Amazing but true–the bar code turned 35 today. I discovered that fact in an interesting article in the New York Times that notes that the first bar codes were scanned on a pack of Juicy Fruit gum on June 26, 1974. Bar codes are now scanned 10 billion times a day–compare that to Google’s paltry 293 million searches/day as of March 2009. Clearly the lowly bar code has some serious mojo and I think that mojo is called “elegant simplicity.”

One interesting tidbit from the article is the fact that when the bar code was being developed by a team at IBM, the only change requested by the MIT  committee charged with reviewing the concept was a change in the font of the numbers below the bar codes to one that was believed more readable by machines. The request was made because the review committee was convinced that the bar codes would soon be replaced by smart machines that would just read the numbers–sound familiar?

As a fan of simplicity, this story resonates because in my experience, simple things endure, complex ones generally don’t. George Laurer, the leader of the team that created bar codes said that the idea succeeded because it was cheap, needed, and reliable–I would add a fourth–bar codes succeeded because they are simple. Scanning a bar code retrieves information from an associated database–that’s it. It doesn’t take a rocket scientist to understand or use bar codes.

For those of us that create information products, we can probably learn from the bar code–instead of layering on more features, more technology, and more content perhaps we should be stripping products down to their bare essence–what’s needed by the user–no more, no less. We should be so lucky as to create digital products that will still be in use in 35 years. While we are on birthdays, did I mention that COBOL just had its 50th?

Share and Send

With apologies to Firesign Theatre, I borrowed one of their titles as I just discovered one of my hallowed notions, the Hawthorne effect, is likely wrong. For the uninitiated, the Hawthorne effect comes from a study conducted at a telephone factory (Hawthorne Plant) near Chicago in 1924. The study was aimed at uncovering the impact of shop floor lighting on worker productivity. To make a long story short, the study found that virtually any change in lighting, brighter, dimmer, etc. led to increased productivity–in other words the very act of experimentation caused an increase in productivity.

I’ve used the Hawthorne effect notion countless times over the years, so imagine my dismay when a recent Economist article informed me that two economists, Steven Levitt and John List, recently found flaws in the study that invalidate the findings–the Hawthorne effect apparently isn’t. More disturbing is having worked for companies that tried to invoke the effect via lighting, muzak (ecch!) and other environmental factors. When I reflect back on my gullibility, I realize that I fell prey to a sort of “truthiness” where my intuition accepted the Hawthorne effect without a rigorous examination of the study results. In effect, the results seemed logical, therefore I believed and even evangelized the concept. Lesson: with the growing amount of information on the Internet generated by “experts”, pundits, and seemingly credible sources, it is increasingly easy to formulate a weltanschauung based on assertions that “jibe” with our way of thinking and therefore become true in our minds. At the moment I feel like one of the bozos on the Firesign Theatre bus, but I’ll try dimming the lights so I can finish my day’s work more quickly.

Share and Send

I am spending a couple of days immersing myself in all things semantic technology at the Semantic Technology Conference, in preparation for an upcoming report on that topic. While government, defense contractors, and pharma companies are prime markets for semantic technology and were all in attendance, several information providers also presented at the conference.

Most interesting session of the day was a publishing panel, with Evan Sandhaus, New York Times; Keith DeWeese, Tribune; Paul Berry, HuffingtonPost.com; Jim Stanley, CBS Interactive; and Michael Dunn, Hearst Interactive Media.  Topic of the panel was how publishers use semantic technology. Evan noted that the NYT started using semantic technology in 1913 in the form of an inverted index of the daily papers, sold bound in cloth or leather as a product. The indexing was done by people, and the NYT still uses a large team of people along with semantic technology to ensure quality of indexing. The NYT uses semantic technology to create topic pages and feeds.

HuffingtonPost.com uses semantic technology to help moderators find and eliminate abusive or offensive posts to their sites. Interestingly, they also are using Twitter as a source for breaking news, a comment that had all of the panelists nodding in agreement. Michael commented that reporters won’t be breaking news stories, they’ll be packaging news updates from Twitter and other sources.

One interesting observation from CNET was that they had used semantic technology to summarize discussions in their communities but found readers preferred the disaggregated view of the discussions. They also use semantic technology as the “plumbing to span content across all of their silos.”

