Page 2 of 4312345102030...Last »

Crowdfunding the future: Mad science, bad science, and the Tesla Museum

September 5th, 2012 by

As the original mad scientist, Nikola Tesla has been a source of inspiration to geeks of all types for over a century. The Serbian-American electro-wizard transformed the world with his development of alternating current (and inspired direct current supporter Thomas Edison to electrocute an elephant), before vanishing into madness and obscurity.

If you’re interested in crowdfunding, you’ll probably know all about Operation: Let’s build a Goddamn Tesla Museum, which has become one of crowdfunding’s most high-profile success stories to date. So far, over $1m has been raised to preserve the Wardenclyffe facility, the site of a particularly fascinating (and ultimately tragic) period in Tesla’s life. But as well as being an inspiring example of grass-roots crowdfunding success, the Wardenclyffe facility provides a useful insight into the pitfalls of funding today’s mad science.

Tesla is my Ohmboy

Tesla’s work was backed by J.P. Morgan and John Jacob Astor (which today would be the equivalent of having Bill Gates and the Sultan of Brunei in your corner). Both were confident that the man who won the “current wars” (without electrocuting a single elephant) would fulfil his promises of wireless power. Alas, the project fell apart in 1905, when Tesla’s alternating current patent expired and the royalty cheques stopped coming.

Unable to find more backing, and with Morgan and Astor growing increasingly impatient, the madness began to overwhelm the science, and Tesla suffered a mental breakdown. Wardenclyffe was foreclosed, and the site has been abandoned ever since.

This sad tale raises an interesting question: if Tesla had crowdfunded his research instead of relying on impatient big wigs, would we now be free from the tyranny of endless power cables (and have an unlimited supply of hats)? Is crowdfunding the key to unlocking the potential of the misunderstood geniuses of the world?

Igor! Check the KickStarter page!

One of the best things about crowdfunding is it gives inventors and innovators access to millions of new investors. Crowdfunding gives even the maddest scientists the chance to fund their zeppelin-mounted death-rays without having to pander to the impatient scrutiny of large, experienced investors. Hopefully this will mean the world will benefit from more inspired innovation (who doesn’t want their very own death ray?).

But the problem with mad science is that it’s very difficult to tell the difference between a crazy-haired genius and a crazy-haired crazy person. This is especially so if you’re an internet-based crowd, rather than a local, expert investor. My concern is that lots of people will lose lots of money funding crazy schemes, not only wasting their money but also discrediting crowdfunding.

Sorting the bad from the mad

Thankfully, sites like Petridish and #Sci-Fund Challenge offer a potential solution to this problem, by promoting the work of established research teams and insisting on university support before opening projects to donations.

Both sites insist on hard science, so if you were hoping to finance your perpetual motion machine, you’d better look elsewhere (and also consider visiting a psychiatrist, or at the very least investing in a high school physics book).

It is said that the difference between genius and insanity is measured only by success. By making sure that the crowd are funding mad scientists and not bad scientists (or just plain madness), sites like these provide a healthy dose of reality. The history of invention and discovery is full of live wires like Tesla. But as any electrical engineer will tell you, every live wire needs to be earthed.

When crowdsourcing goes wrong: Lessons in crowd control

August 29th, 2012 by

If the Discovery Channel ever makes a documentary called When Crowdsourcing Goes Wrong, I suggest you watch it. I say this partly because, as you probably know by now, our industry is full of eye candy. After all, there’s no better way to tone up for summer than working long hours in front of a monitor and near continuous travel. Even more compelling, however, is the fact that attempts at using the crowd can be interesting and occasionally hilarious, for all the wrong reasons.

The most recent example of crowdsourcing going wrong involves a campaign by pizza company Villa Fresh to come up with a new name for apple flavored Mountain Dew. As discussed in a recent post, the process was hijacked by pranksters from Reddit and 4chan, leaving the online leader-board filled with suggestions like “Gushing Granny” and “Fapple.” This is not an isolated incident. As General Motors once painfully found out, sometimes the crowd acts unpredictably.

The thing is, it’s not hard to see why Villa Fresh turned to crowdsourcing in this instance. Crowdsourcing efforts in very similar scenarios have brought excellent results. Alas, in this case, the restaurant chain was left with an embarrassing waste of time and brand credibility, and probably a stern glance from Mountain Dew owners PepsiCo. I wouldn’t want to be the guy to climb Mount Pepsi and tell the bosses up there that you accidentally named their new drink “Diabeetus.”

