Wikimania 2007 wrap-up

Yes yes, I know this already happened months ago … I’m just sometimes very very slooow to get around to putting things on my blog.

Wikimania 2007 wrap-up: constructive criticism, random comments, and a few photos.

Firstly, if you get a chance to go to Wikimania, then you really should go – you’ll have a really great time. I have only been to this one, but the Taipei organisers were amazing – they were very friendly, amazingly helpful, and the facilities with the lecture rooms being so close to the accommodation and so close to the dining areas and all with an ATM and convenience store nearby, was just superb, and all the touches (e.g. the power cables everywhere) were just great, and the other attendees were lots of fun.

Eric recently wrote a good summary of Wikimania, and towards the end invited some constructive criticism. Well, here are my thoughts on a few things that would have made an amazingly good conference even better, although most of this overlaps with wishlist comments that other people have made:

  • Double-sided name badges, so that you can always read someone’s name. I lost count of the number of times I flipped over someone’s name badge, or they flipped over mine.
  • More wireless microphones so that panels and getting audience input is easier.
  • Maybe try some Meraki mesh routers for wireless networking, as per Gerard M’s suggestion.
  • Include a notepad and pen in the conference bag so that people can take old-school notes.
  • A survey form, to gather feedback at the end of the conference. Example questions: Which talks did people like the most? What things worked well at the conference? What things didn’t work so well? What or who would you like to see at Wikimania?
  • Improve the online video experience for people who can’t be there. (Walter raised this point previously):
    • Background: No matter where you hold Wikimania, it’s a given that large numbers of people will not be able to attend. Now, I must point out that other volunteer-run conferences (like have the exact same problem with videos – their volunteers record videos with the best of intentions of putting of them online, neatly labeled by talk, in an open format. During the conference there’s no time to do this though, because there are fires that have to put out, talks to attend, people to meet, parties to go to – so post-processing and uploading the videos gets pushed back to “later”. Then the conference ends, the attendees go home, and the volunteers who have worked their-asses-off start to think about reconnecting with their friends / family / work, with the best of intentions of doing it some time later. But after a few weeks, it hasn’t happened, and it probably never will. It doesn’t really work, and the only time I’ve seen it done really well was at , but that required an amazing and probably unreasonable amount of work from the organizers, and even then not all of the videos were labeled with obvious names.
    • So my suggestion is this: Take some of the money that would normally go towards getting people to the conference (e.g. $1000 or $2000 would probably fly 2 or 3 people to the conference), and instead put that towards paying A/V non-wikipedians to do all the post-processing and labeling and uploading of videos, as soon as possible after the videos are recorded, and give them part of the money only after it’s done. That way, sure, you get slightly fewer people attending which sucks a bit, but suddenly many more people can participate, even if they cannot physically be there, which more than makes up for it.

And now some random comments:

