamentary graph of related wiki. Like in the ProjectSpaceNetwork example picture, but dedicated to just wiki. (To begin with.)

One of the nice things about the WikiNode system is that it doesn’t require that people “do anything,” rather, it only requires that they ‘’consent.’’ If people ‘’consent’’ to being bottom-up organized, then they can be. If they do not, their WikiNode is simply politely refused. That consent, though, leads to involvement with neighbors.

I strongly recommend looking over the SuperProject list, and spending some time reading the linked “small super-projects.”

CategoryWikiAgenda may interest you, and the three major parts- WikiProliferation, MassUseOfWiki, MassCoverageByWiki. (For which we support WikiLandia by way of WikiFarm’’’’’’s, SeedWiki, MotherWiki - the OddWiki,…)

OneBigSoup, and the collaborative editor, since you are into tech. It’s too bad we haven’t written up pages for OnlineOrganizing? or OnlineTools? yet; We put a lot of effort there.

One big thing that we’re doing, and that I believe I’ll contribute to next year, is PICA, the ProamInternetCommunicationsAlliance. I saw you mention that earlier, talking about consensus polling. So, you already know about it.


I pretty much don’t believe in IdeaOwnership? of any kind. Neither my ownership of the GlobalWiki idea, nor CommunityWiki’s ownership of it, nor MeatBall’s ownership of it, nor … anyone individual or group’s ownership of the idea. Ideas think us, they are all products of our HiveMind interacting with our environment. The GlobalWiki is an idea whose time has come so it is unsurprising to me that many many people are being thunk by it in many many different places and in many many different ways. And … InterWiki is not what I mean by GlobalWiki and neither is WikiPedia. Something like OneBigSoup is closer to what I’m finding in my mind.

The idea of an OverlayWeb was mentioned in passing in my earlier comment. I think the OverlayWeb captures the flavor of what I mean by “GlobalWiki”. We can switch on edit mode for the entire web simply by employing an OverlayWeb that twins it. For every page on the BalkanizedWeb there is (or there is the potential for) a twin page to exist in the OverlayWeb. Wikialong is an example of a tech solution that is headed in this direction. Wikalong installs a plugin in your FireFox (I like AccidentalLinking) browser that adds a wiki into your sidebar. For every page that you visit in the main pane, a “twin” wiki page loads up in your sidebar.

Wikalong doesn’t take it far enough though. There is no technical reason why every page on the internet couldn’t have a twin “overlay” page that anyone can edit. If you are browsing the overlay network, you see not the original content, but an edited version of the original content that is personalized for you (via SocialSifting). It may be the case that what is most relevent to you about http://www.dell.com is the original content published by dell. Or it may be that you want to see yellow sticky notes on top of dell’s original content to add useful commentary about various links. Or at the farthest extreme, you might want to view a complete rewrite of the page created by Corey Doctorow or perhaps one by Matt Groening.

While there are big technical hurdles (e.g., how/where to store the parallel “twin” web) the social issues are even more complex. For example consider

  • when do you trust links that have been edited by others?
  • how do you filter spammers and noise machines?
  • how much information about your browsing and preferences do you share with others in order to get more personalized SocialSifting?

I don’t have answers to these questions yet, but I think they exist and are important to discover.

DavidCary: ‘’where to store the parallel “twin” web’’

Rather than storing the entire “twin web” on one huge file server, perhaps it would be better to take advantage of the editable web that already exists.

When I’m looking at one web page, there is a good chance that some wiki exists that is dedicated to that web site. Perhaps it already has a good place to comment on that particular page. But even for people such as myself that suspect it might exist, it’s not easy to find.

Rather than assuming that there are 7 different perspectives on every web site (the ‘’Ford owners view’’, the ‘’Chevy owners view’’, the ‘’bicycle riders view’’, etc.), I would find it interesting to have a list of all the different discussion areas that apply to that web site. I imagine that whitehouse.gov would have dozens of different perspectives, while less controversial web sites would only have 1 or 2 discussion forums dedicated to them.

  • Open-access wiki don’t really need a “twin” web site – people can edit it directly.
  • many web sites (such as mozilla.org and ebay.com and sony.com) already have a wiki (on some other web site) dedicated them.
  • It is fairly easy to start up new wiki on the various WikiFarms.
  • http://aboutus.org/ is a place to leave comments about a web site as a whole. (perhaps aboutus.org would be an appropriate place to list all those discussion forums?)

