Ross Mayfield's Weblog
Markets, Technology and Musings

Subscribe to "Ross Mayfield's Weblog" in Radio UserLand.

Click to see the XML version of this web page.

Click here to send an email to the editor of this weblog.


Saturday, November 30, 2002

Timeshifting for Weblogs

One of the members of the Ryze Blog & Bloggers tribe asked for advice on the message board on how to manage the time to start and run a blog.  Lots of good feedback was posted, here was mine:

They say you play soccer the way you are. I think blogging is similar to this self-organizing sport. You blog the way you are.

When confronted with the chronological format of a blog, the pressure to post is at first extreme. How do I start? What if I don't keep it up? Does this go on my permanent record? But the reality is there is no shame in an empty calendar. Post daily, weekly, monthly or occassionally. You blog when you can.

The question that is most personal is what to blog. As one blogger said, "Find a topic and own it." Finding focus is a sure time saver, but it also contributes to the medium, as its one of specialized voices. The more you post on your domain the better. You blog what you are.

The other part, indeed what this tribe is about, is community. When you have others reading, others you know, their feedback and their own posts spark your own. You don't blog alone.

Make a little plan on how you will start, just begin and you will find your rhythm.

The calendar and chronological format of weblogs is threatening to potential bloggers.  For readers its a great structure, akin to timeshifting with your Tivo (or other PVR device).  Readers can view posts from anytime at anytime.  And as news aggregator functions progress, consumption will be less of a burden.

Manila and other content management systems (CMS) offer the ability to schedule posts.  If this feature was offered to Radio and other weblog applications, it would decrease the pressure to publish by letting bloggers timeshift.  This isn't a presence medium, at least not yet, and such a feature as well as cultural acceptance of occassional bloggers would grow the medium.

1:21:25 PM    comment []


Jon Udell reviews Albert-László Barabási's outstanding network theory book Linked and comments on power law distributions and scale-free networks:

But we are embedded in these networks, we can visualize them, we will manipulate them to gain individual and collective advantage. The sooner we acknowledge this power and learn to use it responsibly, the better.

He also plucks David Gelernter's 1991 book, Mirror Worlds which saw a need for modelling & simulation, "to monitor and debug" -- more on that next week.

9:51:04 AM    comment []

Friday, November 29, 2002


I am attending Supernova, the most interesting conference I have seen in years, put on by Kevin Werbach & Jeff Pulver.

Unified by the theme of Decentralization, its on most of the technologies and business models that have made me an optimist again -- weblogs, web services, collaboration, P2P, grid computing & WiFi.  I look forward to seeing old friends and new ones, let me know if you are going.

And the line up can't be beat:

Jeremy Allaire, CTO, Macromedia
Marc Benioff, CEO
Sergey Brin, Co-Founder, Google
Andy Chapman, EVP,
Narad Networks
Nick Denton, CEO, Weblog Media
Cory Doctorow, Evangelist, EFF
Ann Thomas Manes, Author and Former CTO, Systinet
John Hagel III, Consultant and Author
Dan Gillmor, Columnist, San Jose Mercury News
David Hagan, President, Boingo Wireless
Duncan Davidson, CEO, SkyPilot
Morgan Guenther, President, Tivo
Mike Helfrich, VP of Applied Technology,
Groove Networks
Meg Hourihan, co-author, We Blog: Publishing Online with Weblogs
David Isenberg, Principal Prosultant,
Karl Jacob, CEO, CloudMark
Dan'l Lewin
, Corporate VP, Microsoft
Mike McCue, Chairman,
John Parkinson, CTO, Cap Gemini Ernst & Young
Yatish Pathak, Founder,
SOMA Networks
Howard Rheingold, Author, Smart Mobs
Bill Robins, CEO,
Stencil Group
Martin Rofheart, CEO, Xtreme Spectrum
Sean Ryan, CEO,
Doc Searls, Linux Journal
Clay Shirky,
Dave Sifry, Co-Founder, Sputnik
Narry Singh, VP of Marketing, CommerceOne
Rod Smith, VP of Advanced Technology, IBM
Dave Winer, CEO,

I am blogging the conference, but don't expect to be the only one.