  • The “Truth in Numbers” documentary from Nic Hill will finish filming short after the end of Wikimania 2007. He showed a preview at the party (held at the end of the second day). The audience had been chatting for a while, and had quite a few beers, and so was getting a bit vocal. So what was interesting is that when the video was show, people would cheer or boo, depending on who or what was being shown. They booed at the bit about Chinese Internet censorship – as did I. They also booed at Larry Sanger when there was a clip with a quote from him. Now, I have never met the guy, but I do think he’s getting a bit of a raw deal. What he was saying was intelligent, his point was valid, and I for one am glad that Citizendium exists. Currently I don’t use it, but the fact it exists at all as a competitor is a good thing. So please, cut the guy some slack.
  • Location of the next or future Wikimanias. (Note: Alexandria was picked for 2008 recently, so some of this stuff is now out of date!) Of course there’s a bidding process for selecting this, but people love to speculate and make wild guesses about locations. And why not, as a bit of pointless fun? So here were some of the places bandied about as possibilities for next or future years:
    • Alexandria: This would be close to Europe, be convenient for the Middle East and accessible to members of the Arabic & Hebrew Wikipedias. However it would be hot.
    • Australia: People said to the Australians “when are you guys going to host this?” I said “but the weather in Australia in August sucks! It’s cold, and it can be rainy.” But paradoxically, this just made people even keener – 2006 and 2007 were really hot, so a cool / cold location actually confers brownie points (Average temperature range shown here – 9 to 17 degrees Celsius average in August) … And whilst I would love to have Wikimania in Australia, having seen it first-hand, the effort involved in running and organising this is just huge! All the organisers of this have been absolutely amazing, and I look at what they have done and think “wow, that’s a metric tonne of work … dunno if I could handle that!” Brianna L seemed to have had similar concerns. So the current plan is give it some time, and to bid for Wikimania in 2010, and if we get it (and hopefully even if we don’t) to have a local Australian + NZ Wikimania in 2009 (probably in Canberra) to give us some practice and to put to some faces to names, and to get together in a way that is hard to in a large country.
    • South America: Gets you southern hemisphere + cool, plus sounds like they have an enthusiastic bid team.
    • London: Covers Europe, is accessible from the US, has a strong community. (Possible downside: expensive). But, as a bonus, the devs could have the Hacking Days at Rob Church’s flat / house. Just rock up. Uninvited. No more excuses from him about not turning up – just take it to him. :-)
    • Helsinki was also mentioned.
    • Bidding for the next Wikimania: Will probably have bids for the next 2 or maybe 3 in quicker succession (maybe two months or so apart) – so 2008, 2009, and maybe 2010 will possibly be bid for within the next 6 months. Benefit of that is that people can plan and organise sponsors; also allows people to learn.
  • The breakfasts were in a dining hall, held from 7:00 to 8:30 AM. And between 8:20 AM and 8:29 AM a large number of Wikipedians would arrive (including myself, usually at 8:29), just scraping in before the cut-off, no doubt to the annoyance of the kitchen staff. Definite Wikipedian <–> slacker correlation somewhere in there. Or maybe that’s just me :-)
  • Laptop fashion: Most popular laptops brands among Wikimania attendees were by far and away both Apple laptops & IBM ThinkPads. Having stickers for Wikimedia, Wikipedia, Firefox, Wikimania, and so forth stuck to the lid also brought bonus points, and a conundrum for the Apple owners: Do you add stickers, thus potentially ruining the sleek white plastic or polished metal exterior, but conferring status points? And if yes, do you cover the glowing Apple logo? Oh, these must be difficult and vexing questions indeed! :-)
  • Speaking of laptops, man, I wish I had bought a laptop in Taiwan. Shops were selling IBM / Lenovo X61 ThinkPad laptops (1 Gig RAM, 120 Gig HDD, Intel Core 2 Duo T7100 Santa Rosa processor, nice slim form factor) for TWD$38,000, which is equivalent to AUD$1330. I checked the price for the same model when I got back to Australia, and it’s locally around AUD$3100 (i.e. it was only 42% of the price in Taiwan). For that difference, I could quite happily learn to ignore the Chinese meta-characters on the keyboard.
  • Tip for people planning on attending Wikimania: take lots of business cards. I took about 15 (which I thought would be plenty), but actually about 30 would have been the right number.

And lastly a few photos:

Wikipedians buying fruit at the night markets:
Wikipedians buying fruit at the night markets

Whacky Taiwanese zoo sign: “It’s okay to wave at the Panda Bear, but riding him like a racehorse will make him sad.”
Whacky Taiwanese zoo sign: “It’s okay to wave at the Panda Bear, but riding him like a racehorse will make him sad.”

Yuri being pandered, after falling asleep in a massage chair.
Yuri being pandered, after falling asleep in a massage chair.

Lodewijk takes the Starbucks versus Starbucks-imitator challenge. Starbucks won.
Lodewijk takes the Starbucks versus Starbucks-imitator challenge.

