Post Prediction Markets

I support advertising on Wikipedia

Tuesday, January 2nd, 2007

Wikimedia Foundation is over halfway through a . I hope that when you give you leave the following public comment:

I support advertising on Wikipedia.

Evan Prodromou summarizes a completely unwarranted controversy regarding a matching fund (bottom of page):

All fine so far, right? But a small logo in the donations notice — seen by non-logged-in users on every page of every WMF site — was considered by many Wikipedians and other WMF editors as dangerously close to the line on advertising — or over it. There have been several prominent users who have left the project because of it.

I’m not sympathetic with these folks; in fact, I’m in solid opposition. I think that Wikipedia’s huge amount of Web traffic is a resource that the Foundation is squandering. Traffic like Wikipedia’s is worth tens of millions if not hundreds of millions of dollars in ad revenue per year. That’s money that could go to disseminate free (libre and gratis) paperback pocket encyclopedias to millions of schools and millions of children, in their own language, around the world.

It’s irresponsible to abuse that opportunity.

I strongly agree and will repeat exactly what I said during last year’s Wikimedia fund drive:

Wikipedia chief considers taking ads (via Boing Boing) says that at current traffic levels, Wikipedia could generate hundreds of millions of dollars a year by running ads. There are strong objections to running ads from the community, but that is a staggering number for a tiny nonprofit, an annual amount that would be surpassed only by the wealthiest foundations. It could fund a staggering Wikimedia Foundation bureaucracy, or it could fund additional free knowledge projects. Wikipedia founder Jimmy Wales has asked what will be free. Would an annual hundred million dollar budget increase the odds of those predictions? One way to find out before actually trying.

In somewhat related news, Mozilla just reported 2005 financial information, showing 800% revenue growth:

In 2005 the Mozilla Foundation and Mozilla Corporation combined had revenue from all sources of $52.9M. $29.8M of this was associated with the Foundation (both before and after the creation of the Corporation). The bulk of this revenue was related to our search engine relationships, with the remainder coming from a combination of contributions, sales from the Mozilla store, interest income, and other sources. These figures compare with 2003 and 2004 revenues of $2.4M and $5.8M respectively, and reflect the tremendous growth in the popularity of Firefox after its launch in November 2004.

The combined expenses of the Mozilla Foundation and Corporation were approximately $8.2M in 2005, of which approximately $3M was associated with the Foundation. By far the biggest portion of these expenses went to support the large and growing group of people dedicated to creating and promoting Firefox, Thunderbird, and other Mozilla open source products and technologies. The rate of expenses increased over the year as new employees came on board. The unspent revenue provides a reserve fund that allows the Mozilla Foundation flexibility and long term stability.

An advertising-fueled Wikimedia Foundation could fund dozens of much needed Mozilla Firefox sized projects. And many Creative Commons (which just successfully completed its much more modest annual funding campaign) initiatives. :)

Update: Welcome Slashdot readers. The major objection to ads on Wikipedia takes two forms:

  • Advertising is profane.
  • Advertising would compromose Wikipedia’s neutrality.

A common response to the first is that those who don’t like ads can run an ad blocker. Easier still, those who don’t like ads can log in — there’s little reason to display ads to logged in users, who probably generate a tiny fraction of pageviews. But I don’t think either of these responses will satisfy this form of the objection, as it is basically emotional. Some people object to the knowledge that ads exist, even if not experienced personally. I suppose these people don’t use search engines. It’s a wonder they can stand to use the net at all. I discount them completely.

The second is completely unrealistic. How would third party text ads, e.g., via AdSense, compromise neutrality? There’s simply no vector for an advertiser to demand changes and zero reason for Wikipedians to comply. Wikipedia is not a small town newspaper beholden to the local department store, not even close. It isn’t even Slashdot, which as far as I can tell has not been compromised by years of running ads. To people with this objection: show me a community site that has gone astray due to advertiser influence.

Sponsors, “being managed by Wikipedia staff (like in newspaper ads, i.e. no uncontrolled 3rd party feeds)”, as suggested by Kuba Ober, are far more dangerous than third party ads, because then there is a vector between advertiser and someone with power at Wikipedia.

There may be an opportunity for Wikipedia to completely rethink and remake advertising, or merely compete in some fashion with what some are calling Google’s near monopoly, but now it would make tremendous sense to use AdSense or Yahoo! or both — and I suspect Wikipedia could manage to keep a greater share of revenue than a normal web publisher. Rick Yorgason mocked up what AdSense would look like in the place of the current fundraiser’s donation banner.

