« June 2006 | Main | August 2006 »

July 20, 2006

Aggademia

Last week on Inside Higher Ed there was an article by Scott McLemee describing his wishlist for an 'academic aggregator' that would both a) collate new information from the web of academic publishing into one place and b) allow people to build directories of links on academic topics.

The follow-up discussion on Crooked Timber and elsewhere (Long Sunday, Epiphatic Exhaustion) seemed to agree this would be a useful idea to put into practice and—coincidentally—it was along the same lines as something I'd been playing with already. To see if it's really an idea that could work, I've put a test site online: Aggademia [update: the site's down for now - it should be back in a better form at some point].

Using Aggademia, you can

  1. browse through blog posts (aggregated from most of the top 50 popular science blogs, as published in Nature last week) and vote on items (up/down or 1-5: either works, but only the up/down votes are counted in the 'popular' list at the moment) and
  2. create or join existing groups on particular topics. The owner of a group can edit the official list of links for that topic (using the 'Related Links' box in the 'edit' tab for the group), while other members of the group can make suggestions for useful topic-related links (using 'create weblink' in the group sidebar). To join a group, click 'subscribe' in the group sidebar on the right-hand side.

The aggregation and voting are still very basic and need lots of extra features to be really useful, but this was just a quick experiment to explore what could be done. Comments and suggestions are welcome either in this post or on your own weblog.

As this is my first post, an introduction: I'm Alf Eaton and I started working with Nature's Web Publishing group recently, where I will be experimenting with social software and information management tools for scientists. Previously I have made contributions to online systems for publishing scientific articles, as well as creating the biomedical literature search site HubMed.

July 15, 2006

Article on the Nature Podcast

The July issue of Serials contains an article by me about the Nature Podcast. It also forms part of their Key Issues collection.

I wrote it a couple of months ago, so the facts and figures aren't bang up to date, but my views are pretty much unchanged. And even if you don't enjoy the text, you can always have a good laugh at the photo. ;)

July 13, 2006

Cory Ondrejka visits Nature

On Tuesday we were lucky enough to play host to Cory Ondrejka, CTO of Linden Lab, creators of the amazing virtual world, Second Life. Anyone who's heard me talk in the last year or so about important technological trends has heard me wax lyrical about Second Life. But this time my colleagues got to hear about it from one of the guys who's actually building it. My rough notes are below. There's also some more cogent coverage from Sara Lloyd, guesting on Richard Charkin's blog.

Cory Ondrejka

Online games are seeing exponential growth. Over half of Americans describe themselves as 'game players'. Their average age is rising by about one year per year, so it's becoming more mainstream. It's also a big business, and trade within games is 10-20 times the amount of 'external trade'.

Second Life (SL), however, is not a game. There is no imposed storyline. It's more like Neal Stephenson's Snow Crash.

Second Life is also not a product, it's a platform. The business model is not based on subscriptions. Instead, users pay for land -- anywhere between US$0 and US$30,000 a month, with the people at the high end running in-world businesses.

The land unit is the 'sim' = 16 acres. SL is now approaching 3,000 sims, which is about 3 times the size of Manhattan.

In June 2006 there were 100k users, 340k items worth US$6.5m sold. US$1.5m worth of goods were traded directly between users (without conversion to virtual or real currency). The distribution of transactions by size is a power law, so a few people are trading large amounts. This economy is growing rapidly.

Hernando de Soto, a Peruvian economist, wrote The Mystery of Capital about the importance if intellectual property (IP) in economic development. Tringo is a game that took off in SL (so much so that people complained about it's in-world social effects). IP rights in SL are transferable to the real world, so the inventor licensed them. Tringo will soon be turned into a UK TV game show.

Neil Protagonist is a SL resident who built a whole city devoted to Japanese anime. It took him about 6 months and everything in it is for sale. His real-world wife, SL resident Nephilaine Protagonist, is a graphic artist who makes money in-world as a fashion designer.

On the web, user creation has been mainly limited to text. It's also mainly solo and sequential in nature. Only a very small proportion of users add content to the systems (e.g., ~4% for eBay or the web as a whole, 0.02% for Wikipedia, even lower for Linux). Even though SL is quite hard to add things to, about 2/3 of users create something in any given month.

What are they doing? Mainly 'atomistic construction' -- building stuff from scratch. [Cory showed examples of a UFO and piano.]

SL gets about 140,000 hours of use per day, around 1/4 of which is spent creating things. This is equivalent to 17.5 user-years per day, or a staff that would cost about $650m a year.

In any 7 days, about 15% of residents write code (in a rather difficult language similar to C). The monthly total is equivalent to the amount of code in MS Office.

Some users staged alien abductions, but only infrequently so other people didn't take the reports seriously. ;) This is a good example of completely unpredictable uses of SL.

There was also an SL tax revolt. Users set things -- and themselves, and newly arrived users -- on fire in order to protest against the economic model being used by Linden Lab. The model was changed as a result.

Residents are not all geeks. By hours of use, it's almost exactly 50/50 male/female; the median age is about 33. The older you are, the more likely you are to continue using SL. Women are more likely to continue using SL than men. Most SL users are not 'game players'. 75% are in the US.

SL contains 100s of millions of user creations, 10s of millions of avatars, and millions of scripts, many of which are shared.

