Archive for September, 2009

Unsolved Problems

Wednesday, September 30th, 2009
http://www.juicefeasting.com/Portals/0/juice/green-apples-sm.jpg
This isn’t an announcement that I’m happy to make, but it was an inevitable one: hypios has its first unsolved problems.

When we reviewed the solutions submitted for the communications department problems (the creation of a poster and the design for marketing materials), we were disappointed. None of the solutions was something we could use. Many showed talent and originality, but no single solution was both original and universal. One of the criteria that we listed in the problem description was that the final design would have to address potential Solvers from all sectors.

We were nonetheless tempted to choose a solution, just to have a winner. We would have liked not only to have been able to reward a Solver, but also to demonstrate that hypios can be a great tool for this kind of graphic design problem. But rewarding someone at random wouldn’t have proved anything. So hypios has become the first Seeker to use its right to say ‘No’. Here’s why:

Above all, it seemed dishonest to reward someone whose solution didn’t meet the criteria in the problem description. It would have been nice for us on a commercial level, because we’d like to promote the idea that every problem on hypios finds a solution, and for which one or more Solvers are paid. But this isn’t the case. In many respects, hypios works like a marketplace. Consider your local grocery store. Imagine you were shopping for a recipe that needed red apples (and only red apples). When you go to the store, they don’t have any red apples. Wouldn’t it seem weird if you had to buy green apples or bananas instead, even though you couldn’t use them?

Just as a product can be absent from other markets at certain times, it may happen that a certain solution isn’t available on hypios at a given time. Doesn’t it seem natural that people shouldn’t pay if they don’t find the product they’re looking for? This is why Seekers don’t pay to post problems. If Seekers had to pay a posting fee, it would be like having to pay an entry fee to the grocery store just to see if they had any red apples.

Economically, it wasn’t right for our company to pay for a solution that we wouldn’t be able to use. We were our first Seekers because we wanted to find out what it was like to be a Seeker, but also because we really wanted our problems solved. Our policy is that if the right solution (one that complies with the criteria in the problem description) is not found, the Seeker doesn’t pay. No red apples? We’re not going to force you to buy the green ones. In other words: Seekers that use hypios pay for solutions to their problems, not attempts to solve their problems.

Lastly, if we were to champion hypios’ utility for design problems, other Seekers might be inspired to post their own design problems on hypios. It would be like calling out “Fresh red apples here!” when we only have green ones.

All this only explains why hypios, like other Seekers, won’t pay for attempts to solve problems. There’s another question that we’ve been pondering: Was it only bad luck that, whereas we got great solutions for our computer science problems, the communications problems remained unsolved? Or was it symptomatic of a fundamental difference in the nature of the problems?

We’re preparing our own thoughts on the subject, but we’d also really like your feedback. Until then, check out the three new problems on the hypios marketplace.

(Photo: juicefeasting.com)

Share this post! add to del.icio.usDigg itStumble It!

What Open ID is all about

Friday, September 25th, 2009

When we first invited people to become Solvers on hypios, we were essentially asking them to create new online identities.  We asked them to fill in personal details like their name, occupation, location, and interests.  We gave them the option of adding a picture or avatar, and finally (politely) requested that they share their newly-created profiles with other Solvers by tweaking their privacy settings.  Now, we’ve simplified things by adding a Facebook Connect button.  You can join hypios using the Facebook profile you already have.

Having trouble keeping track of all your usernames and passwords?

It quickly gets annoying to have to fill out a new form with your name and hometown each time you want to join a site. Once you’re signed up, you have to remember which username and password combination you’ve used for each network.  In theory, this protects your privacy.  Social networking sites keep your data to themselves, in their own walled gardens, safeguarding your personal details.  In practice, this can be frustrating.  You may have befriended your colleagues on Facebook, but if you decide you want to contact them through Linked In instead, you all have to create separate Linked In profiles and reconnect through those pages.  You also have to deal with the etiquette of different networks.  As Patty Seybold notes in an Outside Innovation post, an executive who never checks his company email may reply instantly to casual messages sent via Facebook, moving business onto supposedly social platforms.

This poses an obvious problem for companies like hypios, whose platform is based on a social network of Solvers.  According to a recent ReadWriteWeb post by Alex Korth, users’ most significant issues lay in having to sign up for multiple profiles, re-enter and synchronize personal data, and their inability to export this data (even though it was personal information).  hypios’ success as a marketplace for solutions depends on our ability to broadcast problems to a wide network of Solvers.  The more Solvers we can recruit, the better the chances that we can help Seekers find the solutions they need.  If Solvers can join using existing profiles, they can also import the social networks they’ve built with these profiles.