The one question that seemed to challenge the panel was “how do you monetize semantic technology, indeed can it be monetized?” The answers ranged from “still looking” to “semantic technology is a need to have–a cost of doing business.” Semantic technology, at least for these publishers, is achieving success in improving productivity and content usability, but still looking for a role in generating significant revenue streams.

My final observation is that the semantic technology is a giant petri dish of experiments–many will fall by the wayside, but a few will become very important. The challenge for publishers and information providers is to figure out which is which. More to come.

Share and Send

Having put up several finch and hummingbird feeders in our yard, we are now awash in birds trying to out-tweet each other. I’ve noticed that bird tweeting is somewhat democratic as every bird adds to the chorus–from simple clicks and tweets up to full-blown “bird arias.” Apparently that is not the case at Twitter, according to recent research by the Harvard Business Review. The study found that Pareto is alive and well– a paltry 10% of Twitter users account for 90% of all tweets. That sounds more like a broadcast platform than a social network and is somewhat lower than patterns of contribution for sites such as Wikipedia. Moreover, the study found that the median number of tweets per user was one tweet, something that would doubtless make our birds snicker. The study also found some interesting gender-related trends, but the basic usage stats are what caught my attention.

An easy (and cynical) explanation for the results is that Twitter is “lame” and people quickly get bored with it as it is a one trick pony. Easy but not convincing as there are many people that get a lot of value both from tweeting and following others on Twitter. I think a better explanation is that using Twitter (or any other social networking tool) effectively requires significant time and energy–something many of us find in short supply, as well as patience.

Unless you grew up sending Western Union wires to friends and family, you may find it difficult to cram your thoughts into 140 characters of text. Also, the “Twitter beast” needs to be fed and fed often to be useful–you won’t have much of a following if you post once per month. Finally, “your results may vary.” Unless you have serious star power, it will take a while to build a following, so be patient and keep tweeting.

I think it is also important to note that you won’t “get” Twitter (or any other social networking tool) unless you put in some quality time with them. Simply reading about them will not get you there any more than reading about scuba diving will prepare you for diving the Mariana Trench. If you are serious about using such tools in your business, join some networks, do some tweeting, and figure out how (and more importantly if) such tools actually make sense for your organization.

Share and Send

I recently concluded that my thinking on cloud computing is, ahem, clouded. With all of the hoopla about cloud computing, perhaps my thinking was “obscured by clouds” (major apologies to Pink Floyd for that). I’ve been saying that publishers and information providers should investigate cloud computing primarily because of scalability and time-to-market opportunities, but after watching a recent stream of announcements around Google’s App Engine, it dawns on me that perhaps the most important benefit is the rich array of tools and services that are being made available by Google, Salesforce.com, Amazon, and others.

In a relatively short period of time, Google has added or announced a number of new updates including support for Java (soon to be followed by Ruby and Groovy), a version of Cr0n, database import functionality, tunneling using SSL, and XMPP support for messaging. Forgetting the alphabet soup of technologies and acronyms for a moment, what is important is that Google App Engine ( and its competitors) are quickly evolving into application development and management platforms that only the largest enterprises could afford to assemble–that is a real “playing field leveller” for small and medium businesses. Even large enterprises can ill afford the time and energy it would take to assemble such an arsenal of tools and services. And for publishers that are saddled with old legacy technology, cloud computing offers the chance to build new applications with state-of-the-art technologies while maintaining old ones, with a relatively painless cutover. That should be enough to part the clouds.

Share and Send

Interesting read by James Surowiecki in the New Yorker about why businesses cut spending during downturns even when history shows that that isn’t necessarily a good ploy. The gist of the article is that companies such as Kellogg, Chrysler, and more recently Hyundai, have benefited from increasing advertising and R&D budgets rather than cutting them as their competitors did. Studies by Bain & Company and McKinsey, among others, have shown that companies that either increased advertising spend or at least stayed flat grew faster than those that cut advertising, for several years after the downturns.