Slashdot commenter jklooserman summarizes objections from Wikiproject no ads:

  1. Wikipedia’s philosophy is non-commercial
  2. Ads put at risk Wikipedia’s principle of Neutral Point of View (NPOV)
  3. The information that constitutes Wikipedia is wealth for the community

I don’t see “non-commercial” in any form on the Wikimedia Foundation home page. I do see this, in large text:

Imagine a world in which every single human being can freely share in the sum of all knowledge. That’s our commitment.

The next line, all bold, asks for help in the form of donations.

Much more money, hundreds of millions, would speed the arrival of that world and fulfillment of that commitment.

As above, there is no realistic scenario for ads undermining neutrality on Wikipedia.

The third objection strikes me as a non-sequitur. In any case, the point of obtaining more resources would be to increase the wealth of the community — of all human beings.

jklooserman also pointed out that there’s a category of Wikipedians who think that the Wikimedia Foundation should use advertising. Add it to your user page if you agree.

Posted in Advertising, Creative Commons, Economics, Open Source, Peeves, Prediction Markets, Public Goods, Wikipedia | 101 Comments »

Dangerous Optimism

Monday, January 1st, 2007

I never got around to commenting on responses to the 2006 Edge Annual Question — “What Is Your Dangerous Idea?” — as most were uninteresting, not dangerous, or simply lame.

It must’ve been the question, as this year’s responses — to “What Are You Optimistic About? Why?” — make for good reading. I’ll excerpt a few that resonate with themes I go on about.

Steven Pinker on The Decline of Violence:

Even the mass murders of the twentieth century in Europe, China, and the Soviet Union probably killed a smaller proportion of the population than a typical hunter-gatherer feud or biblical conquest. The world’s population has exploded, and wars and killings are scrutinized and documented, so we are more aware of violence, even when it may be statistically less extensive.
…
My optimism lies in the hope that the decline of force over the centuries is a real phenomenon, that is the product of systematic forces that will continue to operate, and that we can identify those forces and perhaps concentrate and bottle them.

James O’Donnell says Scientific Discoveries Are Surprisingly Durable:

But the study of the past and its follies and failures reveals one surprising ground for optimism. In the long run, the idiots are overthrown or at least they die. On the other hand, creativity and achievement are unique, exciting, liberating—and abiding. The discoveries of scientists, the inventions of engineers, the advances in the civility of human behavior are surprisingly durable.

Clay Shirky on Evidence:

We will see a gradual spread of things like evidence-based politics and law — what is the evidence that this expenditure, or that proposed bill, will have the predicted result? The expectation that evidence can answer questions about the structure of society will discomfit every form of government that relies on sacrosanct beliefs. Theocracy and communism are different in many ways, but they share the same central bug — they are based on some set of assertions that must remain beyond question.

Jamshed Bharucha on The Globalization of Higher Education:

We are all better off when talent is realized to its fullest—even if it crosses borders.

I didn’t count, but I think the subject mentioned most often was climate change, with solar power as the thing most were optimistic about. My favorite take on climate change was Gregory Benford on Save The Arctic:

So: despair? Not at all. Certainly we should accept the possibility that anthropogenic carbon emissions could trigger a climactic tripping point, such as interruption of the gulf stream in the Atlantic. But rather than urging only an all out effort to shrink the human atmospheric-carbon footprint, my collaborators and I propose relatively low tech and low expense experiments at changing the climate on purpose instead of by mistake.

If we understand climate well enough to predict that global warming will be a problem, then we understand it well enough to address the problem by direct means.

There are also several good entries on health, life extension, and also networks-will-change-publishing — but my, isn’t the last relatively boring?

One last favorite, on human enhancement, Andy Clark on The End Of The ‘Natural’:

Second, the biological brain is itself populated by a vast number of hidden ‘zombie processes’ that underpin the skills and capacities upon which successful behavior depends. There are, for example, a plethora of such unconscious processes involved in activities from grasping an object all the way to the flashes of insight that characterize much daily skilful problem-solving. Technology and drug based enhancements add, to that standard mix, still more processes whose basic operating principles are not available for conscious inspection and control. The patient using a brain-computer interface to control a wheelchair will not typically know just how it all works, or be able to reconfigure the interface or software at will. But in this respect too, the new equipment is simply on a par with much of the old.
…
In sum, I am optimistic that we will soon see the end of those over-used, and mostly ad hoc, appeals to the ‘natural’. May we all have a thoroughly unnatural New Year.

A highly agreeable toast.

…

Many of the responses contain very rough predictions, reminding me of prediction registries, an idea Robin Hanson has said would obtain 80% of the benefits of prediction markets (I doubt the number is that high) and also promoted by David Brin. I think prediction markets and registries are almost entirely complementary.

I like Brin’s point that “One advantage of registries is that they can be involuntary.” A pundit can only avoid inclusion by effectively not making predictions (which may include being wishy-washy and imprecise). I conjecture that DiscourseDB (I mentioned previously) is a model of what a prediction registry would look like — just imagine cataloging “will” rather than “should” opinions, and add evaluation.

…

I’m surprised that none of the responses (I could have missed one) took the (unintended?) bait offered by combining the 2006 and 2007 questions: Is optimisim dangerous?

That depends on the subject of optimism. I think people tend to be dangerously optimisic about the outcomes of authoritarian processes, including both obvious societywide authoritarianism and conscious decisions made by individuals, but dangerously pessimistic about decentralized processes, including listening to external advice at the individual level.

Via Boing Boing, Marginal Revolution, or EconLog, all of which appeared in one batch of feed updates.

Posted in Apartheid, Health, Peeves, Politics, Prediction Markets | 1 Comment »

Most important software project

Sunday, December 10th, 2006

I don’t have a whole lot more to say about Semantic MediaWiki than I said over a year ago. The summary is to turn the universal encyclopedia into the universal database while simultaneously improving the quality of the encyclopedia.

Flip through Denny Vrandecic’s recent presentations on Semantic MediaWiki (a smaller pdf version not directly linked in that post). There’s some technical content, but flip past that and you should still get the idea and be very excited.

I predict that Semantic MediaWiki also will be the killer application for the Semantic Web that so many have been skeptical of.

Yaron Koren also says that Semantic MediaWiki is “the technology that will revolutionize the web” and has built DiscourseDB using the software. DiscourseDB catalogs political opinion pieces. Koren’s post on aggregating analysis using DiscouseDB. Unsurprisingly this analysis shows the political experts making bad calls.

Koren also has created Betocracy, another play money prediction market where users create claims. It looks like Betocracy is going for a blog-like interface, but I can’t say more as registration obtains a database error.

One prediction market and Semantic MediaWiki connection is that making data more accessible makes prediction markets more feasible. Obtaining data necessary to create and judge prediction market contracts is expensive.

On that note Swivel also looks interesting. Some have called it data porn. Speaking of porn, see Strange Maps.

Posted in Open Source, Prediction Markets, Semantic Web, Wikipedia | 13 Comments »

Iraq withdrawal and civilian casualties

Saturday, November 25th, 2006

I don’t follow Iraq closely, but recent headlines seem to indicate a turn for the worse and that withdrawal of U.S. troops is now on the table.

It should not have been difficult to predict that invasion would turn out badly, but politicians make the same mistakes (less charitably–tell the same lies) repeatedly, in particular when it comes to war (one reason why).

Among all the tragedies of the Iraq war, a small one is that there was no set of conditional prediction markets to consensus check (an analogue of “fact check”?) likely outcomes. An arbitrary expert can always be countered with another arbitrary expert. The nice thing about prediction markets here is that they converge to a single consensus probability (or set of interlinked probabilities for a set of claims) given the possibility of arbitrage. Faced with a market that says what a politician wants to do will probably have ill effects, the politician can ignore the consensus, but can’t counter it will an equivalent, as can be done with any expert.

So should the U.S. withdraw its military from Iraq? Unfortunately I do not know of a conditional market set up to guess the impact. Iraq-related markets I found:

Unfortunately all of these are play money markets and all only concern U.S. troops. What about Iraqi civil war or economic performance? Fortunately we can use one of these markets as an input for a conditional market that attempts to guess the impact of withdrawal on Iraq. I used the second, as it maps directly to a probability, unlike the first, and is not deemed to be an incredibly long shot, unlike the third.

The Iraqi Body Count currently says a lower bound of 47,781 Iraqi civilians have been killed since the invasion. I assume if that lower bound moves to 100,000 or greater by the end of 2007, a civil war has occurred or is in progress.

So I set up Iraq withdrawal and civilian casualties on Inkling, with four stocks:

  • USLEAV07 true AND >= 100k IBC EOY 2007
  • USLEAV07 true AND < 100k IBC EOY 2007
  • USLEAV07 false AND >= 100k IBC EOY 2007
  • USLEAV07 false AND < 100k IBC EOY 2007

I set the intial price of the first two at 12 each and the second two at 38 each, reflecting the 24 percent chance of substantial troop reduction given by Newsfutures traders and a 50/50 chance of civil war (I don’t know of a probability source for the latter). In theory prices should move to whatever traders think the probabilities actually are regardless of their initial settings.