How openID standards can make your identity portable

Fortunately, online identities are becoming more and more portable, thanks to open ID initiatives.  OpenID standards allow you to create one account for several sites. The ID can come from an email account (on Google or AOL, for example), a social network profile (your Facebook or myspace account), or another single sign-on (SSO) provider.  You sign in by giving your password to the ID provider, who then confirms your identity to a website or application.  For example, if you want to use your Facebook account to log in to the hypios network, hypios will never see your password. Instead, Facebook checks your password and confirms your identity to hypios, keeping your account secure.

Some application providers have also become ID providers, namely Google and Facebook.  Last December, both companies launched ‘Friend Connect’ features, which allow users to log in to different websites using previously-created identities.  However, as blogger John McCrea pointed out in his post about the launches , Google’s connect feature is built on open-source technology, while Facebook’s uses its own proprietary sources.  And while Facebook initially focused on integrating with major sites like Digg, Google promised to add a social component to any and every website by making it easy to cut-and-paste its code.

Who can you trust with your identity?

Of course, all of this depends on your trusting a provider with your personal information in the first place.  There are privacy, security, and practical concerns:

  • One irate former Facebook user, Leif Harmsen, told a New York Times blogger that “Facebook does everything to make you more dependent,” such that “it is not ‘your’ Facebook profile.  It is Facebook’s profile about you.”  In fact, Facebook ended its Beacon program after users complained that purchases they’d made on partner sites were automatically posted as ’stories’ in their Facebook profiles.
  • Besides concerns over commercialization of personal information, there are security worries.  Just as an OpenID provides users with access to multiple sites, it gives hackers the same freedom.
  • More practically, if an ID provider service goes down, users will be unable to sign on to any of the sites at which they use their IDs.

Why ‘open’ ID doesn’t mean ‘open to the whole Internet’

We think it’s important to remember that having an OpenID doesn’t mean that your profile becomes more open, though.  (Even if Facebook is encouraging its users to make some elements of their profiles open to everyone on the Internet.)  Though more and more websites will integrate social networks, ID providers will only reveal the details you’ve chosen to give out. It’s the same information, just displayed on more pages around the Internet.

As Alex Korth of ReadWriteWeb points out, however, ‘open’ IDs may eventually lead to interoperable IDs and a ‘web of identities’ much like the ‘web of data’ that will form the semantic web.  Just as data will point to other data with URIs, social connections will point to each other through OpenIDs.  Depending on how much information our friends share, a typical search query on the web of identities could poll social connections’ favorite books or vacation destinations by looking at purchase histories or status updates.

It could also help Solvers use each other as resources, whether by polling connections about the best new textbooks or soliciting advice on a scientific problem from like-minded people.  At the very least, it could help Solvers find others who have worked on similar problems or have matching interests.

Given our interest in the semantic web, we think this could be a really exciting way to connect with people.  No matter what ID provider you choose or how you use your online identity, we hope that openID integration will make it easier for you to become part of our network.

Photos from mandamonium and Arkangel on Flickr.

Share this post! add to del.icio.usDigg itStumble It!

More about our Solutions…

Friday, September 11th, 2009

winning hand

I promised to be back with a longer post after reading through all the solutions, so here it is.

We’ve finally evaluated all of the submitted solutions for the computer science problems posted in July, 2009.

And as a founding hypios member, I must say that I’m proud to see the platform keeping its promises: Not only were our problems solved, but, mostly, they were solved with great elegance.

When talking about hypios with potential Seekers, we often use the following arguments:

1. The Solver network on hypios is big enough to find someone with the right competencies to solve a Seekers problem.

2. There is a good chance for surprises: Seekers have access to what we call “360° points of view”: Solvers have viewpoints on the problem that all differ from the Seeker’s. This means that the right solution(s) often won’t look like what the Seeker envisioned (and in some cases not even like what he asked for). Seekers will get, as it were, a near to total set of options.

As for now, these arguments were more like predictions of how hypios would work. And the solutions submitted gave us evidence supporting both of our arguments: we had winning solutions from Solvers with unexpected competencies and winning solutions that challenged our own vision of the problem.

So as it seems we not only practice what we preach, but we preach what actually works in practice.


Share this post! add to del.icio.usDigg itStumble It!

Open Innovation made by LEGO & what others can learn from it

Wednesday, September 9th, 2009
LEGO2
Happy birthday, LEGO Mindstorms!

