Deriving knowledge from information

Google’s Knol is now open to everyone, causing turmoil in the blogosphere, particularly among those interested in the old battle of Turning Information into Knowledge.

An old battle that is! Leonardo’s famous notebooks were a way of organizing information into knowledge. Today, Google claims that it’s not easy enough for people to share their knowledge. Enters Knol.

Knol is seen by many as a direct competitor and threat to Wikipedia. Google’s official message underlines a key difference: the idea behind the Knol project is to highlight authors. Each article (dubbed Knol) is expected to be authoritative about specific topics. Unlike Wikipedia, Knols are driven by authorship. “Moderated collaboration” is also available, allowing authors to work together — yet, as far as I’ve seen, most articles are written by a single author and are set for “Closed collaboration”.

The blogosphere is currently boiling with posts on how Knol will fail and never defeat Wikipedia. I am personally not interested in that soup. I am more interested in the social implications of a platform like Knol. So I decided to plunge, choose a topic, and compare Wikipedia’s and Knol’s articles on that very same topic. The topic I chose is something I’ve been increasingly interested about: Bone Marrow Transplantation. Here are the articles.

After carefully reading both articles in a learning mood, here are some of my impressions:

• Displaying the author’s name and portrait makes little difference to me as a reader. Both articles are quite extensive and contain probably more information than I was looking for.
• I “trust” both articles’ content equally.
• The single author vs. community of experts debate seems to make little difference to me, at least at first sight. The content is exhaustive on both platforms. I expect this to be the case on most topics, as experts in any fields are already influenced by the collective knowledge (learning in Universities, working in teams, collaborating in research projects, reading other experts’ articles, etc.). The exception to this could be very recent topics, news or discoveries, in which case the community has (in theory) a higher chance of updating the online information faster. But collaboration is already taking place, even if it’s not within Wikipedia.
• Knol’s rating system is a way of tapping collective intelligence. The Common Internaut is already used to rating information online, from restaurants and movies to articles on online newspapers.
• If I’m serious about acquiring knowledge on a specific topic, I would without a doubt read both Wikipedia, Knol, then keep on searching other sources of information. I would even take a walk to the local library and browse physical books.

One clear difference between both platforms, as far as I see it today, is that Knol appears to make it easier to anybody to write articles. Why? Simply put, a Knol resembles more a blog post than a Wiki page. Whether we (tech geeks) like it or not, Wikis are a bit scary to the Common Internaut (I can hear the screams of blasphemy already). In a sense, Knol lowers the barrier to entry for writters.

Continuing with the blog comparison, take a look at Knol’s article on Bone Marrow Transplantation and scroll down to the comments area. Read the first comment — it looks and sounds like a blog comment, doesn’t it? This particular comment is not related to the article, but to the author and his team. It discusses a story, a life and death situation — something definitely important, at least to the comment’s author and her family. This is something that would rarely take place in Wikipedia, or would it? Is that useful? Is that noise? Is that information? Is that knowledge? Should a collective panel of experts decide?

Will the quality of Knols be tainted by the commercial interests of the platform? Only time will tell. In the meantime, at least to my eyes, there is no single authoritative platform deriving knowledge from information. Knol and Wikipedia will certainly coexist for many years.

To regain the love of my fellow wiki-loving friends, here’s a geeky ending for this post:

Where is the wisdom we have lost in knowledge? Where is the knowledge we have lost in information?

— T.S. Elliot, The Rock

Where is the information we have lost in data? And where is the data we have lost in noise!

— Gale Moore

Advertisement

Paris — Gargoyles

Things that really matter

For a while now I’ve been trying to focus on things that really matter. That is highly subjective, of course, and not an easy task. But with so much knowledge and resources on our hands, focusing on projects that target only an exit strategy is no longer appealing to me. While life takes us on a quest for success and external happiness, we can choose to stop for a second to realize that the really important things are in front of our very eyes.

As I often say it, the future is already here, it’s just all broken. That is a negative statement, I am aware of it. But I strongly believe that technology professionals have the moral obligation of fixing their mess, not just substantially contribute to entropy while focusing on financial growth.

While Google advocates not being evil, a new trend is rising: not being evil isn’t cool anymore, being good is the new black. I’m not fully convinced that this falls in the things-that-really-matter bucket. Yet it’s definitely something that keeps start-up folks motivated. Still, isn’t it just a fashionable facade in front of our crave for growth?

There are many things to fix — the list is long. I’ve been recently investigating one of them: Education; more specifically, the North American education system. I had the chance to meet a few American and Canadian Educators that think outside the box, yet (with a couple of exceptions) these forward looking people still appear to me as living in the past. A little research shows a few ongoing efforts to improve the Educational System, but I believe that what is required is not just a bit of dusting but a major re-thinking of what Education is all about, a true revolution.

While doing a bit of research, I bumped into this video that presents the Education system as we know it, as we lived it. I hope you enjoy it as much as I do.

Evil Whiteboard

Who said that remote white boarding had to be sexy and virtual?

Flowering Gardens

Butterflies

GHOSTS

I have always admired Trent Reznor‘s artwork. I have also admired Marcel Achard‘s creations since I met him many years ago here in Montréal. While very different in style, they both produce multi-sensorial, viscerally emotional and personal content.

Tonight the moon is full, and my inbox is calming down — Easter weekend started. Yet, I just got a surprise: Marcel sent me a short video he’s producing in the context of the GHOSTS collaborative project. Get your hypothalamus ready, dim the lights down, raise the volume, then click play:

Pitching ideas

Pitching your ideas to investors is exciting. You have a few minutes to share your ideas, prototype or product. Your goal is to get them to want to know more about yourself and your story, and to convince them that both your people and ideas are worth investing in.

Assuming that you have a good team, presenting people is the easiest part. Pitching ideas, on the other hand, is much more difficult.

After presenting our latest and greatest adventure in technology to a couple of VCs, I can openly say that we did not succeed in pitching the idea. On one side, I hear excited customers telling me every single day how much more fun their job is, how much time and money I save them, and how they just could never go back to their previous ways of working. On the other side, I don’t hear much from these groups of investors. Bad sign!

Trying to come up with a model to describe the situation, I bumped into an article by Bill Buxton. In his article, Bill Buxton introduces a law that he calls Gradual Granularity Refinement Law. This law states that the granularity at which we distinguish meaningful differences gets finer the more our familiarity with a subject grows.

The GGR Law is about familiarity with a subject. If you are pitching an idea, and you have both a gut feeling plus feedback telling you that the idea is good, it probably is. If you managed to come up with a clever product or idea, it’s because you have enough familiarity with the problem to model things in a different way. You probably felt the pain and were brave enough to do something about it.

The GGR Law is also about granularity. Different groups of people may have felt the pain. But the more familiar you are with the problem, the easiest it is for you to distinguish meaningful differences, and therefore come up with clever solutions.

What does this really mean? You are familiar with a problem space, you have felt the pain, you have a good idea, and developed a good solution. You are making some money out of it, and you have a great team. Now you find yourself in a conference room pitching your idea to investors. How can it go wrong?

Unless you apply the same level of innovation and creativity to your pitch as you did to your problem solving, your story will just not be interesting enough for the people across the table. Why? Because it’s unlikely that they are as familiar with the problem space as you are. Your explanations may be good, and they may understand the concept at an intellectual level, but unless they understand it at a visceral level, they will just not want to know more about your idea.

I have an appointment with a very interesting group of investors in a couple of days. This particular group seems to be quite familiar with the problem space, unlike the previous ones. Plus, I now know what to do, so it should be fun.

Wish me luck! 😉

Arrivederci Eolas!

In 2006, the folks at Micro$haft changed the way Internet Explorer embedded control objects, affecting in particular how web developers deal with Flash objects. The reason: the Eolas patent dispute. The outcome: web developers started growing considerably more gray hair than before. Yet, it contributed to the flourishing of Flash embedding Javascript libraries, and to endless discussions on how can you best embed flash content.

These discussions were not only about circumventing the side effects of the Eolas patent dispute. They were about providing alternative content, about simplicity of use, about standard compliance, about cross-browser support, and about graceful degradation and progressive enhancement. Still, now that Microsoft announced that an IE update scheduled for April, 2008 will remove the click to activate functionality, reverting the software to its original design, I am trying to figure out what it means in practical terms. After all, the click to activate message is probably the main reason why many web developers adopted embedding libraries in the past years.

From an embedding point of view, not much changes really. Web developers still have to embed the beast and provide graceful degradation. Apple’s iPhone still doesn’t support Flash, as far as I know. Plus, accessibility is being enforced by laws in many countries. There are no two sides to this story; we still cannot rely on markup-only methods.

From the simplicity of use side of things, using a Javascript library is IMO the best way of embedding an interactive beast in your web pages. Decoupling is good, we were told in school, remember? A clean, well tested Javascript library takes you right there in no time.

From a standards compliance point of view, Micro$haft is pushing very hard for people to upgrade their browser to IE7 while working on IE8. This means that web developers may finally start worrying much less about IE6, and start worrying more about IE8. Is that good news? I’m not sure about that yet. I am happy to see that IE8 passes the Acid2 test. I am even happier to hear that HasLayout is as good as gone in IE8. Yet, while I do agree that the DOCTYPE switch is broken, I still can’t manage to get my head around Micro$haft’s version targeting mechanism proposed for IE8. It seems to me that every time Micro$haft proposes a new way of solving a problem, web developers have to expect a new avalanche of issues to be on top of. I sincerely hope I’m wrong this time.

While EOLAS falls into history, object-embedding Javascript libraries are here to stay. Today, the authors of the two most compelling Flash embedding libraries have joined efforts into the SWFFix SWFObject project. In practical terms, adopting this library is probably the most forward-compatible action item in my list. For the rest, it seems to me that regardless of headaches and hours of suffering caused by IE6 and the EOLAS dispute, it will all fade into nostalgia like Polaroid’s film.

SAYONARA, EOLAS!

Back to Montreal, with Python fun.

I’m back in Montreal recovering from DEMO 08, where I had the chance to meet super interesting people and to show Cozimo to the press, entrepreneurs and VCs. Now that Cozimo is officially out the door, I’m busier than ever. Luckily enough, it’s not just about answering I forgot my password support e-mails… some interesting developments are forming and there’s still energy left for midnight hacking, so I can’t complain.

Still, I can’t wipe out from my brain the vision of Monique Savoie menacing me: “Don’t you dare hiding for a year ever again, or else!“, so I decided to go to the very first Montreal Python event. And it was fun!

David Goodger, architect of DocUtils and renowned Python activist, entertained us with one of his old-time passions: polyform puzzles.

David briefly described algorithms involved in solving these puzzles…

… but I wished he focused a bit more on stuff like this:

It was an interesting presentation, and a good way to start Montreal Python’s series of events. Montreal is buzzing more than ever, and it feels really good to be part of such a vibrant community. Kudos to the organizers for a well organized event.