Chaos as people tore the wiki ball apart. Sadly, nobody burst out of the ball.
Chaos as people tore the wiki ball apart. Sadly, nobody burst out of the ball.

Delphine and Cormac jokingly go head-to-head: Taiwanese toys at 50 paces!
Delphine and Cormac go head-to-head: Taiwanese toys at 50 paces!

Wikimania 2007: “Wikimedia Board Plenary Session”

Discussion notes: “Wikimedia Board Plenary Session”
Present: Frieda, Erik Moller, Jimmy Wales, Florence, Kat Walsh, Jean-Bart.

Normally 7 board members, 4 elected and 3 appointed.
Florence / Anthere joined in 2002 on EN and FR. Joined board in 2004. Chairperson.
Kat – Law student, US based, free culture.
Erik – from Berlin, Germany. Was journalist & writer. Concerned about managing potential of volunteers to do what they think is necessary. Most tech-focussed member.
Frieda – Italian, joined 2002, in 2005 was one the founders of the Italian Wikimedia chapter. Was elected a few weeks ago.

Q: Are you happy with the mix of non-elected and elected board members?
A: Most happy with having some appointed people. Erik would prefer all members to be elected.

Q: Print on demand. Any more details? Will we be able to get PDFs for free?
A: Want to make available to other wiki installations. Don’t want to get into specific details, want to avoid a vapor-ware situation. Working with a German company for this. [More info and some source code was released recently, several weeks after this session].

Q: How should people address the board? How do people communicate with the board?
A: “I dunno.” :-) Join the mailing lists, join the local chapters. Need the community members to push things to make them happen, because the board gets so overloaded. There are many things they would like to do, but they are only 7 people. Post-script: Contact: Cary Bass: User name: Bastique, has been hired by the foundation to be the volunteer co-ordinator.

Q: What do you think are some of the highest priority items?

  • Frieda: More communication.
  • Erik: Organising volunteers and co-ordination volunteers.
  • Jimmy: Building infrastructure of the organisation, organisational structure, to take advantage of opportunities.
  • Kat: Organisational structure, less crises.
  • Jean-Bart: Lowering the barriers to entry.
  • Ant/Florence: Usability, explaining our values.

Wikimania 2007 talk: “The Sharing Economy”

Talk by Joi Ito – “The Sharing Economy”.

Wants to have the .wiki TLD. Interested in the blogging revolution. Creative commons licenses.

We are very difficult as an audience! Very varied level of knowledge. Audience will argue back.

The Internet: “The Stupid Network” – David Isenberg (used in a paper); “Small pieces loosely joined” – David Weineberger (used in a book).

Internet created by small teams connecting together. When the Internet fails, it’s usually because of large companies and governments. The innovation is best when it is small pieces. Free software works because the cost of failure is very low – this allows ideas that would normally not be tried in large companies. E.g. Google was most exciting when it was small, and most innovative when it was small.

Professional versus Amateur:

  • The era of the professional had a large barrier to entry… We began to associate professional = good. Professionals are paid.
  • Amateur derives from the word “to love” – doing something because you love it. In English the perception is that amateur = low quality. There is a mismatch here.

Economists – think the more money you have, the happier you are. Utility function. The idea that you are trying to earn as much money as possible. People in the financial business associate intelligence with money. Yet often people are motivated more by factors that don’t have anything to do with money.

Also there is a difference between pleasure and happiness:


  • Drugs
  • Gambling
  • Winning lotto
  • Good weather

You can buy pleasure. We adjust to and quickly get used to pleasure.


  • More than enough is too much.
  • Comes from compassion – giving and receiving love.

You cannot buy happiness.

Amateurs make decisions based on happiness. Professionals maybe more based on pleasure.


  • Hollywood oversimplifies (Shock! Horror!). Pirates sometimes are not stealing; they are in fact fans that are addressing a need that is not being made (e.g localisations that don’t exist; movies which they simply cannot buy in their region).