Why does it work so well and what can other companies learn to get their Open Innovation programs started?

At ten years old, LEGO Mindstorms is one of the oldest and most successful Open Innovation (OI) programs ever launched by a corporation. At LEGO, new product developments come not only from designers; customers are also encouraged to contribute to their design.  As far as I can see, their main, perhaps their only rewards, are recognition within the LEGO community, the pleasure of seeing their idea become a LEGO product, and a prominent place on the Mindstorms nxt website, dedicated visits in LEGO’s R&D labs and access to special beta-versions of some products. This is not bad, but it surely doesn’t explain the tremendous success of the program: there has been a vast amount of contributions from all over the world that have made LEGO’s Robotics System the best-selling LEGO product ever. But why does LEGO’s tool for open innovation succeed when so many others don’t? And what can others learn from LEGO? Points 1-3 explain the spontaneous hacking of LEGO robotic products which started off LEGO’s program and the program’s ongoing success. 4-6 are really about how LEGO rides the OI wave.

Why LEGO succeeded

1. LEGO makes toys. LEGO doesn’t produce weapons, it doesn’t produce soap or soup, it produces toys. LEGO has been around long enough for all of us to have played with it when we were kids. We know how they work, we like what they do, and maybe we wish that we were still playing with them right now (especially since they’ve gotten even cooler). The memories attached to LEGO (as with other toys) are unusually positive and emotional. Campbell or Clorox, to name two other companies which recently started to try to transform customers into innovators, fail to spark the same kind of joyful memory and enthusiasm as LEGO. Stefan Lindegaard wrote two great posts on the mistakes they seem to make on their OI platforms. I think that they could more specifically learn from LEGO. Here’s how:

2. (Nearly) all LEGO products are customizable, educational and technical. Customization goes very far at LEGO: on the LEGO factory website you can design your own houses, animals and spaces using something close to the software that LEGO designers use. If you wish, you can order the parts to your virtual construction and LEGO will send them to you. Parents who buy LEGO prefer toys that help their kids learn and develop, as opposed to toys that are just ready-to-use once they’ve torn them out of their packaging. Building with LEGOs makes kids active and apparently more technically inclined–they can either follow the directions to build things like police stations and pirate hideouts or use the pieces to invent their own model. LEGO seems to encourage kids to leave the beaten path. So kids who like LEGOs (and especially LEGO technic) are more likely become tech-savvy adults (and parents) that are easy to tap for the Open Innovation effort.

3. The gap between user and developer is relatively small with LEGO. As opposed to soap (or soup), you buy LEGOs because you value the challenge of building and the idea of making your own product. Most products are used in a much more limited sense. To use soap (or Windows, actually) you don’t need any of the skills that you would need to develop it. For a few years, to use computer software (this was true until MS-DOS) you needed some rudimentary knowledge of programming. Before Windows, computers were products where the line between developer and user was thin, which made it easier to appeal to users for help in development: Open Source communities emerged. And making a product more “user-friendly”, generally means to make it less (obviously) technical, hiding the technics far below the user interface.

4. LEGO’s OI challenges are embedded within a community. With videos, blogs, pictures from users, downloadable pictures, and online support, the Mindstorms Nxt platform gives you the feeling that, 1) participating in a LEGO challenge makes you part of a worldwide community of shared interests and pleasures, and 2) that the company really cares about you and your ideas. An online form like the one on the spartan Campbell site just doesn’t give you this feeling.

5. LEGO’s challenges are precise while remaining open, e.g. “we challenge you to build a robot based on an arcade game”. We’ve written about the problem of formalizing a problem before. But to keep it short: the key element is specificity. Even if you are open for various kinds of proposals, it won’t do just to say “Send us anything you think could be useful.”

6. LEGO’s successful adaptation of offline toys and their design on an online platform gives you the feeling of entering a universe when you enter the website. This is something many companies aim at, although few of them suceed.

The lessons other companies can learn

Now, the question is: what can companies do if they don’t have a product as well-suited to OI as LEGO’s?

1. Be conscious that marketing a product for co-development is not the same as marketing the product. Emphasize the play. Try hard to show the playful aspect of your products: create animation videos, drawings, shoot videos with engineers passionate about the project, etc.

2. Identify the biggest draw.  Spot your company’s product that is most customizable, playful, technical and start your initiative with that. This is the product with the right audience for a participative effort. LEGO started with LEGO technic, but extended the program to include all kinds of LEGO bricks with the LEGO factory that demands less technical skills.