Sports examples: golf, baseball, tennis (an app that replays shots from Wimbledon games). The SL Yacht Club is one year old. (Slow-moving vehicles work well, and the water is pretty.)

Simulations and training: Used for first-responder training for Dept of Security in the US. Also for working out the logistics of storing emergency medical supplies. Simulation of schizophrenia hallucinations help doctors and families better understand the disorder; this took one user a month to create. 'Shockproof' is an area built by stroke survivors for 'cognitive recovery'. (Engaging your brain is very important following stroke.)

Anshe Chung is a real-estate baron in SL. She buys virtual property from Linden Lab, makes it look better, divides up the land and sells it on. As a result of the publicity she's received, she now has a lot of competition.

SL is full of amateur-amateur education. E.g., the Black Library, which invites writers to come and add to the library (and the bar helps attract them).

Skydiving is also popular. The people who first popularised it make money selling parachutes.

SL helps to reduce the cost of learning by enabling:
- Legitimate peripheral participation (i.e., watching experts)
- Situated learning (i.e., providing context)
- Heterogeneous learning (i.e., providing variety)
- On-demand learning

Last semester, 17 universities used SL for distance learning. These activity are continuing to grow. 50 universities own land in SL for research or other purposes. The 'International Space Flight Museum' was built by NASA engineers in the their spare time and contains (virtual representations of) all the NASA manned launch vehicles.

Virtual classrooms. The 'hypercello' allows composition of music. Drawing classes in SL that allow people to draw in the real world, upload the images to SL and discuss them with (virtual) classmates.

Exploratorium: A live in-world web cast of the last solar eclipse was accompanied by a permanent virtual display explaining the solar system.

Publishing: The 'Metaverse Messenger' is perhaps the most popular in-world weekly newspaper. O'Reilly's 'MAKE:' magazine also has a presence in SL. Phil Torrone, the editor, has a (real-world) watch that tells him when someone arrives on their (virtual) land so that he knows to log on.

SL Community Convention: 6 people arranged a real-world conference, with all the planning taking place in SL. The conference itself was streamed by video into SL. One user was present in the real world and in SL!

SL movie theatres allow people to watch movies together and talk about them (which you can't do with web streaming).

The Beyond Broadcast Conference (held by the Berkman Center at Harvard) took place in the real world and SL simultaneously. SL participants could ask questions of real-world speakers and could chat about the conference afterwards with other SL residents.

Play versus work is a false dichotomy: One user played live music in SL and put out a hat for tips to make money.

An international AI experiment involved participants creating their own programmed virtual fish and adding them to a collective shoal.

The rules are different for bits than atoms, and amateurs are better at appreciating this than experts are.

There is a recent Harvard Business Review article about marketing to avatars (N.B. not merely using virtual worlds to market in the real world). The conference organised afterwards to discuss the piece was done in SL. This got participants to understand much better the qualities of virtual worlds.

Cory Doctorow held an in-world book signing. He releases his books under Creative Commons licences. He challenged SL users to make in-world book readers (i.e., applications that enabled his books to be read in SL). People queued up to get Cory to sign their books.

Larry Lessig did something similar. Interestingly, people who met Larry in SL, talk about actually meeting him (not talking with him via SL). People who see pictures of their avatars consider them to be pictures of themselves. Avatars look like people (generally!) and are therefore conceptualised as people.

Examples of clothing and music stores in SL run by real-world companies, presumably as marketing tools. The BBC has adopted it too. For example, Newsnight broadcast from within SL and so did the One Big Weekend concert.

In short, SL looks like the web in 1994-5, with early adopters starting to do interesting stuff and some others scratching their heads in bemusement.

July 02, 2006

Peer Review Debate concludes, Peer Review Trial gets interesting

The final batch of contributions to Nature's peer review debate were published last week (while I was away at the Databasing the Brain workshop in Oslo, so excuse my tardiness). It's another fine set, but I have to single out the contribution from Charles Jennings, a former Nature colleague of mine. Charles is open-minded, analytical and articulate, attributes that every scientist should have, but which many of us fall short of. His opening sentence sets the tone:

Whether there is any such thing as a paper so bad that it cannot be published in any peer reviewed journal is debatable.

This is a very important point: peer review is not one thing but a whole collection of different practices. To some journals it means sending the manuscript for detailed review by three or more independent experts, to others it means an internal editor reading it through for obvious howlers. Charles goes on to argue — quite convincingly, I think — that the primary purpose of peer review is not to ensure that bad papers don't get published, but that each paper gets published in the journal that it deserves (and, furthermore, that this serves a useful purpose). He then proposes ways in which the various costs and benefits of peer review could be analysed. It's brilliant stuff so read it.

Meanwhile over at the peer review trial, it's good to see some really meaty comments coming in. (As I write, this seems to be the best example, but manuscripts are coming and going all the time, so that will change). I'm not knowledgeable enough to say how useful these comments are to the review process, but it's interesting to see the conversation (perhaps) starting to take off.

"Nascent Web publishing efforts have their genesis in a burning need to say something, but their ultimate success comes from people wanting to listen, needing to hear each other’s voices, and answering in kind."
Rick Levine
The Cluetrain Manifesto

Subscribe

Subscribe to this blog's feeds:

[What is this?]

Recent Comments

Powered by
Movable Type 3.2