‘’the social issues are even more complex’’ Yes. We’ve talked a little about them ( MindswapTrustOntology ResponsibleSponsorTrustNetwork ReputationBudget ReputationCurrencies WebOfTrust InvestmentWebOfTrust ). Perhaps someday we’ll figure out the perfect solution :-). Or at least a better one than we have now.

new: 2005-09-18 12:45 UTCLionKimbro:

FridemarPache: If you’d like to advocate WebAnnotation, I invite you to comment on this page.

I’d like to point out that, even using WebAnnotation, the same community questions and issues still arise: “Who am I annotating ‘’with?’’” “Who do I ‘’want’’ to annotate with?” “Who’s attention will I draw, when I annotate here?”

Conservative Christians may not want to share an annotation environment with Conservative Muslims, for example.

CliquesAndCommunities, still a reality. It’s just another medium for talking. It has its unique strengths and weaknesses. And there are a million forms of WebAnnotation. In a limited way, we’re already doing WebAnnotation, because we link to pages on the web. If you made a FireFox tool that said, “CommunityWiki mentioned this page” while you were browsing the web, that would be one step closer to more explicit WebAnnotation. (And so on.)

CommunityTiedToOneTechnology is another thing to take into consideration. If a community uses two interfaces, there are the people who prefer the one interface, and the people who prefer the other. Now you have two different spaces for conversation, and you have two communities. Did we EnlargeSpace, or did we just make a new one?

“What am I annotating? Why? And with whom? With what tools?” Important questions. DavidCary’s comment from 1 month ago, just before this, reflects this.

new: 2005-09-18 12:45 UTCLionKimbro:

I’d ‘’love’’ for CommunityWiki to standardize on and integrate with a web annotation system. It’d be a wonderful path of discovery and innovation. But I don’t think DiiGo is it. We haven’t even standardized on a SocialBookmarking system, (for a number of reasons,) and even that is “simple.” There’s even OpenSource software for it.

If we ever do an OpenSource GroupServer, things might change.

new: 2005-09-18 12:45 UTCLionKimbro:

Do we want to conceal our discussions and conversations from people who don’t use DiiGo, or any other WebAnnotation system?


FridemarPache: Thank you Lion for the invitation: I share the vision of openness and there are ways to combine the tools, such that peers, who don’t use DiiGo, can benefit from it. DiiGo is currently in its malleable phase, therefore I am so much engaged in it. “Nuff said”. More in the future, when the current community metamorphosis phase is over. Good luck. – FridemarPache 2006-12-15 00:06 UTC

PatrickAnderson: This reminds me of http://Annozilla.mozdev.org which I installed just last night:

‘’This is the the Annozilla project, designed to view and create annotations associated with a web page, as defined by the W3C Annotea project. The idea is to store annotations as RDF on a server, using XPointer (or at least XPointer-like constructs) to identify the region of the document being annotated.

FridemarPache: Hi Patrick, I visited your website and am impressed (not only) by its unobtrusive links . In the critical phase of the current community metamorphosis I won’t expand (as far as I am concerned) WebAnnotation discussion, so I am pondering to visit PatrickAnderson and asking you something on your installation. Is that ok? Or do you suggest to postpone my question until later. – FridemarPache 2006-12-15 13:43 UTC

PatrickAnderson: Hello Fridemar. It is ok with me for you to do as you please, but I still don’t understand the rules here at CommunityWiki, so I withhold 80% of my comments because I calculate they would cause anger even though my intentions are pure. If we could ever get PlainTextWiki going, things will be much smoother, because anyone that didn’t appreciate a citizen’s actions could choose to not subscribe to their input… but that is some time away for now.

See also: Wiki:XenophiliaByChrisPhoenix Wiki:AnnotationEngine

TwinPages: MeatBall WardsWiki


Define external redirect: IdeaOwnership PhraseAnnotation MachineCodeBlock OnlineTools OnlineOrganizing

EditNearLinks: OpenSource ColdBlanket DiiGo CritLink OneBigWiki DiigoService WikiFarm PatrickAnderson AccidentalLinking EnlargeSpace OddWiki MeatBall GraphViz SeedWiki AboutUs WikalongExtension CategoryRecentChanges WebOfTrust TwinPages AnnoteaProject WikiCities


The same page elsewhere: