I’ve never been much of an online shopper. I avoided shopping online as much as I could. Until recently.

I discovered Amazon’s Mechanical Turk just before I left India. I created an ID with my India address and forgot all about it.

Until a few weeks back. I made some money on MTurk by participating in surveys, transcription, stuff like that. And then I wanted to transfer it to my US account.

Nope, nada. You can’t. You can only get it delivered to your India address. Which will take six weeks. And $4. And you can’t change the address unless you delete your account (and lose your money) and create a new one.

So the only other option I have is to convert it to Amazon.com balance. Which I did.

And after buying stuff for friends and family on the website, I now have a modest amount left to spend on myself. Let’s say I have $15.

I want to know what I can get for $15 in books. What sort of combinations. Woody Allen + Milan Kundera? Artemis Fowl + Le Petit Nicolas? What am I missing out on? What can I buy which is better?

So… this can be considered a knapsack problem. It occurred to me after my situation reminded me of this xkcd cartoon:

So what is the Knapsack problem? I’ll explain the simpler, more common 0/1 Knapsack Problem. So you have a set of items. Each item has a weight w and a value v. The knapsack can hold at most a weight of W. The problem is to choose which items to fill in the knapsack, such as to maximize the total V in the knapsack, while keeping the total weight under W. You can learn more here.

My Amazon.com problem can be considered a knapsack problem with the weight w of each item as its price, and the value v as how much Amazon thinks I’ll like the item. That of course is possible using collaborative filtering (yes! I know a term!) and other techniques. They can also use their ‘frequently bought together’ feature here.

Would others find it useful? Yes, if they are on a fixed budget like I am. Or if they want to buy just enough to be eligible for Free Super Saver Shipping.

Would it be possible in real time? I’m sure something can be worked out there.

And…. crowdsourcing this… any recommendations for nice books worth buying under $15?

On an aside, it’d be nice if WordPress.com had suggestions for most-frequent tags, and asked if we want these tags to be converted into categories.

And… Happy Thanksgiving! Wish you the best of Black Friday deals!

WELCOME TO THE DIGITAL DIGEST

GREETINGS, geeks and welcome to a brand new section The Multifarious: the Digital Digest, a selection of regurgitated slurry from this week’s web-and-tech-related news.

Nothing clever like a theme here, but instead a 2p pick-and-mix of chewy, gelatine-and-E-number-packed goodness from the dingy newsagents that is the Internet.

MP predictably slams ‘violent’ computer game: millions of gamers groan inwardly

Yes, I’m talking about the <ahem> ‘row‘ surrounding the release of a little game called ‘Call of Duty: Modern Warfare 2′. Trumpeting the news that certain levels of the 18-rated FPS contained content that was – gasp! – quite violent, MP, Keith Vaz, threatened to raise hell in Parliament on behalf of the curtain-twitchers of Leicester. You’ll be very surprised to find out that the root of all this manufactured outrage started life in pages of the Daily Mail – fancy that! This hysteria accompanies the revelation that a French teenager, who was recently convicted of planning to murder his classmates and teachers, was a ‘video games enthusiast’ – something that the media took great pains to point out.

The UK games industry provides 1000s of jobs and remains one industry in which the UK can still claim to be a world leader: meddlesome politicians looking to score cheap shots before a general election should perhaps bear that in mind.

Live the future now (and look a teensy bit of a plonker)

TED (that’s Technology Entertainment and Design, in case you didn’t know) is a global forum where the world’s leading boffins mass together to give long talks about how clever they are in various mind-meltingly difficult subjects. Wait, come back! It’s actually very good and often turns up gems like this one, where MIT student, Pranav Mistry, brings us one step closer to a ‘Minority Report’-style future of holographic interfaces, controlled by simple gestures. The most amazing thing is that he’s turned the paradigm that each future application (as science fiction would have you believe) would require a different piece of hardware on its head, by employing an amazingly simple idea.

DEATH TO IDEAS!

Nice pictures of how the fragile butterfly that is a creative idea can be trampled beneath the hobnailed jackboot that is the process of getting out to the public.too many laboured similes, I know.

Google Library Scan FAIL

So, as Google’s plans to digitise every out-of-copyright book rumble on, questions are being asked as to whether or not they should be the sole publishers/custodians of such a vast repository of knowledge. This suggests that the big G is not quite ready to assume that mantle.

Adidas World Cup

Real-life German footballers act woodenly in a lavish intro to this high-concept World Cup digital promotion from Adidas Germany. But hold on! What’s this? Heavens! It’s actually a game! Plays a bit like Subbuteo, but slightly less fun. Thanks to James M for this first piece of World Cup digital marketing on the radar.

Usability and User Experience…zzzz

No, DON’T fall asleep: this might actually help your clients a little happier and your job a little easier to boot. Anyway, I happened across a couple of interesting sites that provide a very cost-effective alternative to expensive focus groups and UX labs:

http://www.fivesecondtest.com/ & http://www.feedbackarmy.com/

With Fivesecondtest.com, you can upload a page design and either set up a quick memory test, where a user is displayed the design for 5 seconds and has to list 5 things he or she remembers about the design, or a more involved click test, where a set of click-based goals are set for a user.

The memory test could be useful for simple A/B testing e.g. working out the best location for a promotional container on a page based on which design has the best recall; the click test works in a similar way, but specific questions can be asked (although this is a premium feature). You can send the test design to an anonymous horde of testers, or circulate a secure link among people you know. At the entry level (11 results and no custom questions), this application is totally free, but if you need a bigger sample, then it creeps up a little bit ($12 for 46 results). I think it makes use of Amazon’s ‘Mechanical Turk ‘ service for the pool of testers – useful in its own right if you’re desperately hard up!