There is a culture gap:

  • Remix – combining pre-existing material in new and interesting ways
  • E.g. the grey album (black album + the white album)
  • We are currently stifling free speech of remixing video.

Unregulated uses of things, like books – things can currently do:

  • Read
  • Sell
  • Sleep on a book
  • Burn a book
  • Fair use (tiny little grey zone)

Mostly only people who were copying a book and selling it were triggering copyright law.
Most uses of books are free, and unregulated.
Most regulated uses were commercial.

There is a big difference when we go from analogue to digital. Suddenly we are triggering copyright law all the time, because with digital material things get copied ALL the time, and indeed NEED to be copied to work. E.g. a DVD is copied into memory to be displayed; a web site’s pages are copied over the network and into memory when it is displayed in your browser. In the digital world, copies are EVERYWHERE and are REQUIRED for it to work.

Law + Technology = DRM.

Amazon is saying things like “let’s charge per page or per view”.

Now people are saying, “let’s charge second hand book stores”. It’s a creeping form of greed. Companies are not about happiness, they are about money.

There are also people who want “no rights reserved”

Creative commons is in between these two ends of the spectrum.

Creative commons = “Open source for content”. It’s also a “user interface for copyright”.

CC non-commercial no-derivatives is NOT a free license – how can you say it’s free when there are so many restrictions and regulations? – can take this into court – has a legal code. Can add metadata to material to indicate the license restrictions.

With the Internet, want it to be open – it’s GOOD that your enemies and people you don’t agree with can use the Internet – it makes it a much more interesting place and a more useful thing.

Ensemble. The importance of diversity. How a few negative comments can damage group cohesion. How to get different people with very different backgrounds to work together. A sense of unity and shared purpose. A lot of this ensemble gets lost in mailing lists and IRC and the minutia of day-to-day remote stuff – it’s why conferences and get-together are so valuable.

People who work in an online community care about the rules WAY more than people in companies. Most employees will never read their company rules, or comment on them. Studies have found that paradoxically, places with more rules tend to have happier people (because people know where they stand and what’s acceptable). (Personal comment: There has a be a limit to rules otherwise, like the real legal system, ordinary people will need specialists / lawyers to navigate all the rules. E.g. I’m personal slightly doubtful that more rules and guidelines would be constructive for the English Wikipedia).

Wikimania 2007 talk: “Visual identity and Visual Consistency of the Wikipedias”

Talk: “Visual identity and Visual Consistency of the Wikipedias” by Guillaume Paumier. Link to talk’s page.

Wikimedia names and brands.


  • One colour per project (e.g. wikibooks one colour for all languages that wikibooks supports)
  • Official name and localised motto and transcription.

Choosing good logos:

  • Good message
  • Good graphic quality
  • Fitting in the visual identity of the organisation. Pertaining to each other, yet unique.

Stats on other wikis:

  • Around 85% of wikis are like the Wikipedia’s visual identity.
  • In 74% of installations only the logo is different.
  • The default is MediaWiki skin is Monobook, so most MediaWiki wikis look like Wikimedia sites, because it’s the default.

Want a different visual identity for Wikimedia wikis. Want to change the default layout.
Showed some proposed mock-ups for skins, as a starting point for discussions.

Designing a new skin is a challenge. The skin currently is rather overloaded / cluttered – has many links, disclaimers, interwiki links. Want to simplify the UI. Only 4.6% of visitors are editors, and most of the links are relevant to editors. Want perhaps to have a way of switching between viewing and editing mode.

Question of usability. Site needs to work well for everyone. Clean and simple UI, with easy discoverability.

  • Nav bar between the projects.
  • More visible links
  • Thematic consistency
  • Pull down JavaScript menu for actions.

Q & A afterwards:
Q: Why no pull-down menus currently?
A: Client compatibility issues, what happens if JavaScript is not enabled.

Idea of maybe having a small Wikimedia logo on the Nav Bar, to establish that these sites under the Wikimedia umbrella.