There are two major problems with this experiment. First, a spike in violence may make troop reductions more (or less) likely, which makes it harder to divine the impact of troop reductions on violence.

Second, Inkling markets are sometimes at great variance with others or common sense, e.g., Hilary Clinton is given a 28 pecent chance of winning the 2008 Democratic nomination, others have her around 50 percent.

I surmise that there is something wrong with Inkling. That something could be just that it has no users. I set up this experiment on Inkling because it was trivial to do so, but I’d really like to see Tradesports/Intrade set up real money contracts along these lines.

Update: The first problem can be removed by ignoring deaths through April 2007. I will create a new market reflecting this…

Iraq withdrawal and civilian casualties (improved) is running with the following stocks:

  • USLEAV07 true AND IBC >= 40k May-Dec07
  • USLEAV07 true AND IBC < 40k May-Dec07
  • USLEAV07 false AND IBC >= 40k May-Dec07
  • USLEAV07 false AND IBC < 40k May-Dec07

Update 20061127: The improved market is now actually running, was previously held for admin approval.

Update 20061211: Followup posted at Midas Oracle.

Posted in Economics, Iraq, Peeves, Politics, Prediction Markets | 4 Comments »

GPL Java

Monday, November 13th, 2006

Sun announced today that it is releasing all of the critical pieces of the Java platform under the GPL. This is fantastic news, as a huge number of important and exciting projects are built on the Java platform and now they can be completely free as in free software. Read Tim Bray on the announcement and lots more blog commentary via Tailrank.

This should have happened years ago but as of yesterday it happened sooner than I expected. I set up a play money prediction market on Inkling (the first of two) asking whether Java would be open sourced by the end of this year. The price slowly declined from 60 in May to 20 in late October, then spiked to 70, with a last trade at 81.76 this morning.

I judged the contract at 100, but probably shouldn’t have — much of the code won’t be released until early next year. Oops. Good thing Inkling markets are play money and zero oversight, or Chris F. Masse would rightly castigate me.

Posted in Open Source, Prediction Markets, Programming | 1 Comment »

Electability predictions

Saturday, November 11th, 2006

On rare occasions interrupting an activity to check feed subscriptions saves time, e.g., when the activity is writing a blog post on electability implied by prediction market contracts for nomination and election, and a post by David Schneider-Joseph on Nominatibility and Electability shows up that says most of what I wanted to say:

The fact that, in the real world, 2008.PRES.GIULIANI divided by 2008.GOP.NOM.GIULIANI happens to equal 72.2 simply means that, in those scenarios where Giuliani actually ends up being nominated, his electability averages 72.2. But his abstract electability, given the hypothetical scenario in which the Republican Party nominated him without considering other candidates, is not necessarily the same.

This far out one should not read too much into electability implied by prediction market price ratios, but they’ll be interesting to follow anyway, and on primary election or caucus days, and the eve of nomination even moreso, a power-hungry partisan would do well to pay heed (at that point scenarios where candidate A versus B gets nominated differ little excepting that candidate A or B gets nominated).

Of course I’d really like to see a party that nominates the candidate whose nomination is predicted to best further outcomes preferred by the party — platform as a utility function — nomination by . If a party’s preferred policies are not predicted to lead to a party’s preferred outcomes, a futarchist nominating process could lead to the nomination of the candidate most likely to lose!

Posted in Blogs, Politics, Prediction Markets | 3 Comments »

Defeatist dreaming

Sunday, October 22nd, 2006

Jimmy Wales of Wikipedia says to dream a little:

Imagine there existed a budget of $100 million to purchase copyrights to be made available under a free license. What would you like to see purchased and released under a free license?

…

I was recently asked this question by someone who is potentially in a position to make this happen, and he wanted to know what we need, what we dream of, that we can’t accomplish on our own, or that we would expect to take a long time to accomplish on our own.

One shouldn’t look a gift horse in the mouth and this could do a great deal of good, particularly if the conditions “can’t accomplish on our own…” are stringently adhered to.

However, this is a blog and I’m going to complain.

Don’t fork over money to the copyright industry! This is defeatist and exhibits static world thinking.

$100 million could fund a huge amount of new free content, free software, free infrastructure and supporting institutions, begetting more of the same.

But if I were a donor with $100 million to give I’d try really hard to quantify my goals and predict the most impactful spending toward those goals. I’ll just repeat a paragraph from last December 30, Outsourcing charity … to Wikipedia:

Wikipedia chief considers taking ads (via Boing Boing) says that at current traffic levels, Wikipedia could generate hundreds of millions of dollars a year by running ads. There are strong objections to running ads from the community, but that is a staggering number for a tiny nonprofit, an annual amount that would be surpassed only by the wealthiest foundations. It could fund a staggering Wikimedia Foundation bureaucracy, or it could fund additional free knowledge projects. Wikipedia founder Jimmy Wales has asked what will be free. Would an annual hundred million dollar budget increase the odds of those predictions? One way to find out before actually trying.