Feedbackarmy.com carries on in the same vein, this time with the ability to ask open-ended questions as standard, the twist being that you can send a link to a functioning website instead. The price for 10 responses? A very wallet-friendly $10. Luvverly jubberly.

And finally – man plays Guns ‘n’ Roses ‘Sweet Child O’ Mine’ on an…

…how can I put this ‘unconventional instrument‘. I wish that John Lewis had decided to use Mr Handman’s version instead of this rather drippy cover. Anyway, I guess you could say it’s a real ‘parpy’ hit! Snort.

So, that’s it for now. Next instalment is due in a week’s time when I will be serving up some hot nuggets of web goodness.

Until then…KEEP WATCHING THE SKIES

Elisabeth Oppenheimer blogs about CrowdFlower and GiveWork:

CrowdFlower.com is a site that uses Amazon Mechanical Turk’s technology to create work for refugees in Africa. AMT, as noted, allows companies to parcel out tasks that are simple but can’t be done by a computer. One problem with AMT is that it’s hard to check answers for quality; people may not understand the task, may not speak the language well, or may just blow through the task to rack up the payment. There are different strategies for dealing with the problem; this article, for instance, describes how to design certain types of tasks for maximum effectiveness. What Crowdflower does instead is charge a premium to have independent users double-check the work.

CrowdFlower has started a project called GiveWork. CrowdFlower employees train refugees in Africa to do AMT tasks. iPhone users who have downloaded a free app can then donate a minute or two to double-check the work. (In fact, AMT users frequently say that they’re doing tasks not solely for the money, but because the tasks are easy and amusing, like solitaire. CrowdFlower pushes that angle to iPhone users.) Once the iPhone user has approved the work, it’s sent off the the company that requested it, and the refugee is paid. You can check out more details on CrowdFlower’s website.

This is cool: it gets real work in the hands of people who can complete it and for whom the money is desperately important. There are plenty of things to worry about with ubicomp—labor standards, the disaffection that comes with assembly line work, doing a piece of a task without being able to evaluate the moral valence of the whole—but the potential should be nourished.

Read more .. ..

I am fascinated by alternative employment. And I don’t mean Suzie swings on a pole instead of climbing the corporate ladder. I define alternative employment as anyone engaged in commercial exchanges that are not main stream. It is not interesting job descriptions I am after, but people who trade their skills in the marketplace in a different way than gathering a paycheck by providing a service or selling a good in a traditional manner. This includes a lot of new and innovative models developed or at least propagated by the internet as a marketplace—people who live off the advertising hits in their blog and people who prefer to barter than sell or get paid.

More specifically, it usually includes some combination of freelancing and self-employment. I am intrigued with providing a good or service in the context of creating your own market.  There are a great many things that would be neat to do for a living, but to find someone who needed that cool thing done every day for eight hours, five days a week, month after month, year after year was highly unlikely. Instead, that one rare task is requested once by thousands, and you might be able to make a living doing that one cool task, and maybe even on your own schedule.  This is the power of the internet; it has the potential to corral that market for you.

Even more amazing is that some new websites allow freelancers a huge customer base and varied tasks tailored to, or tailor-able to their skill sets. In some cases, it allows the purchaser to taste test or shop for the best—all the while paying for the service. In the past, if a company wanted a brochure written, it would hire a technical writer for a lump sum or by the hour.  The writer would deliver the brochure by the deadline and get paid.

Now it is cost-effective enough to post the task, have five technical writers do the task, pay all five and use the best of the five submissions—and get a better product while saving money in the end. In other words, it redefines or at least questions the efficient market theory in relation to commerce because money is spent multiple times on duplicate tasks. Economic theory relies on goods and services trade favoring the most cost-effective method. In theory, the economy has no waste or excess profit because all goods are sold at the best price to the consumer and all goods are of the highest quality. Of course, as workers and consumers we know this is not true; there is asymmetrical information, all processes in production or delivery are not efficient and goods pricing is, well, not always scientific.

The notion that task duplication could be more economicallyefficient than doing something once, at least in the service of quality, is a paradigm shift. As a creative professional, I am more likely to understand production or service providing as a non-linear process; many times you start out not knowing what the finished product will look like or how long it will take. Even in the cases you do, the path to get you there is unclear. It often involves many of the classical education methods—trial-and-error, model building, research—none of which are efficient. If learning were about efficiency, there might possibly be one best way. Or as we have discussed here, maybe not.  But ways of learning is fodder for a future post.

Growing up you may have heard things like ‘do it once and forget about it’, ‘give it your best shot’, ‘do it right the first time’ or something like that. The main premise here is that if you put maximum effort out applying all your knowledge, will likely get you the best product. But is this true? Our discussion seems to point toward doing something many different ways, or getting many opinions at least, before settling on one course of action.

Three sites have popped up on my radar of late in regard to alternative employment:  Elance (www.elance.com) , Mechanical Turk (www.mturk.com) and Associated Content (www.associatedcontent.com). Associated Content is a site for freelance writers. It brokers writing tasks, usually very specific ones, to people who might know about a specific, esoteric topic, like how to make your own all-natural Halloween face paint. Methinks even Martha Stewart would be scratching her head on how to cover that one. These very specific writing tasks now have a marketplace, and people with very abstruse interests or knowledge can cash in if properly matched.

Elance is a large, freelance agglomeration.  It collects interested freelancers from around the world and people with tasks seek out the talent. Individuals can register as service providers or buyers. It is like looking at a massive resume database; most searches are done around specific skills, like programming languages or design specialties. If you wanted to start your own business, you could do that with very little overhead by starting here.

Mechanical Turk might be the most innovative of the three. It was started by Amazon almost by accident as an in-house project. Instead of finding talent and the appropriate task, it lets the worker choose. Mechanical Turk has the power to break tasks down into discrete parts, which might be the secret to its efficiency. As mentioned above, it might allow more than one person to work on and submit a task. The tasks can be very specialized, akin to technical writing, or very inon-intellectual but time-intensive that a few hours of websurfing could accomplish.

I find these three sites notable because they also all offer the prospect of working from home, on your own schedule.  The old saw that if you picked a company and ‘worked like hell’ it would pay off in twenty years with a fat paycheck and corner office is long outdated. Workers are not trusting corporations because loyalty is fractured. Workers are finding progressive, lifestyle-congruent ways to make money and enjoy their lives, sacrificing as they choose. The internet is helping.

I recently spoke with a friend who is retired from the Navy. We were talking about the government’s health care plan debate. I told him you needn’t worry:  you have government-sponsored health care for the rest of your life; you will be taken care of. He said: ‘As well we should. We were the ones who were miles under the sea away from our families for twenty years.’  He was right in a way; he sacrificed for our country and should receive the best. However, there was a strain of entitlement in his voice. It is a concern that I hope to not ever have to feel—that after my days working are done, I still feel owed something for my years of work.  I want to feel, for the most part, that every day my job and life are a fair deal.

There are scores of people peeling off from the rat race and taking their own excursion to define their career and what it means to work.  First there was trade. Then it expanded exponentially, 2.0, with the internet. Now the internet is getting finely-grained in its applicability and match-making capabilities as people get the hang of its power. I am interested to see where it heads, and hope it only makes many people feel a little less frustrated, a little more adequately compensated, and a little more free.

Bicycle Built For 2,000 is comprised of 2,088 voice recordings collected via Amazon’s Mechanical Turk web service. Workers were prompted to listen to a short sound clip, then record themselves imitating what they heard.

O projeto final é aqui no site.

“People are imperfect. What we have learned through the ages, though, is that combining lots of people creates a better end result.”- Eric Raymond, The Cathedral and the Bazaar.

Though the term Crowd-Sourcing has been mentioned to point of meaninglessness (much like the word “innovation”), we’re finally seeing it’s ‘power’ take roost in web applications surfacing of late that are actually useful.

CrowdSpring has been of interest lately, pissing off seasoned designers as they see small jobs go to fresh college grads and inexperienced web monkeys. It’s an interesting concept that has flipped the traditional design business model of it’s head. Future Paul Rand’s of the world won’t be able to charge a million bucks for an IBM logo-power has been democratized with employers & those without a design background, not so much the designers used to another business model entirely.

What’s far more interesting and a glimpse into the future of our increasingly socialized web economy is Amazon’s Mechanical Turk. Named after a fake chess playing machine that hid a human inside it, it’s a fairly simple concept: be a micro-job hub for tasks computers cannot do. For example, compiling evidence on Google’s superiority to Bing, reviewing  first hand what people think of when they see your face, where people like to be touched in intimate situations, and seeding MasterCard’s “Priceless” app with local deals(McCann NY did this and their postings on MK are the largest available).

From a research and planning perspective, these examples only scratch the surface on uniting global agencies creative brain power for a variety of tasks and projects. I can only imagine how this new found economy of opinion and insight sharing will build upon itself.

Monetary and business possibilities aside, Mechanical Turk has an incredible altruistic vein, with thousands of users using Google Earth to aid in the location of adventurer Steve Fossett last year. Though no one found him prior to his crash site being located in September, it’s a testament to the expanded power of the herd and it’s availability to anyone with an idea and a computer.

I wonder if they’re research companies that make use of Mechanical Turk. If I were a usability testing company, I’d be taking this head on.

Today at Harvard Law Schools’s weekly Berkman Center lunch, Aaron Shaw presented into the potential  Amazon’s Mechanical Turk(AMT) holds for social science and the culture that surrounds it. His talk drew upon research-in-progress from the Berkman Center’s Online Cooperation group, in collaboration with Daniel Chen and John Horton.

Although the presentation itself, cheekily entitled “HIT me baby one more time, Or: How I learned to stop worrying & love Amazon Mechanical Turk,” was a bit light on statistics, the conversation within Berkman’s community around the issues of labor laws, privacy, methodology and technological potential were fascinating, as always.

Aaron Shaw at Berkman

As Shaw noted, the origin of the name for  Amazon’s Mechanical Turk lies in a chess-playing “automaton” that was no mechanical creation at all, but instead a clever contraption that hid a chessmaster inside. Amazon’s version farms out small tasks — or “HITs” — that require a human to accomplish.

As an aside, I have to note that, as Peggy Rouse pointed out in Mechanical Turk, Powerset and enterprise search, there may be considerably more to Amazon’s strategy than the creation of a crowdsourcing market for simple tasks. She thinks Mechanical Turk may play a role in enterprise search down the road. She’s a canny observer, I’d recommend reading her thoughts.

Early in his presentation, Shaw offered up a shoutout to Andy Baio (@waxpancake) who asked two questions late last year in “Faces of Mechanical Turk“: “What do [Amazon Turk users] look like, and how much does it cost for someone to reveal their face?”

Credit: Andy Baio, Faces of Mechanical Turk

As Shaw noted, however, when it comes to the Turk,  no public, trustworthy, aggregate data is available. What evidence is available derives from self-selecting surveys and experiments. Those samples showed a large number of women, from many countries of residence (although mostly in the US & India). Speculatively, he noted that the age of users appears to be low, while education and income is high.

Shaw posited that the geographically component is likely correlated to Amazon’s requirement that users hold a US banking account.  As a result, Shaw’s research relied upon whatever his team could collect on the Turk or through interviews with users and Amazon executives.

So, does the Mechanical Turk work for its users? Sometimes. Shaw noted that once you get a few people performing a given task, the accuracy rate for completion goes up overall, providing the example of machine-learning algorithms.

As he noted wryly, it’s “Not all bots, cheaters and scripts.”

Task selection and design is important to that success rate: skill matters, on both sides. It’s not just the skill of users and their ability to follow instructions – success also relies upon the skill of the creators of the HITs. Social scientists — scientists of any stripe, really — recognize the issue here in experimental design.

The uses of Turk cover a broad spectrum, though by nature each represents some form of crowdsourcing. Amazon itself used to Turk to generate product descriptions, questions and answers, thereby “spamming itself,” as Shaw put it.

Spectrum of users of Amazon Mechanical Turk

How else is the Mechanical Turk being put to use?

• The Extraordinaries: “micro-volunteer opportunities to mobile phones that can be done on-demand and on-the-spot”
• CastingWords.com is using it for transcription
• AaronKoblin.com uses Mturk to create art. For .02, he pays users to draw a sheep facing left. He then sells sheets of them  for $20, some portion of which is donated to charity.
• Also noted: oDesk, reCAPTCHA, Threadless, Aardvark, liveops

Aside from commercial, artistic or volunteer uses, Shaw believes that Mechanical Turk has considerable potential to enhance social science.

Specifically:

1. As a pool of subjects for randomized experiments
2. As a pool of inexpert raters for distributed observation, or “coding”

Advantages to labs?

Low cost of use, ease of paying subjects, speeds, diverse subjects (potentially), one HIT = one person, workers do not (usually) interact.

Experiments can consist of contextualized real-effort tasks. As the Turk has created a real labor market, as for text transcription, there’s utility in many areas, like canonical games in economics and paired surveys.

In other words, its neither reducible to a manifestation of the “Internet hivemind” or some sort of “latter day child labor,” at least in Shaw’s view. The online conversation around the presentation, which included Esther Dyston, was more skeptical on the latter point, noting that the potential for skirting labor laws was not inconsiderable. Shaw readily conceded that the issue is salient, although he sees such labor issues as “downstream,” he expects to see more given that the “tension is so clear, so stark.”

Shaw has been advised by Yochai Benkler while at Berkman, who evidently considers the Turk to be of use for content analysis for distributed observations. In this context, the ability for researchers to randomly assign HITs for raters to code objects is helpful. Shaw brought up Klaus Krippendorf, of UPenn, in the context of understanding some of the theory here; I’ll need to go do my due diligence in understanding Krippendorf’s work.

Yochai has noted that specific groups involved in distributing computing types, like SETI, have performed admirably. According to Shaw, in fact,“The Knights who say “Nee” perform quite well when measured against other countries with distributed computing.”

I also heard about the “Turkopticon,” a Firefox extension that allows users to submit feedback about HIT creators. Although Shaw said that it is not widely installed, there’s clearly a step towards community self-policing.

When asked about the utility of using the Turk for searching for missing computer scientist Jim Gray or searching for Steve Fossett’s plane, Shaw immediately recognized the value but hadn’t examined the data sets in question at length.

The question itself begged for a follow up, given the release of Chris Andersen’s “Free” this week: How and why are users motivated to provide hits when altruism is involved? Is work of higher quality when there is money involved?

Shaw offered a cautious affirmation, though with reservations: Payment vs free is “such a loaded issue in society. The symbolic value of money or donation is humongous.”

A Berkman Fellow in attendance, Chris Soghoian, noted that his advisor pays 5-10x the market rate and gets email about when the next task is coming, along with decent results.

In Shaw’s view, there needs to be “a more serious examination of the question. Experimental evidence of research suggest sub-populations of people who would respond differently. Some people will be motivated by doing good, others don’t care, want the .05. We need better ways to test. It’s situation-specific.”

As he wryly noted, “We’re not all homo economicus.”

As usual, this was an excellent lunch.You can view the archived video of the presentation as a .mov.

Following the presentation, Aaron wrote me to add the following:

“Daniel and John’s contributions to the field of experimental research on online labor markets include

1. recognizing that AMT could serve as a venue for experimental studies;
2. conducting the earliest labor market experiments on AMT;
3. solving a bunch of difficult problems so that they could make valid causal inference based on the results of these experiments.”

I have to note one other organization I learned about today: “TxtEagle.” TxtEagle is a innovative concept for active “mobile crowdsourcing,” distributing small-scale jobs via SMS and payment the same method. 

In other words, microjobs with micropayments. The mobile platform’s founders recognize that there are more than 2 billion mobile phone users in the developing world that could potentially be leveraged to perform tasks. The BBC wrote that “txteagle is changing the dynamics of outsourcing labour.” Hard to disagree with that.

Comment tisser des liens toujours plus forts entre les consommateurs et les marques? Il existe des milliers de solutions possibles, inutile de les répertorier.