Q: How to proceed?
A: Some usability testing by a uni / college could be useful.

Add an experimental non-default test skin. Would be okay to add this to MediaWiki core.

Wikimania 2007 talk: “Wikia”

“Wikia” talk, intro by Angela, some from Gil, majority from Jimmy Wales.

  • Wikipedia = non-profit reference material.
  • Wikia = Approaching very different material (e.g. gaming wikis; Wikis on really complicated & convoluted shows like “Lost”; POV / personal perspectives). For-profit.

I.e. A magazine focussed wiki instead of neutral reference-focussed wiki.

  • Licensing terms
  • Content censorship
  • Let the communities decide

Why would you use Wikia?

  • No installation or maintenance
  • Stable and reliable hosting
  • Large communities so you are not alone.
  • Single-sign-on.
  • New features. (Details?)

Gil – background is eBay.

  • ArmchairGM.
  • User pages – have pictures.
  • German Wikipedia has concept of “people I trust, and people I don’t”.
  • Can add friends, can track activity of friends and foes.
  • Can see the Geo-Location of other users.

People contribute for 2 reasons:

  • Social activity, make friends, etc.
  • Care passionately about the topic at hand.

Showed the Wikia WYSIWYG editor (Personal comment: I think the momentum on this has been lost because of the delay in releasing the source code externally; rather the momentum now seems to be behind FCKEditor).

Open or closed? Future of search:

  • Free access – search engine project – announced Dec 2006.
  • A lot of confusion about this.
  • How things are rated and ranked is proprietary. Considers Google to be a problem, since search has to be trusted. Building the stack (AKA the LAMP stack), and break out the similar requirements for search.

Open source search:

  • Public algorithms for ranking (i.e. the maths that lead to those results)
  • Make a crawl of the web publicly available.
  • Bought “grub”, a web crawler.

Search advertising is a proprietary biz model. There will be ads. Wants mechanisms to prevent Google-bombing. (Why? Large-scale distributed google-bombing is kind of rare, although SEO techniques for messing with keywords are painfully common…)

Opposed to Google’s yielding to the Great firewall of China to censor content.

Wikimania 2007 talk: “WikiHow case study”

“Challenges in a growing wiki : The wikiHow case study” by Jack Herrick (Wikihow founder). Link to talk’s page.

Things people do on the Internet, top 3, in decreasing order:

  • Research buying things
  • Read the news
  • Find out how to do things

Get good feedback from people using their site (showed some feedback messages).

Wikihow started Jan 15 2005 (Wikipedia was Jan 15 2001). By the end of May had:

  • 700 articles
  • 13 active editors

Travis is the engineer.

First article – “how to ride the elevator”. :-)

Jan 2006:

  • 9000 articles
  • 162 active editors.
  • Quality standards rising.

Jan 2007:

  • 15000 articles
  • 250 active editors
  • Existing admins redefine who should become an admin

June 2007:

  • 20000 articles
  • 522 active editors
  • 40 admins

Things want to improve in wikihow

  • more languages
  • more articles. Don’t have red links so much.

Doing okay on:

  • traffic
  • Equality on gender participation
  • Hand patrol every edit.

Wikihow works towards making editing wiki editing easier. Have tried to solve the licensing of uploaded images. Built a tool/extension that links into flickr. Shows the acceptably licensed images based on a search term, and with a click will upload these to Wikihow, and add all the right licensing information. Have written the extension for this. Would like to see this extension or something like it enabled on Wikimedia Commons. (Q: Is this extension in SVN? Update: thanks to Tgr – Yes, it’s the ImportFreeImages extension ).
Also use templates for entering information.

For-profit wikis: Good or evil?

  • Have advertising, which is the biz model & how they pay for everything.
  • Registered users = No ads
  • Anon users = Get ads, but they are minimal
  • This blended approach seems to work okay.

This is how Silicon Valley views user-generated content: Showed Evan P’s crowd sourcing slide.