Recently we’ve offered advice on how to bring crowdsourcing into manufacturing, and how to use crowdsourcing to get help from your community. Here we offer some Villa Fresh-inspired ideas for effective “crowd control.”

Know your objectives.

It’s always necessary to have a solid aim, and when you think about it, Villa Fresh actually achieved its goal to “name a new flavor.” In retrospect, a better objective might have been “name a new flavor something appropriate.

Vet the crowd.

One of the best things about the crowd is its diverse thinking and creativity. In order to stimulate this unpredictable creativity you need a wide and mixed audience. But this doesn’t mean you can’t vet the crowd in certain instances. In Digitalkoot we did this by only allowing volunteers to actually participate once they had demonstrated that they were genuinely trying to help (one malicious player wasted hours deliberately typing in incorrect words that never did any harm to our results because he or she did not pass the vetting stage).

Actively guide the crowd.

Guiding or setting boundaries for the crowd often produces better results. A previous Mountain Dew-organized campaign on Facebook was more tightly controlled and much more successful for it. Guidance can work wonders too. When it comes to design contests, the more feedback you give the crowd the better the results are likely to be.

There’s a Kardashian-sized but coming, though. While minimizing the risk of the crowd behaving badly is good, you still have to maintain the integrity of the whole process. Heavy-handedness in certain circumstances will stifle creativity and potentially alienate the crowd. Vet the crowd too much, and you may submit to expert bias (or end up with no crowd at all). Being too rigid in your aims will deny you the occasional flash of unexpected brilliance.

It’s a tricky balance to find, but there have been plenty of examples of companies getting it right. And admittedly getting it wrong can be entertaining. Glass of “Diabeetus”, anybody?

Summer Blockbuster, in cinemas now: The Document Processing Knight Rises

August 21st, 2012 by

microtask_document_processingAs regular readers of this blog know, there is nothing we like more than discussing strange and new types of crowdsourcing. From weird music-related experiments to the incidence of expressions such as “I need to” during the Mad Men era, we try to keep you informed with what is going on across our industry.

Every now and then, however, we use this forum to talk about something much closer to our home and hearts: ourselves.

For the last few years Microtask has focused its efforts on solving the traditional problems associated with document processing. Basically we have been trying to magically transfer those nearly illegible, dusty piles of paper forms that fill the dark comers of your office into digital format on your computer.

This summer…(pretend I am saying this really slowly in that movie preview voice), Mike Rotask rises again, to save the world from the evil problems associated with document processing, and banish those paper forms you hate, forever.

Ok, the animation is not quite Batman, but it does describe how our incredimagic team of oracles and sorcerers thinkers and engineers managed to triumph over the traditional impediments to document processing. Enjoy the magic of cinema, and give us some social media love!

Search me: what Mad Men and brave moles can do for historical records

August 15th, 2012 by

Ever since we began helping the National Library of Finland correct mistakes in its old newspaper archive, I have noticed myself developing a slightly anti social interest in historical texts. I say ‘anti social’ because of its effect on conversation: what I have found is that while most people claim to be interested in history, the best way to get unwanted guests to leave your house after a dinner party is to start discussing the technical challenges presented by digitizing historical records.

To overcome this problem in the Digitalkoot project, we sacrificed thousands of cute moles: if volunteers failed to enter the correct words the poor moles fell to their digital deaths. (While we felt bad about manipulating people’s love for cartoon animals, we can still sleep at night because it was for a good cause.)

How to get people interested in history (without hurting animals)

This experience is the reason I admire historian Ben Schmidt’s recent success in getting people enthusiastic about history and digitization.

What Schmidt has ingeniously done is take two of today’s most popular TV shows – Mad Men and Downton Abbey – and check their dialogue for historical accuracy against millions of texts published over the last few hundred years. (For example, apparently even the phrase “I need to”, so common in Mad Men, is not something that people said in the 1960s).

This study was made possible with Google’s Ngram viewer. As we discussed about a year ago, Ngram allows you to chart how many times a different combination of words or letters has appeared in Google Books’ huge corpus of 5 million texts published between 1500 and 2008.