8:19:27 AM    comment []

Thursday, November 28, 2002


Thanks to my family for making it all worthwhile

Thanks to my kids for making me believe in the future

Thanks to my friends for making it fun

Thanks to my colleagues for making it interesting

Thanks to the peace we have

Thanks to the year that taught me what I am working for

10:43:59 AM    comment []

Wednesday, November 27, 2002


Im a little swamped before the holiday and then off to a conference in Florida for next week.  Fixed the link to email me in the last post on the survey (thanks Al).

The initial response from the survey has been great.  Will add Meta Blogs, Media & Journalism and Educational Blogs as initial categories.  Almost all of the feedback on categories is that they are "just right" with the exception of Business Blogs (pretty broad), and Other (too broad) and some strong votes for dropping Collaborative Blogs.  Probably won't drop it just yet.  Will add these categories and open up the survey to the larger audience of members.

This initial feedback is that the majority want to be listed in multiple categories, but most see value in categorization so long as its dynamic.  I am beginning to like the idea of dynamic categorization through an open continual survey process.  

Thanks all -- Have a great Thanksgiving, Ross

4:02:49 PM    comment []

Tuesday, November 26, 2002

Categorizing Blogs

A number of great people in the Ryze/Blog tribe have suggested breaking down the blogroll into categories of blogs.  Because the list is growing, categorization is necessary to continue to be a useful resource.  However, because membership is dynamic so must be the categories and I don't want the categories to be arbitrary. 

So I created a little survey.  And its here before I spam all the tribe members.

Here are the initial categories:

  1. Business blogs
  2. KM blogs (Klogs)
  3. Marketing Blogs
  4. Information Technology Blogs
  5. Web Design Blogs
  6. Biotechnology Blogs
  7. International Blogs
  8. Writing Blogs
  9. Political Blogs
  10. Personal Blogs
  11. Collaborative Blogs
  12. Other

3:16:53 PM    comment []

The Autonomic Udell

If there was ever a set of metaphors subject to potential marketing its the biological:

Being and nothingness. Jon Udell: "'[A]utonomic' for IBM has become what '.Net' is for Microsoft: an umbrella marketing term that encompasses everything and nothing in particular." Good analysis by Jon as usual. I think both MS and IBM are onto the Next Big Thing, but they can't figure out how to explain it. The marketing efforts make the confusion worse. One can only hope that the companies listen to the technology and the markets as they go forward, rather than letting the slogans define what they build. [Werblog]

IBM has provided an 8-point definition of autonomic computing and if it sticks to it and journalists like Jon Udell remain vigilant, the value of these functional concepts are less like to dilute.

As grid computing takes hold, autonomic computing plays a vital role in enabling scalable service.  Without complex adaptive systems, patterned after biological theory, managing the sheer number and interplay of elements, applications and policies becomes untenable.  As processing, storage and applications move towards the domain utility service they would experience the same complexity constraints as utility bandwidth service. 

Telecom carriers have been in the business of transforming utility elements into services for some time now.  And in absence of adaptive systems they have experienced systemic failure.  At the network element level, the vast majority of inventory records are inaccurate, despite advances in auto-discovery, leading over 85% of customer orders failing to achieve straight through processing to the point of provision.  As a result, network and service management is largely a manual process.  People don't scale well.  And failure to automate results in high SG&A costs and an unsustainable business model.  Autonomic computing uses decentralized agents to adaptively manage a system, and in contrast to centralized management, scales.

Back to marketing abuse.  Last week I recieved a press release from Cloudmark claiming they cracked the Genetic Code of Spam.   Essentially, Cloudmark uses collaborative identification of spam, the spam message is inspected for common semantics (a "spamGene"), which are aggregated as "spamDNA" and fed to a Bayesian classifier which determines when a message is blacklisted. 

I was about to slam them, but called my friend and life science expert Zack Lynch for some fact checking.  It is a stretch to call Cloudmark a complex adaptive system, let alone "evolutionary."  And its a little rediculous to say they have cracked the genetic code.  However, if you assume that the people enlisted for spam identification, those on SpamNet, are acting as autonomous agents and you buy into Bayesian statistical qualification as evolving, the metaphor holds to some degree.  