So with such evidence, why would companies continue behaviors that sub optimize their performance during economic upturns? Surowiecki points to economist Frank Knight and his work on risk versus uncertainty. In essence, risk describes a situation where you have some idea of the range of possible outcomes, while uncertainty entails a total lack of knowledge of what might happen. Businesses are quite used to managing risks, but when faced with uncertainty as most currently are, their instinct is to manage what they can which is primarily spending. So businesses are, in fact, acting rationally, just perhaps not wisely, when they cut advertising and R&D spending.

We continue to beat the drum for our “no guts, no glory” theme as we believe that publishers that are “pushing through” to the digital side and making investments in light, agile technology and superior user experience, will be much better positioned to ride the wave of the inevitable upturn in the economy. Companies that, on the other hand, are purely focused on cost cutting and riding out the storm will likely get caught flat footed by their more-nimble competitors. Now that’s risky business.

Share and Send

Remember “one hit wonders?” Sure you do–name any two songs by Norman Greenbaum–”Spirit in the Sky” and..??..Same for Shocking Blue–”Venus” and …?? It’s safe to say that most one hit wonders did not set out to achieve that dubious honor, so what’s the deal? More importantly, is there a lesson that applies to the web?

Thanks to research by Fang Wu and Bernardo Huberman of HP Laboratories and their paper Persistence and Success in the Attention Economy,  we find that there is a sort of parallel. They researched the “success rates” (measured as a threshold of view counts) of uploaders who submitted multiple videos to YouTube over time. Counter intuitively, the overall success rate for most uploaders actually declined after an early success–in other words, there are a lot of modern day one hit wonders on the web. This inspite the fact that the actual quality of uploads often improved with multiple submissions–something that cannot be said for Mssr. Greenbaum’s efforts. Possible explanations include loss of novelty and appeal to the broader audience, although the researchers caution that they have not verified that.

Lesson to be learned? I suppose one is that more can be less and in some cases, more is definitely not better. If your content strategy includes blanketing your customers with information in hopes of building loyalty and viewership, you might want to at least sanity check that it is working. Second lesson is that achieving a “string of hits” is difficult. Most long-lived bands have either hit on a formula and made it work with fairly subtle adjustments (AC/DC), or have continually re-invented themselves as markets and tastes have changed (ZZ Top). Either approach can work, just make sure you know which one you are pursuing.

Share and Send

Lately I’ve been musing about some of the content-related actors that make up the  Web 2.0 ecosystem ranging from the snarks, to the ranters, to the illuminati (not those Mason guys, I mean the “people who light up our lives through their brilliant blogs and tweets”), to the “just plain folks” who like to write and express themselves. It seems like a hopeless jumble, hence my effort to find some measure of structure. I finally decided that a value chain construct was the ticket, and this post attempts to apply the notion of a value chain to the creation and evolution of content on the web. What is odd about Web 2.0, is that arguably the most valuable content is at the head of the chain, while the least is at the tail end, making it a sort of “de-value chain.”

At the top of the chain, creating the most value,  are “originators”, those people who actually contribute unique and meaningful content, whether it be music, words, photographs, whatever. The truly gifted originators are the illuminati I noted above, of which the most gifted can create content which will make you sing “you light up my life.” Note: this is the oldest trick in the book and a cheap ploy to make this piece sticky–you are now doomed to listen to Debby Boone in your head for eternity while recalling this piece. Thanks to the Heath brothers for that one!

Next down the chain are “elaborators”, who take a one way content transmission and turn it into a two way or multi-directional conversation by adding commentary, critique, and additional information and perspective. These folks are kick-started by the spark provided by illuminati (see how this is working) which stimulates additive, but original thinking. Some might argue that elaborators should be at the top of the chain as they broaden the conversation, but in fact, they need the raw material of the originators to do their thing.

Next level down from originators are “synthesizers”–no not the Edgar Winter kind, but instead people who take other peoples’ content and connect it together or mash it together to create new content. Synthesizers may or may not add their own thinking to the mashed up content, but they do add value by interleaving otherwise disparate content to create a new and unique combination.

Finally we get to the carp of the Web 2.0 ecosystem–people I call “snarks.” They troll the web looking for the work of any and all of the above actors, adding often misspelled and grammatically-challenged disparaging remarks. These range from infantile to unintelligible,  but have in common a singular feature–a total lack of content value, hence their place at the tail end of the de-value chain. Imagine if factories worked this way–you’d start with a new car and end up with a wreck covered in graffiti. Web 2.0 is a truly amazing phenomenon.