3. There are products more up or down the line from users to developers. If you have technical products that require greater competence for use, use them for you OI efforts.

4. Create a community. Show that you care about outside innovators. LEGO lets you ask questions, makes users interact, mentions those who have contributed to an idea. Campbell, on the other side, states in their contract that they will never explain the refusal of an idea and that they will only pay if your idea becomes a patent (ouch!).

4. Break it down. Be precise and appealing in your problem statement. Otherwise, you’ll lose your time reviewing answers that are not on-target, if you get any. Choose to create discrete, precise challenges rather than going for a catch-all. A catch-all statement will look like you don’t know what you want, and that means that others won’t know, either. Think about #2 and 3 to choose the right product for your Open Innovation début.

5. Create a universe. A picture of the CEO and a “Submit your ideas” button is not going to do the job. Campbell, for example, could rely on Andy Warhol’s taste for their soup and give its site more of a “pop” look to  give you the feeling that their soups are iconic and that contributing to the development of such a product is rewarding. Clorox could put up some animations of the products they already have, show drawings, present the first formula that led to their first product, etc.

All this obviously takes a lot of work and time, probably even someone in the company who is only in charge of Open Innovation. Companies that are impatient and want to find out what they can learn from the outside right away, and who’d rather like to start experimenting with opening their corporate context with more limited costs and effort, should try to use a website with a community that is already in place, with people that signed up because they are interested in sharing their ideas and solving problems wherever they may come from, a place designed from top to bottom for problem solving and innovation, by people who only think about how to make the experience of problem solving more rewarding. A platform like this.

Klaus-Peter Speidel

Seeker Site Unveiled . . .

Friday, September 4th, 2009

We’ve got to thank, as always, hypios’ marvel of an engineering team for one more job well-done. They’ve gotten seeker.hypios.com up and running—with nary a glitch (well you know how things go). Moving forward, this site will be Seekers’ one-stop solution-shop. Here they can manage the entire problem submission process, from formalization to evaluation. Seekers can easily invite participants from within their home companies to pariticipate in communication with Solvers, improving problem precision, or evaluating solutions. Our aim was to make collaboration tools as flexible on the Seeker end as on the Solver end.

In general, what we’re aiming for is a seamless interface between Seekers and Solvers, with little or no need for hypios team support. That doesn’t mean we don’t want to help you (of course, we’d love to, and hypsters are always standing by) but, hypios believes that a social marketplace for ideas should minimize mediation. And the social network is a great tool for interactions. We think Seekers and Solvers are the relevant experts and they have all the information that they need to reach the right solution. What we’re here for, is to provide the tools and the security—the transactional architecture. And we’re sure it’s the best out there.

So, Seekers, hop on board, take a look, and let us know what you think. If you encounter any problems, glitches, bugs please send us your feedback.

Oussama Ammar
President

Share this post! add to del.icio.usDigg itStumble It!

Opening The Box

Friday, September 4th, 2009

Here we are! The deadline for the first problems has come and gone, and today we’re going to open Pandora’s box and start reading the solutions. It’s a great moment for us: being our own first Seeker gave us a profound insight into the mechanics of the platform. This alone was worth the money we dedicated to the problems.

After ironing out countless wrinkles, we now feel ready to offer to our first real Seekers the smoothest hypios experience… But did the platform work out as expected? Have our problems been solved?

A brief post, sorry: but I have a lot to read tonight. I hope you will surprise me.

Pandoras box, by Andrew Junge

Pandora's box, by Andrew Junge

Share this post! add to del.icio.usDigg itStumble It!

The Great(est) Race: Netflix, Crowdsourcing and the Winning Predictive Algorithm

Tuesday, September 1st, 2009
raceWe’ve been watching, over the years, a certain contest with great interest.  Netflix is an American DVD rental company that revolutionized its industry with the simple idea of letting people select and order films online and then receive them in the mail. It was a huge hit.  What they always wanted to do better, however, was predict the kinds of movies that their customers would like, based on other films they’ve seen and the rating they assign it (customers are sent a brief survey to evaluate recent rentals), in order to push films towards customers that they’re more likely to enjoy.

This search for a predictive algorithm is the first reason for our interest, and we are not alone. Predictive algorithms are on many companies’ wishlists. Now that we have access to all this data, like social network profiles that list a user’s preferences and interests (plus the site’s internal trackers that record users’ behavior patterns), the thought is that this massive amount of data should allow us to predict what someone will like or dislike, purchase or ignore.  (more…)