Internet, le village global, offre cette particularité de relier la terre entière, a fortiori de trouver réponse à chacune de nos questions, voire même une main d’oeuvre pour chaque travail.

Amazon est à ce titre célèbre pour avoir réussi à convertir ses clients en travailleurs à travers le crowdsourcing. A différents niveaux, ces derniers notent, écrivent des critiques, alimentent le site en recommandations… Depuis quelques mois, le géant du livre en ligne propose d’aller plus loin en salariant certains de ses clients. Les tâches proposées sont basiques – tris d’images, reconnaissance de fichier, etc. – mais permettent de rapporter quelques dollars au pronetariat et économiser un back-office énorme pour la marque. Tout le monde y gagne.

Pour aller plus loin sur ces marques qui font travailler leurs clients, voici une veille Né Kid rédigée il y a quelques mois :

Découvrir le service Mechanical Turk d’Amazon.

The Galaxy Zoo is a neat little site with a long list of prestigious universities behind its development. It harnesses the power of human cognition to allow users to look at pictures of galaxies and classify them for future research. Humans are way, way better at this visual processing tasks than computers, as it turns out.

The site itself is easy to use. You register, look at a one-page tutorial, and then you are off, counting the arms of spiral galaxies and flagging unusual astronomical features in photos. The bulk of the galaxies I classified were smooth and dull, but that just made the occasional brilliant spiraling galaxy more exciting. There is an option to check if you see anything unusual – I daydream that when I press this option a fleet of goggle-wearing scientists in lab coats receive an emergency page and scurry to a computer to see the anomaly.

This is a great game for wasting time – just fun enough to hold my attention and easy enough a child could do it. In fact, in a Scientific American podcast (click here to listen) with Yale astrophysicist Kevin Schawinski, one of the project’s creators, Schaminski said that was one of the great things about the Galaxy Zoo – parents get online and classify galaxies with their children, getting them excited about scientific research.

The Galaxy Zoo isn’t the only project out there that harnesses the power of the internet and the human brain. Called human-based computation, these projects usually seek to accomplish what otherwise might have been the world’s worst temp job and turn the task into a game shared by a huge network of people. The ESP Game makes image search engines work better by having two players look at an image and guess what descriptive words the other player uses to tag the image. The tags that the players agree on then become associated with the image and improve its searchability. 

Can you count the spiral arms? Of course, you are a human.

ReCAPTCHA is the bookworm cousin of CAPTCHA, those annoying squiggling letter tests that verify your humanity for all sorts of online activities. ReCAPTCHA puts a handwritten word already known to the computer alongside a untagged word from manuscript. The user types both in, verifying their humanity and coding word in a manuscript in one fell swoop. Best part is, putting reCAPTCHA on your site is free. Although perhaps both CAPTCHA and reCAPTCHA might fall into disuse if Slate’s Chris Wilson is correct that increasingly sophisticated bots are learning to read these images as well as humans – or, possibly, that spammers get around this security measure simply by outsourcing the human cognition overseas.

Some of these sites will even pay the human workers. The Amazon Mechanical Turk is the only one I am prepared to say is not a scam. Not a get-rich-quick scheme by any means, but you can put your human brain to work and earn yourself a few Amazon dollars through this site.

If you are looking to tap the power of your computer rather than your brain, a list of active distributed computing projects that use a portion of your computers CPU can be found on Wikipedia here.

We’re working on a new product here at Grocio and it may just present you or someone you care about an opportunity to make a little in between money.

What’s needed:

• An Amazon Mechanical Turk account;
• Fast, fast, accurate, accurate typing skills;
• Ability to spend 15 or 20 minutes of focused time to punch out each assignment.

Rather than describe the assignment here… Better to direct your attention to our assignments on Mechanical Turk. If it says someting like “Your search didn’t match any HITs”… that simply means all our assignments are checked out. No worries, there are plenty and I’ll edit this posting once it appears we have some predictable scheduling to offer.

“Amazon Mechanical Turk is yet another way of aggregating service providers and buyers in one place ‘online’ where you can make money depending on your area of expertise.

Business week mentions “people can offer payment for businesses and services that require a human element rather than computing power”.

ShortTask is a marketplace for tasks, aiming to match-make “solvers” and “seekers”: people who have time to complete brief tasks and people who need small tasks completed.

In many ways, ShortTask is similar to Amazon’s Mechanical Turk (MTurk) service. Like MTurk, ShortTask is seeking to address tasks that aren’t large enough to necessitate the hiring of a contractor or part-time employee. This marketplace is largely oriented around activities that can be completed from home — ideal for web workers — and currently numbers around 61,000 available tasks.

Activities include transcribing audio and video files, annotating images, copywriting, commenting on blog posts or online forums, data entry and online research. Seekers simply publish a task, set a payment level and deadline. Solvers can explore the site and select the tasks they’d like to complete. ShortTask takes a 10 percent commission on each transaction.

ShortTask might be a useful way to find additional work. However, I’m uncomfortable at the payment levels set in the marketplace. Salon published a great article in 2006 examining the labor law implications of Mechanical Turk, “I make $1.45 a week and I love it!,” which suggested that MTurk-like services are undercutting the minimum wage.

For example, today ShortTask has listed a task to transcribe a 22-minute video, within four hours, for $5.50. That’s about $1.30 an hour, or the equivalent of $200 for a month’s work. That particular task is one of the better paying tasks listed at ShortTask; the worst fast-food service jobs will pay you better.

If services like ShortTask expect to flourish, they’ll need seekers to pay fairly, so that a reasonably sized population of solvers can be sustained. More significantly, the marketplace operators should at least adhere to minimum wage levels.

People have fought for the right to be equitably compensated. Codifying that in the services we use to find work is both essential and moral.

What do you think of marketplaces like ShortTask?

Unlocking Value in Online Communities

RESEARCH OVERVIEW.

[ Presented at MIT Innovation Lab Meetings, March 19-20, 2009, hosted by Prof. Eric von Hippel ]

Online collective approaches to work is a nascent area of inquiry, but one that is growing in importance. The research presented at the MIT Innovation Lab meetings aimed at contributing to our understanding of this phenomenon by looking at three contrasting cases :  the Gutenberg Project Distributed Proofreaders (self-organized, not paid), Facebook Translations (firm-sponsored, not paid), and the Amazon Mechanical Turk (firm-sponsored, paid).   The study of these cases comprised a survey exploring the motivations of the online communities associated to each.

For much of the last decade, the literature has focused on widely acknowledged self-organized public endeavors such as Linux and Wikipedia. Most recently, however, a growing body of evidence has emerged of successful firm-sponsored private endeavors using online distributed work processes in other industries. Salient examples include Amazon’s Mechanical Turk, Google’s Image Labeler, Facebook’s Translations.   This is the focus of my research (Villarroel et al 2007; Villarroel 2008).

As an example, on December 27, 2007, Facebook launched Translations, an initiative inviting people to join a “community of translators” to make Facebook available to everyone in all languages. This online collective initiative was supported by an online distributed application allowing Facebook users to translate words or phrases from the Facebook platform itself, which other users could vote upon to judge their accuracy. Within four weeks, 1,800 members of the Spanish speaking community had translated Facebook entirely. Within one year, the community had translated the Facebook platform into 100 different languages and dialects, without requiring formal contractual ties to the firm.

Although successful, the approaches to online collective work vary significantly from one case to another, ranging from self-organized to firm-sponsored, collaborative to competitive, unpaid or paid, and differing in implementation according to problem complexity, among other factors.  My ongoing research seeks to develop a taxonomy and study the effectiveness of the different approaches.

MIT Innovation Lab Meeting Presentation

We have been spending time talking and working with folks who have been using Mechanical Turk.  So we finally put it to some good use as part of a project to learn more about people’s desks. 

So we asked 28 people (we inlcluded 10 of them above), to share pictures of their workspace along with something they would like to change about their desks.  We asked them not to tidy or change anything and so we received responses throughout the day, from a variety of countries. The whole process took less than 12 hours. 

I guess everyone’s desk accumulates drinks over the day.

Before Google was a verb, Twitter was an avocation, and Facebook was a way to spend your evening, it was easy to know where to get your information. Step 1: Go to your local library. Step 2: search the catalog (and before that, peruse the card catalog). Step 3: compare the citations for the term you are searching to see if they came from “reputable” sources such as a major newspaper, a mass-market book by a “famous” writer or historian, or if the same facts were corroborated by different sources. How can you get similar fact-checking on Internet sources when the news itself might only be minutes or seconds old?

The easiest way to trust what you read on the internet is to rely on the old Mass Media (WSJ, NYT, etc), and ignore the web sites or sources you don’t know or trust. This is a reliable indicator for major events that are well-covered by the traditional mass media. However, the NYT doesn’t help you with the small decisions you make every day. But an interesting new web site aims to improve the ability of your personal network and of the personal networks of others to arrive at the Truth (from your perspective). Caterina Fake, co-founder of Flickr, explains that Hunch can provide the right answer for you for the question you asked.

Hunch operates on two levels. The first level, popularly known as Collaborative Filtering, can provide reliable recommendations “liked” by the crowd that are more accurate than individual predictions and produce recommendations for future “likes” based on present behavior. Hunch also asks you questions about You — so that it can take the crowd-sourced recommendations and give you a lens through which to view them that matches your personal views — and “guesses” or creates a Hunch tailored to you. This second level of filtering matches the large number of random (or not-so-random, depending upon the question) answers from the crowd with your actual likes and dislikes. I think this is the beginning of a critical thinking ability for the internet that can help you to identify unknown sources of content as valid or invalid (with the help of networks like Hunch).

What are the implications of getting a personalized version of the “Truth”?
It’s a little scary on the one hand to think that you are seeing the “Right Answer” to your question based on your likes and dislikes. On the other hand, the only person who can accurately rate an answer to your question as correct is … You. Hunch’s grand potential is that it could take the questions individual people have, distribute them to others to validate, and then take the “right answers” and syndicate them out in bulk as “crowd-sourced” news.

Imagine a world where you can ask your Hunch network to help you decide (provide you with options, not tell you what is right and wrong) whether an individual piece of content, or a personal decision to take the freeway instead of local streets, or whether the hamburger or the chicken is most fattening. Ok, you say, — what’s different about this from Mechanical Turk, Yahoo Answers!, or any one of the other answer-my-question sites that have sort of succeeded, sort of failed in the past few years?

Two differences are interesting to me that will make services like Hunch succeed: scale and mobile computing platforms. Caterina Fake has already determined how to build a large consumer service to scale. Flickr executes billions of interactions for millions of global users, and does so in a way that is pleasing to use and consume. Mobile computing platforms (I’m thinking of iPhone, but others like Android are on the way to becoming ubiquitous as well) make the idea of “asking a question” to your Answer Network more plausible at the moment that you need it.

One final thought on this issue. Could Google + Twitter produce this harmony of information, albeit in a more brute force, “ask my friends and index the result” sort of way? The folks at ReadWriteWeb recently covered this development, and it will be interesting to see where it leads. In the meanwhile, Hunch is an interesting and new way to answer the question: Can You Trust What You Read on the Internet. The answer? Maybe.

So, this idiot wanted to pay for fake enthusiastic products’ reviews. Not enough: to review clearly BAD products without even using them and to mark the real, bad review as “not useful”. Big time screw up, baby Mike.

Oh, boy. I hope the Belkin guys kicked his ass out of the door. Badly.

Another hard day at the social web zoo. The good news: at least one less dope around the interwebs.

Michael Bayard, a Business Development Representative for Belkin (A manufacturer of audio, video and computer cables, power protection, wireless networking, and other mobility accessories) thought that it would be a brilliant idea to pay people to give five star ratings to Belkin products on Amazon.com, Buy.com, and Newegg.com.

Bayard decided to use Mechanical Turk (a site that where people can post tasks that computers cannot perform, and other individuals can get paid to do that work) to hire people to give positive ratings to Belkin products. 

How stupid does one have to be in order to think that that could end well. Did it not occur to him that this was a very public way of hiring people to lie about Belkin’s crappy products?

Belkin – Your products are big fat turds. All of the polish in the world won’t fix that problem. Stop paying people to say you make great products and actually invest in creating great products. If you make good stuff you won’t have to pay people to say they are great. They will do that on their own. 

Arlen Parsa (The individual who broke this story) – I salute you. This online ecosystem only works if people fight to preserve some level of honesty and integrity. Well done. 

Michael Bayard – I present you with the first ever Geek Whisperer’s Über Idiot Award. You’ve earned it with your exceptional jackassery and total lack of thought. As you write your resume you should remember to include the award. You’ve earned it.

Update

In the interest of fairness, here is Belkin’s response. I would pick it apart some, but I don’t think it’s particularly necessary. They lost me as a customer a long time ago because their products are subpar. This has only served to reinforce my view of the company.

HIT services are going to be a hit this year in the mobile space – as a result of the influx of photos with app-specific uses. Photos aren’t just to share anymore – they’re for processing… for example, Amazon’s own Remembers service.

Amazon’s Mechanical Turk service for HITs: http://aws.amazon.com/mturk/

After twice trying to find a doctoral program that satisfied my intellectual curiosity while giving me the tools and credentials I need to become a university-level researcher and teacher, I’ve decided that the time is now for me to finally make it happen.

To anchor this desire to a tangible goal, I will give myself until Tuesday, December 1, 2009 to prepare and submit all of my applications to doctoral programs. From this point forward I will refer to this date is my “PhD-Day.”

Why this date? Simple: of all the doctoral programs that interest me, December 1, 2009 is the first application deadline for fall 2010 enrollment. This is the date on which I will finally take that “one giant leap” into my long awaited career in academia.

Although circumstances beyond my control were partially the reason behind my pulling away from my doctoral pursuits, I know now that I was also not clear enough about my goals. I just wanted a doctorate and did not give any meaningful consideration to the discipline in which it was anchored and how that would impact my future career options.

Previously I thought having a PhD qualified you to teach any subject, but I now realize that, with rare exceptions, the discipline in which you earn you PhD is the discipline in which you will concentrate your research and teaching.

Knowing the purpose of a PhD is to, as one of my colleagues comically suggests, know more about less, I must start with a question of “what” first, then determine “how.” I have therefore stopped first looking for a program (the “how”) that I will then try to make work with my interests (the “what”).

Instead I will take the opposite approach and first determine the topics I want to research and teach (the “what”) and then find a program that offered opportunities to study it (the “how”).

After evaluating what most interests me, I realized it had been staring me in the face the entire time: social media. I have previously mentioned my interest in this burgeoning topic in previous posts and in my list of research interests, so this is not breaking news by any means.

However, I have finally embraced the idea of studying it academically so I can understand it as a researcher and not just as a user.  Specifically, my research interest is to investigate the impact of social media on the creation and distribution of information.

What is social media?

I define social media as follows:

Social media includes information generated with and shared by individuals using various web-based tools including blogs, message boards, video sharing sites, wikis, chat, IM and similar technology.

I also feel it is related to concepts such as crowdsourcing and collective individualism.  Social media also touches on the idea of distributed computing, though in the case of social media the “nodes” are human and not computers.

In a more abstract interpretation, social media could also cross into the realm of artificial intelligence – especially as the tools we use to connect socially online continue to become more intuitive and personalized to each user.

The main use of social media is knowledge sharing among individuals for the greater good. However, it can also be leveraged (or exploited, depending on how you look at it) for commercial gain. Of course, marketing in this medium is not without its challenges and it certainly can’t be done in a traditional way (e.g. forced and artificial vs. the natural, organic feel of true social media).

Beyond products, people who participate in social media often market ideas or even products by the information they share (consider the metoric rise of Barack Obama who, despite your political persuasion, was impressively effective at using social media in his presidential campaign).

Social media can also be used as a training and development tool.  As a father to a child diagnosed with autism, I also wonder how social media might help my son learn social skills and share information in a virtual environment. As a parent, I have already been impressed by the power of social media to connect me with important information and individuals focused on autism.

I am also fascinated by the thought of using social media to enable many individuals to complete parts of a larger task (what first piqued my interest in this was when Steve Fossett went missing and there was an attempt to find him using Amazon’s Mechanical Turk, even though that effort was not successful in finding him).

Clearly, social media can be both a communication tool and a marketing channel. I am interested in social media in both of its forms. It intrigues me that technology can so intensely enhance our human experience.

How will I study social media?

My investigation into current doctoral programs that include social media revealed it is relevant to several disciplines. Information Technology/Computer Science and Communication are the two I have most frequently found. Social media is also relevant to the disciplines of Psychology, Marketing and Sociology. Given the impact social media has had on the workplaces, Management is also a reasonable discipline in which to study it.

It makes sense that social media crosses into several disciplines — it is quite pervasive, but can also be investigated from many different angles. Perhaps what angles I want to study, or maybe how I want to study social media, will ultimately dictate the discipline within which I will investigate it further.

At the moment my assumption is that I will most likely be studying social media either within a Communication or Marketing program.

Where will I study social media?

Given my practitioner mindset and entrepreneurial orientation I would like to be able to teach in a business school. To do that I will need a PhD from an AACSB-accredited program.

However, given my background in communications and journalism I wonder if Communications would be a more suitable environment (especially since I am not as interested in traditional business subjects like finance and economics)?

I am still evaluating my options, but right now my top choices include the following (in alphabetical order):

• Claremont: PhD in Management and Information Systems (Interfield)
• Claremont: PhD in Management and Organizational Behavior (Interfield)
• UCLA: PhD in Marketing (Anderson School of Management)
• UCSB: PhD in Communication (Technology and Society Emphasis)
• USC: PhD in Marketing (Marshall School of Business)
• USC: PhD in Management (Marshall School of Business)
• USC: PhD in Communication (Annenberg School for Communication)

Aside from the obvious criteria of being accepted into a program is the issue of funding. Having already borrowed my way through an MBA program, my goal with the PhD is to get the cost of the program covered while also earning additional income through fellowships and other related methods.

Why do I want to earn a PhD?

I have always wanted to understand why and how certain things work (or don’t work). Whether I am contrarian by nature or unquenchably inquisitive, I was never satisfied with a surface level answer about anything. My problem was, and remains, not having the proper “tools” with which to conduct proper inquiry.

I also fundamentally enjoy creating and sharing knowledge. Looking back to my years in journalism, I think the desire to craft a story and share it with readers is related to the same idea. Notably, I recently learned the first academic paper I wrote and presented was referenced in a book called “MBA in a Day” and an article I wrote about non-profit fundraising five years ago in Marketing News (the bi-weekly trade paper of the American Marketing Association) was cited in a recent academic paper.

I was energized knowing that something I wrote helped someone else create something of their own. I want to be an active part of this process. On a related note, my experiences at academic conferences were unquestionably positive and motivating. I relished those opportunities to exchange ideas and information with difference people, creating knowledge in the process. This is why I am so endeared to the concept of “generative learning,” which Peter Senge defines as learning that “enhances our capacity to create.”

My long term purpose in embarking on this undertaking is to secure a position at a university where I can engage in active research while still teaching. I have been adjuncting online and in person for roughly 18 months now and have thoroughly enjoyed the experience. It has been a very liberating and affirming time for me, especially when I continue to get positive reviews and comments from students. But I also want to be adding to the academic dialogue, not just guiding students to a basic understanding of what has already been produced.

Also, although I have no pressing desire to go back to the corporate world, I am open to partnering with industry on research and also potentially consulting on the side. I just really don’t want to have to worry about red staplers and TPS reports! Even when I was in industry I tended to approach things in a more intellectual way than most. I even had two managers with whom I had good relationships tell me I was definitely “an academic” and would do well in that world.

I am a thinker and a tinkerer, but not a hard-core corporate type. I enjoy discussing and debating a topic sometimes more than “doing” whatever that topic is related to. For example, I enjoy the concept of branding and understanding how people develop allegiance to a brand, but don’t necessarily want to go launch a branding campaign.

How will I stay focused on my goal?

Staying focused on achieving my goal of earning a PhD will be challenging, given the various personal and professional obstacles I will need to overcome to see it through. At the same time, I find myself thinking about a PhD with increasing frequency: it is something I must do, not just something I want to do.

Nevertheless, there are plenty of distractions. Ironically, during the past eight months, I have repeatedly encountered one kind of distraction while driving to and from my adjunct teaching job: a delivery truck with graffiti on the roll top door that reads “PhD.” I am sure this is some tagger’s initials, but for me it represents and reminds me of my dream: a PhD.

What makes it more significant to me is that I have seen it numerous times — driving north or south, in the morning or afternoon. Usually I encounter it on or near the Grapevine portion of Interstate 5. I am unsure where it is driving to or coming from, all I know is I have seen it numerous times — at least a half dozen.

Perhaps this truck is my albatross — or maybe its just coincidence? Maybe it was sent by the “PhDMV” to keep me on track?

Below are two photos I took of the truck on March 25, 2008 while heading home from DeVry (southbound on Interstate 5). Following the photos is a short video clip I filmed the morning of October 6, 2008 while heading north between the base of the Grapevine and the split between Interstate 5 and Highway 99 North (near Lebec, CA):

What are my next steps?

Now that I have defined and committed to this ambitious goal, how do I intend to achieve it? My next steps include the following:

• Thoroughly research the PhD programs at the aforementioned schools.
• Read “The Craft of Research” and write a specific research statement.
• Begin reviewing the academic theories relevant to social media.
• Speak with colleagues and mentors to understand my options.
• Start writing my statement of purpose.
• Explore grants and scholarships.

See you in a year on PhD-Day!