So there you have it, a complete end-to-end taxonomy of Web 2.0 content participants. And note, in that bizzaro world of the Web 2.0, the value actually decreases from the head to the tail of the chain.

Share and Send

One of the challenges of being an analyst, even a part-time one such as myself, is that you have to keep coming up with new material. I’ve been beating the drum for agile publishing since I wrote my first piece on it back in 2005 and it is starting to feel long in the tooth (even though many are just now discovering it). I’ve been musing lately about “what comes after” agile and I’ve decided it is “proactive adaptation”–the art of anticipating events and reacting to them before they happen. When you stop laughing, let me explain.

One thing I like about agile is that it helps you “skate to the puck” faster than traditional waterfall approaches. In fact, if you are really good, it can even help you skate to “where the puck is going” (sorry, I promised to never use that hoary cliche again but I lied). But what if the puck is making seemingly random turns and movements? That my friends is “continuous discontinuity,” the reason that I believe proactive adaptation is the “next big thing.” If you predicted what is going on in the world today, say a year ago, you can stop reading–you are already there, so to speak–a veritable Black Belt Proactive Adapter. For those that didn’t, much work lies ahead.

For starters–the tools: think mash-ups of trend discontinuity analysis, un-forecasting, black swan mitigation, Taleb’s counter-factual reasoning and others yet to be imagined. I’m being a little facetious here, but only a little. Conventional forecasting and scenario planning have a place and I’m an avid supporter of both. That said, neither help much when black swans such as 9/11 or the current economic meltdown occur. Far more difficult than finding and learning tools is the ability to create a culture able to suspend judgement and belief–in fact the latter is so hard, that some would argue it makes proactive adaptation virtually impossible. More to come.

Share and Send

After spending the afternoon chatting with vendors at O’Reilly’s Web 2.0 Expo, my overall conclusion is that the initial wave of Web 2.0 innovation has crested and some people “caught the wave”, some didn’t. (My office looks out at Mavericks, hence the gratuitous surfing metaphors). The vendor hall included the expected mix of cloud computing, collaboration, and other Web 2.0 applications. What was striking was that most of the showings were variations on the core Web 2.0 themes of mash-ups, blogs, wikis, and social networking–in short, no game changing, breakthrough stuff.

Some interesting vendors and products:

• MotherApp: builds mobile applications with an interesting twist–you build a site in HTML, they send you a completed app in one business day using their MotherApp mobile app generator. Cost is $3K for the app, and $1K for each mobile platform.
• JackBe: company I covered a while back that sells enterprise mash-up tools. They report business is booming as enterprises realize they need to get more agile to deal with the economy. They were and still are a “pick to click.”
• Oneeko: Flash-based screen sharing with Skype integration. No software download, instant setup, and nice UI–what’s not to like about that.
• bluemango: on-line learning systems including Screensteps Live, a tool that enables quick and easy creation of on-line tutorials and guides.
• Static: Integrates twitter, Facebook, and MySpace to enable you to manage and broadcast content to all three at once.
• Besttoolbars: IDE and SDK for building toolbars and plugins to customize your clients’ user experience when using your website or other online apps. Impressive client list with Skype, IBM, Intel, and others.
• Yola: simple (and I don’t use that word lightly),  browser-based drag and drop website construction tools for small companies.
• Microsoft’s Surface: I hope they make a consumer version before I need a new coffee table…

Note: these mentions are not endorsements of the companies or products–merely vendors I found to have interesting stuff. My overall takeaway: Web 2.0 is “mature.” I know that seems like an oxymoron, but many Web 2.0 companies are falling into more traditional modes of business: those that have real products are succeeding, those that don’t are going away; tools are finding homes in both enterprise and customer facing applications; and organizations are able to point to business benefits gained through the use of Web 2.0 technologies. That said, there was good energy and buzz at the conference–Web 2.0 may be reaching maturation, but it isn’t stodgy. Irony moment: spotting a two inch thick book in the O”Reilly store on the twitter API–that’s a whole bunch of tweets…

Share and Send