Showed the wikihow biz philosophy:

  • A community service 1st, and a biz second.
  • Let go.
  • Go slow.
  • Build community.
  • Share it.

Other things:

  • Will try hiring someone for the German wikihow site, because it’s not growing organically. This is a seeding process.
  • Trying to create a for-profit company focussed on a public good.
  • If their biz model works, then it will be a good example for other wikis.
  • Wikihow is forkable. Right to fork both the source code ad the content of the wiki. Thinks this changes behaviour in a very positive way.
  • Have modified MediaWiki quite heavily. Source code available.
  • Have not faced any lawsuits yet, Lawyers have told them there are no problems until get much bigger. (e.g.: 10 times larger).
  • Q: do you feel any pressure to share profits? A: There are sites that pay people to create content currently. However does not think that paying volunteers will wreck the whole concept of wiki content creation. Totally different dynamic between work (I’m working on this to get my points to make a buck), versus “I am creating this for fun and for good”.
  • Interwiki linking is there, but it hardly every gets used.
  • Want to get a WSYIWIG editor happening. Thinks the Wikipedia is too hard to edit. E.g. only 10% of folks can figure out wiki syntax. But 100% of the people should be able to contribute. E.g. The WYSIWYG stuff from Wikia or FCKeditor. Will try both of these and see how they work.
  • Suggested license for starting a wiki: CC-BY-SA. Used a difference Creative Commons license and has sometimes slightly regretted this (used the non-commercial CC license?).

Wikimania 2007 talk: “Virtual and national cultures”

Talk: “Virtual and national cultures: Wikimedia, projects and organisation” by Delphine Ménard.

[If you’re watching the video of this, then skip forward a few minutes until the slides start working, because there was an A/V glitch at the start with getting the slides to display]

One of the funniest edit wars on the French Wikipedia was over “endive” (a plant) – because of “chicon” versus “endive” (two different names for the same thing – Endive in France and Chicon in Belgium). However the page now says “chicon” and “endive” all the way through.

Yoghurt versus yogurt – which spelling do we use?? On the English Wikipedia, seems to be a first come first serve approach – whoever writes the page first gets to determine what the page’s spelling is.

How much do real life cultures impact the Wikipedia?

The Spanish wikipedia is one of the few wikipedias that calls admins something else – namely “bibliotecarios” – which means “librarians”. Which kind of makes sense – like librarians, they keep the place clean, stop bad behaviour in the library, and make it welcoming to people, and help people when they need help.

The German Wikipedia banned “Fair Use” outright pretty early on (not recognised in the German legal system). English allows it. French is somewhere in-between these two perspectives.

[[Henry the Navigator]] – same article, on 3 different wikipedias. German: just the facts; Portugal: has facts, plus says he was good; The English one has a dispute about whether he was a homosexual or not!!

Village pump: In French this is called “Le Bistro” – the café – more informal.

Request for Admins – on the French, you CAN self-nominate. On the English one, it’s far less common to self-nominate.

How culture affects local Wikimedia organisations: Today we have 10 official chapters, in 3 continents. In the US: Do we have a US chapter, or a state-based chapters? Versus Argentina, that wants to have a country-wide organisation.

Q&A: Heard from Indonesian wikipedia: European conflicts that spill onto the Indonesian articles. E.g. conflicts over Geographical name of something (which comes first). Have to try to work together to find a solution that is acceptable to both sides.

Link to talk’s page.

Wikimania 2007 talk: “Enhancing wikis with social networking tools”

Talk: “Enhancing wikis with social networking tools” by Evan Prodromou (surname pronunciation guide: “Pro-Dro-Mo”, the “u” is silent)

WikiTravel founder

Keiki co-founder

Q: What makes wikis great? A: Cumulative effect – everyone can edit. Easy and adaptive. Progressively improvement. Get something close to consensus opinions.

