UPDATE: Ward left Microsoft
I don't need to explain wiki to this audience. It;'s so tiny it doesn't need explanation, but you don't understand it until you have been there and done that. It's you and the community that participates that makes it real, gives me perhaps too much credit. My hope is that wiki becomes a totem for a way of interacting with people. Tradition in the work world has been more top down, while wiki, standing for the Internet, is becoming a model for a new way of work. Largely driven by reduced communication costs, it changes what needs to be done and how it's going to get done. I hope that the wiki nature, if not the wiki code, makes some contribution.
A wiki is a work sustained by a community. Often asked about difference between wiki and blog. Something tangible is ve The blogosphere is the magic that happens above blogs -- the blogosphere is a community that might produce a work. Whereas a wikis a work that might produce a community. It's all just people communicating.
One's words are a gift to the community. For the wiki nature to take whole, you have to let go of your words. You have to be okay with that. This goes into the name, called refactoring. To collaborate on a work, one must trust. The reason the cooperation happens is we are people and it is deep in our nature to do things together. Important to make a distinction. Cooperation has a transactional nature, we agree it is a mutual good. Collaboration is deeper, we don't know what the transaction is, or if there is one, but if I give of myself to thsi collablration, some good will come out of it. You have to trust somebody to collaborate. With wiki, you have to trust people more than you have any reason to trust them. In 1995, it was a safer environment, don't know if I could have launched wiki today.
Refactoring makes the work supple. Word borrowed from mathematics, not going to change the meaning of the work, but change it so I can understand it better. Continuous refactoring. Putting a new feature into a program is important, but refactoring so new features can be added in the future is equally important. The ability to do things in the future is something that I consider suppleness, like clay your hands that accepts your expression. Programs and documents get brittle very quickly. Wiki imagines a more dynamic environment where we accept change, with the aid of a computer not make that dramatic, embraces hypertext which lets a document start small and grow while always being the right size. When there are two ideas in the page, split them into different pages with new names, so a third page can reference both. This is built into the web in some sense, it's just exploited in a wiki. Phenomenal that so much as been done in a tiny text interface, writing an encyclopedia. I have to apologize as a computer scientist that we have to go through that, but also says how strong the desire is for people to work together, but I look forward to the day where we don't have to do it just this way.
I was in favor of anonymity when I started this. Anonymity relieves refactoring friction. Have learned that people want to sign things. But try to write in a way where you don't have to know who said it. But when someone who is not in a giving mood uses anonymity (spammers), that abuse can drive us away from anonymity. But I hope we can drive the ill-intended out without having to give up the openness. Can one trust the anonymous? If you think of trust as believing people will behave in the way they did before, it seems dependent upon identity, but it may not be imporant to know if online behavior is consistent with offline behavior. But knowing what is going to happen when you give something away is significant.
The web has been an experiment in anonymity. Conscious design of low level protocols. Lots of identity infrastructure has been created to make it an online shopping mall, which makes it unpleasant for all of us because the machinery isn't that great.
Result: people can and do trust works produced by people they don't know. The real world is still trying to figure out how Wikipedia works. A fantastic resource. Open source is produced by people that you can't track down, but you can trust it in very deep ways. People can trust works by people they don't know in this low communication cost environment.
Result: the clubby days of friendly internet are over. Lots of technical questions about to sustain something we have experienced in a more complicated environment.
Opportunity: reputation systems for the creative (non-transactional). Reputation systems are an umbrella term for where the computer keeps more track over who you are and trys to make that visible in controlled ways to other people. eBay as an outstanding example, creating a space that didn't exist before. Again, going back to collaboration vs. cooperation. Doing this well depends upon excellent collaboraiton between the scientific community and the practitioners. Hopes this symposium becomes the center of this exchange.
Opportunity: organizational forms supporting creative work. The form we have today is a legacy from GM. Corporations aggregate and deploy capital to make things happen. Necessary back when communication was more expensive in this country. Top down hierarchies make communication work when it is expensive, I hope that wiki can be a flagship in this move in the industry to produce computer support for this kind of work and evolve organizational forms.
Eugene Kim asks about the conflict between anonymity and reputation. He calls it an opportunity because it isn't reconciled. The first thing we think of with reputation will be wrong and has adverse impacts. Do it by watching the impact it has on people in the area of creativity. Doesn't have to be complicated, but careful with what it reveals. If you walk in
Richard Gabriel: reputation can be attached to an individual or to something, such as words. The reputation can be attached to the words can enable anonymity. Ward says great, idea -- take notes.
On moderating change in the original wiki over the past year, and the tools he created for it (the following is probably only of interest to wiki moderators)...
Less than a year ago, my site, which had been growing exponentially, got to the point where things were happening on the site, spam. Where a dedicated set of volunteers had been defending the site, the posting of spam links put a load on them that wore them out. Combined with that you had arguments about page content, he left it to be a community standard, but bathroom humor pages were fended by programs made by programmers, increasing page edits. Finally had some people that wanted to behave by taking from the community. Had all three going on this year. Next thing I know, I get lots of email asking to shut it off, go read-only and call the experiment done. In the midst of this I was trying to have a career, but this was on the brink. Neat thing about exponentials is that they go on and go on and then suddenly a crisis. Didn't want to go into a programming contest. Created a community within a community, put the smallest amount of login I could. Captchas, or if you are from a place on the net known for abuse, you had to go through people that were close to Ward. A group of people that were not anonymous, asked their postal address to make a strong connection between the cyberworld and real world. Also asked a modest but non-trivial contribution to the running of the site, $100. People were willing to chip in. There are about a dozen people on the site that I consider Stewards.
The site hit a plateau, changed programming purposefully, would not be disappointed if it didn't keep growing. Will have value as a site, but if I made it read-only it would ruin the way people really read it. The question is what you want to change in the world before you plateau. Inspired by Clay Shirky's a group is it's own worse enemy to develop the Steward process and foster a core community collaboratively. One question is how open it should be, the Free Speech question, people submitted paragraphs between this is about programming or this is about free expression, turns out by letting people sort them, the community is more about free expression. 1/3 or 1/2 of them help moderate changes. Tried to send the password out through a trust network, but there are modern ways to do this better.
Shows backend housecleaning tools: Controls, Browsing, Editing, Visualizations, Queries and Monitors. Originally Recent Changes was just another page, so people would annotate them, so it was informative, not mechanical. The only editing people did in recent changes was the deleting older changes or silly gaming, so made it computer generated. Shows a dashboard for within seconds, minutes and hours where you can see number of edits to see when there is a massive change, click on it to see the diff, and have revert button. After reverting something, gives suggestion for what else you might revert (changes from the same IP address). Big guns for an edit war are good for the good guys. Visualization of how much posting is going on, a grid with days as rows and hours by column with the edit count as the variable. Shows a cool Sparklines-like view of activity per IP address over time. Shows some more advanced tools I think we are not supposed to blog about to tip his hand to spammers.