Along with the many fascinating results, the great thing about this project is that it has managed to get the media interested in Ngram again. My own Ngram-like analysis using just normal old Google Search found that after some initial excitement in 2010 and 2011, almost everyone had forgotten about Ngram until Schmidt found a way to include it in the same sentence as Don Draper.

Don’t forget about the crowd

The reason I bring this up is that I am disappointed with the lack of interest the world’s institutions have shown in the technology we now possess to preserve, analyse and search historical records. (Even Google Books seems to have lost momentum in its efforts digitize the world’s texts.)

While OCR technology still makes too many mistakes when digitizing texts, we know that people are actually willing to give their time to correct them, assuming we can find a way to include cute animals or handsome Mad Men in the tasks.

As we found with Digitalkoot, the gains from such projects actually go beyond the statistical results: not only have hundreds of thousands of people freely volunteered to help, it has also been a great way to get them interested in the historical records themselves. The only downside of this is that bringing up the digitization of records is no longer a reliable way to make the last dinner guests leave my house.

See hear: crowdsourced subtitles for everyday life

August 9th, 2012 by

microtask_backstroke_of_the_westAs anyone who has watched a few movies with unofficial subtitles will know, quality control can be a bit of an issue. But even in the world of terrible, inaccurate subtitles, nothing comes close to the awful poetry that is Star War the Third Gathers: Backstroke of the West (better known as Star Wars Episode III: Revenge of the Sith). The homemade subtitles that accompany this pirated version of George Lucas’ blockbuster are so wildly mangled that it has achieved cult status. (To give you a taste, in this version “Obi Wan Kenobi” is ingeniously translated to “Ratio the Tile”).

Thankfully, the mastermind behind Backstroke of the West has absolutely nothing to do with an incredible piece of subtitling software in development at the University of Rochester.

I see what you mean

The software was created to provide deaf people with real-time subtitles for everyday life. Currently, deaf people are forced to rely on expensive professional transcribers, who are only available at certain times. The Rochester research team, led by Walter Lasecki, instead rely on Mechanical Turk users to provide live translation for a fraction of the cost, day or night.

As well as helping deaf people navigate the noisy world, this crowdsourced transcription system is perfect for travelers who don’t speak the local language. The researchers plan to launch a smartphone app called Legion:Scribe this October, which will help you pay too much for souvenirs in almost any language you can think of.

The possibilities are truly exciting. The app could transform your smartphone into a Star Trek-style Universal Translator, letting you talk to absolutely anyone. If the researchers manage to hit their goal of a near-instant transcription, you could even watch a play, lecture or movie, no matter where you are in the world.

But what prevents you from getting nonsense like Backstroke of the West? The reason Star Wars Episode III came out so mangled in this translation (apart from the input of the lizard-being who replaced George Lucas in 1989) is partly because it’s the work of just one person (apparently this person translated what he heard into Chinese, then fed it through a machine translator to get the English subs that are so hilariously mangled).

Listen and earn

Legion:Scribe breaks up the stream of sound into bite-size chunks, and members of the crowd transcribe each chunk for a small fee. Just as we’ve found at Microtask, the researchers noted that smaller tasks led to improved accuracy, and that by letting multiple members transcribe each chunk, the work can be automatically verified. As Mechanical Turk does not have such a system in place (unlike some crowd computing companies we could mention), the researchers had to create their own algorithm, which is currently in the beta-testing phase.

The beta version of Legion: Scribe has a 74% accuracy when transcribing ordinary conversation, compared with an average 88.5% for professional transcribers. This is a significant difference, but the researchers aim to tweak the software and close the gap before the final version hits the app store.

The Legion: Scribe project is an inspiring illustration of what human computing can achieve. By combining innovative research with the existing power of the crowd, the Rochester research team isn’t just helping millions of deaf people to communicate. They may also demolish the spoken language barrier entirely.

Like the Descriptive Camera, which I discussed back in May, the Legion:Scribe app has gone from idea to reality at an astonishing speed. The fact it already generates reasonably accurate results demonstrates how far our understanding of distributed labor has developed in the last few years. Let’s just hope that it won’t prevent entertaining disasters like Backstroke of the West popping up every now and then!

Page 2 of 4312345102030...Last »