…. but wikis can’t do everything! Don’t over apply wikis. Example the wikiclock – lets users update the current time, manually. Surprisingly accurate (only a few mins behind)

Wikis are not good for:

  • Automatable jobs

  • Personal opinion

  • “Protected” content & structured content

2 Types of wiki Communities:

  • community of practice (lets get something done)

  • community of interest (shared areas of interest)


  • 65% of WikiTravel users are engaged for 1 day or less (1 edit – never see again)

  • 95% for less than 1 month

  • This accounts for 70% of edits – so get lots of content from anon people.

Want to retain users, keep users engaged.

Features that work well with wikis:

  • Social networking

  • Blogs

  • Photo sharing

  • Forums

  • Social bookmarking

I.e. the whole web 2.0 playbook.

This works well, and complements wikis well.

Tagging / concepts / categories. Gives a way to associate wiki content with non-wiki content. Unity of content (wiki) versus a multitude of chaotic content (social side)

Don’t repeat yourself: People don’t want a new blog, or a new Flickr site. Want tagging. Done via RSS, FOAF, web APIs.

Can do various ways of gluing stuff together (Likes Drupal for example).

Case study #1: WikiTravel Extra.

  • Travel is very personal

  • Opinions

  • reviews! Want to capture these.

  • travel photos


  • Drupal

  • Shared login via OpenID authentication system (was already using at WikiTravel)

  • Lots and lots of plugins

  • Custom software and glue code to bring it all together

Showed WikiTravel Extra:

  • same logo & look and feel as WikiTravel

  • lots of different content and blog posts.

  • Have geographic forums

  • Have photo sharing

Results: Have had good feedback from Extra.

Case study #2: (the name is Hawaiian for “child”).

Decided to create a free content parenting guide, open to everyone, edited by everyone.

Even more that travellers, parents like to share their experiences.

There are privacy concerns (e.g. sharing kids photos)

Have no existing wiki community. Need to focus more on content production, less on the wiki community.

Results gave a demo. Just opened today! Launching in French and English.


Q: Will be made into a paper book?

A: Hopefully, yes.

Link to talk’s page.

Wikimania 2007 talk: “MediaWiki API”

Talk: “MediaWiki API” by Yuri Astrakhan

Summary: “It’s all about the data, stupid!”

Yurikbot: More than 3 million edits.

The API adds a new layer of access. Allows new clients to access data (e.g. JavaScript browser extensions, Standalone apps like vandal fighter, data gathering for researchers).

Does not use the HTML rendering code.

Example of this; The navigation pop ups extension – this uses the API (because it is faster).

Current situation:

  • We have login

  • Can query existing data

  • Multiple output formats.


  • Change data (this is currently in development by the Spanish Vodafone folks, among others)

API is very modular – can add things, and they will just plug in.

Can get some of the following:

  • Page information

  • Lists (e.g. list of backlinks to a page)

  • Metadata (servers)

  • Can get multiple types of information all in one query

  • Conveniences to avoid gotchas – e.g. normalisation, resolving redirects, etc.

(gave some demos of these on the live Wikipedia).

Tries to be very quick – only does the work that you ask it for.

Gave API performance figures: API gets 2% of Wikimedia site hits. Yet uses 0.1% of CPU load. I.e. API is 20 times more efficient than the main UI. However at the moment is a read-only API, so would expected better performance that the main UI (which also has to do writes).

Most of the hits are coming from the Open Search module currently.

Future: Want to add unit tests for the API (Note from me: it’s coming, but don’t hold your breath!).

Asked API users to use GZIP compression when calling it – added recently to PyWiki – this saves lots of bandwidth.

In the Q&A time, I asked the Vodafone folks about the API write capacity they’re working on, and very roughly when they thought it might be ready: Their estimate: Should be ready around Christmas (i.e. 4.5 months). They want it for their mobile phone customers, who want to be able to modify the Wikipedia, as well as view it on their mobile phones.

Demo time: Showed some examples of some API code (e.g. added a simple module, showed how to add new supported formats to the API).

Link to talk’s page.