Via Boing Boing via /.

Posted in Blogs, Creative Commons, Economics, Intellectual Protectionism, Open Source, P2P, Peeves, Prediction Markets, Public Goods, Wikipedia | 6 Comments »

Prediction market aggregator

Sunday, September 24th, 2006

Chris F. Masse points out Smartcrowd, a blog that gathers prices from several markets as the primary component of its commentary. I’d really like to see a service that only gathers prices for related contracts from several markets in an automated fashion, but Smartcrowd’s apparently manual index on GOP control of the U.S. House is a useful start.

Masse’s summary and comment on U.S. House control contracts are contradictory:

[real-money political prediction markets predict a GOP-controlled House while play-money political prediction markets predict a Dem-controlled House.]

So the crowds at Casual Observer and Newsfutures currently favour Democrats to win the House of Representatives, while the crowds at Tradesports and WSX suggest the Republicans will retain control of the House of Reps.

But WSX is a play-money market.

Aggregation should highlight a problem with play-money markets — play money is not fungible, so one can’t arbitrage between play-money markets, effectively reducing their size. I say should because there’s a pretty big discrepancy between Betfair and Tradesports real-money prices for US. House control. I’m guessing that with more active markets price difference among real money markets would shrink. There should be mountains of evidence one way or the other for sports bets. Anyone know?

By the way, Masse’s collective blog on prediction markets isn’t really launched yet but you may as well subscribe preemptively. Same for his insider blog which has a clever tagline (“the sidebar blog of prediction markets”).

Update 20060926: Masse points out Oddschecker, which does what I want for sports bets (hopefully they’ll expand) and a paper that has some evidence for lack of arbitrage opportunities between real money exchanges. See the comments for details.

Posted in Blogs, Economics, Politics, Prediction Markets | 7 Comments »

Long tail of (electoral) politics

Saturday, September 23rd, 2006

Nick Gillespie interviews Chris Anderson:

Anderson laments that national politics has yet to become part of the Long Tail. “I wish the system would put forward politicians that I could vote for,” he says.

I wouldn’t expect it to. At a minimum you need something like approval voting or at the extreme delegable proxy voting. I’ve always found such reforms curious but distracting, as I don’t know what their impact on policy outcomes would be, and I suspect they’d be small. However given that voters are not outcome oriented I wonder if being able to make a closer to their ideal expression when oting would make voters happier, at least for time they are in the voting booth.

But the real long tail of politics isn’t about elections at all. Even if I can vote for my ideal candidate, or vote directly on every issue, at the end of the day I will still get policies approximating those of George W. Bush and John Kerry. That’s like being able to order any of millions of books at Amazon but always getting the current #1 best seller delivered regardless of your order.

The real long tail of politics is decentralization and arbitrage. Lots of people say “Bush isn’t my president.” Why can’t that be true? Declare yourself Venezuelan, Hugo Chavez is your president. It should be (almost) that easy. If that seems extreme and disruptive, at least executive power should be curtailed, for surely it is the antithesis of long tail politics. And being able to live and work in any jurisdiction should be a given.

Via Boing Boing.

Posted in Apartheid, Economics, Politics, Prediction Markets | No Comments »

9-11 repeal: impeachment

Monday, September 11th, 2006

I don’t follow the nitty gritty and scandal du jour of U.S. politics, but I’m getting a stronger sense that part of 9-11 repeal should be impeachment. Glenn Whitman says it well:

Maybe I just have Bush Derangement Syndrome. But I find myself agreeing with James Wimberly and Mark Kleiman: there exist more than sufficient grounds for impeaching George W. Bush. In his recent statements about CIA detainees, he essentially confessed to violations of U.S. law. The laws in question provide for criminal punishments – of 20 years or more – for acts of torture and violations of the Geneva Convention.

Send the abominable person to prison!

FX’s play money Bush impeached or resigns (anytime before his term ends) last traded at about 17, more or less where it has been in its year of trading. The closest real money analogs I know of would be Intrade’s Rumsfeld.Resign.Dec06 at bid/sell/last of 16.0/17.8/16.0 and Cheney.Resign.Dec06 at 4.1/4.5/4.5.

Posted in Politics, Prediction Markets | 2 Comments »

« Older Entries
Newer Entries »