Wikimania 2007 talk: “Special:Contributions/newbies: User study of newly registered user behaviour”

Talk: “Special:Contributions/newbies: User study of newly registered user behaviour” by Brianna Laugher. User name : pfctdayelise (pronounced as: “perfect day elise”). Most active on Commons.

Why are we interested in what new users do? New users are really important to community growth. Wikis have to keep growing – they grow or they die.

Interested in the English Wikipedia as it is the trailblazer, and it’s the first point of contact for many people with wikis (so still want a good impression for wikis in general).

Why do new users sign up? Do they think it’s a social site? Do they want to add articles about their employers?

What were other people’s experiences? How did other people interact with you when you first join? When did you first feel that you were part of the wiki community?

The attendees of Wikimania are the success stories.

Did a study of 1 day’s worth of new users. Picked a random day, and just observed all users on EN who registered on that day. What happened to those new users?


  • Deleted edits are not available through the MediaWiki API (would like this; would also like to be able to get a diff of an edit via the API, which currently is not possible).

The day: Feb 1st 2007. 10641 users signed up. This is a fairly typical number.

Showed some of the usernames:

Some were crazy, and you could be sure that they were not coming back.

Not every account represents one person. For example, there were a lot of Stephen Colbert usernames!

Total number of edits that they made:

7000 of the users out of 10000 had zero edits (i.e. 70%). They did nothing with their account.

How many edits until you are part of the community? 50? 100? 1000?

Only 5% of user made more than 10 edits.

Images uploaded by these users:

1329 images uploaded. 40% of these images were deleted.

Some very troubling cases:

  • 58 images uploaded. Of these 57 were deleted. Seems soul destroying for these people – it’s a lot of effort to upload images. Suspects that the majority of these issues are copyright problems.

Wikipedia = social networking site? To test this looked at talk page. 14% of the sample users had a talk page. Had a look at the edit summaries to the talk pages of these users:

  • 14% of the people got a page deletion notice (e.g. you added something not notable).

  • 36% got a vandalism warning.

  • 19% got an image deletion message.

  • 20% got a welcome message.

  • Some people left a talk message to themselves.

Some people seemed to use Wikipedia as a help lifeline. Kind of disturbing messages about people asking for help with bad domestic situations.

Community dynamics:

  • “All the low-hanging fruit has been picked” – Andrew Lih

  • Backlogs

Discussion of user warnings that we show to people:

“Don’t bite the newbies”. We have template warnings. They are pre-recorded messages. They are officious. It’s not people talking to other people. Templates use the royal “we”.

{{uw-test-2}} –> {{uw-test-3}} –> {{uw-test-4}} –> blocking

Not good system for socialisation and introducing people to Wikipedia and what we are about.

Are new users potential Wikipedians, or are they just pests mucking stuff up?

[[Wikipedia: WikiProject user warnings]]. It would be nicer to have [[Wikipedia: WikiProject user socialization]] – instead of scalding people, would be good to socialise them.

What are the goals of Wikipedia?

New users DO muck stuff up. How do we reward good behaviour and encourage people?

There is no page that identities new content adding users (not just spelling or formatting changes), especially by new users, so they can be encouraged and rewarded. Should we add this?

Everyone has something valuable thing to share. The hard thing is find what that is, and get them to share that stuff.

There is a wiki ethos.

“Laugher’s Law”: If you are going to act as is X is not allowed (existing social restriction), you may as well stop letting people do X (introduce a technical restriction).


  • Change the community attitude.

  • Recognise and reward and promote GOOD behaviour, rather than punishing BAD behaviour.

We treat too much stuff as vandalism. Sometimes people are just confused, or don’t know what to do. Not all people are vandals that we currently call vandalism. E.g. blanking a page can be someone who really knows their stuff (but doesn’t know wikis), and knows that the page as presented is just wrong – so they blank it out.

People being templated to death is due to admins being overloaded. It’s why they were created.

Link to talk’s page.