From a business model persective, decentralized user feedback also enable the model to scale. These individual users experience vendetta-like satisfaction for pressing the "Spam" button, gain spam protection and enhance the enterprise solution. As the spam wars escalate, competitors with centralized management like Brightmail may experience diseconomies of scale.

11:18:02 AM    comment []

Dimensions of Communication

Jerry Michalski posts some basic dimensions of communication to his Wiki:

All the communication technologies we use -- telephones, newspapers, radio, IM, e-mail, mailing lists, TV, books -- are mired in historical cruft that keeps us from seeing clearly what to build next.  It is useful to go to first principles, then reexamine whatever communication task you have at hand.  So let me suggest the following basic dimensions of communication:

  1. Timing: instantaneous (IM, phone, TV), delayed (e-mail, TV) or slow (books, movies)
  2. Audience: one person (phone call), two (conference call, chat), a dozen, two hundred (mailing list) or everyone in the world (Website, Weblog)?
  3. Mode: text, voice, video, audio, 3D -- some mix of the above?
  4. Length: short (IM) or long (movie, book)?
  5. Persistence: evanescent (IM), temporary but retrievable (e-mail), published (Website, book) or pretty permanent (stone stele)?
  6. Production Level: casual (ASCII e-mail), formal (proposal) or polished (movie, book)?
  7. Identity: how much of your identity do you want to reveal? which identity (work, private)?
  8. Permission: who can see or use the material? 

If the purpose of this theoretical framework is to help differentiate existing modes of communication and identify what to build next, other dimensions would help understand when to shift modes:

9. Centralization: Decentralized (IP-based modes)/Centralized (broadcast & publishing)

10. Whole Costs: cost of the service and user time costs.  Borne by the sender (broadcast & publishing; weblogs), the reciever (email) or both (IM, POTS)

11. Signal to Noise: the level of advertising or spam saturation in a medium.  Broadcast and publishing models reach a natural equilibrium.  Others, particularly when costs are borne by the reciever, are saturated with spam.

9:02:51 AM    comment []

Monday, November 25, 2002

OneNote Personal MetaData

Microsoft announced the launch of OneNote in their Office Suite at Comdex.  Steve Gillmore highlights the potential importance of this product: "Mark my words: OneNote is the new center of the Office universe, relegating Outlook, Word, Excel, and PowerPoint to the edges of the architecture in a single leap. Billed as an Office add-on, in reality it's a smart device programmed to transform Office from a suite of applications to a grid of interactive components."

OneNote is Microsoft's platform for personal meta-data.  The ability to add uniform meta-data in either text, handwritten or audio format on top of web or office documents is very powerful when combined with search and organization features.  If the personal cost of annotation is low to produce requisite volumes of link-rich meta-data, the uses are manifold. Microsoft is supremely positioned to own personal perspectives of information, as Google owns the collective perspective.

You might recall all the companies during the boom that were going to create "Post-it Notes on the Web."  Unfortunately, companies like ThirdVoice opted for more viral models that enabled their use like graffiti on public sites, instead of being useful annotation tools. 

On the web, the closest thing to a link-rich annotation tool may be weblog and today weblogs provide one of the richest sets of meta-data.  Anil Dash wrote a great paper on how Microsoft is entering the weblog market.  And OneNote will be a key component.

There tons of fantastic ideas of how weblogs will evolve into a micro-content client.  A OneNote is link-rich micro-content with ease of production browsable through a number of Office applications.  Microsoft is further along in fulfilling this vision than we would all like to admit.


10:41:49 AM    comment []

Friday, November 22, 2002

Immersion at the Edge

In September, Sony released its VR headset for the Playstation 2 in Japan.  Price point: $500.






Just a couple of years ago, similar motion-sensing stereoscopic headsets cost ten time that amount.  Microsoft's Xbox Live kit ($50) includes audio headsets.   If the future isn't fortold by porn, it is by gaming.  As Moore's Law drives down the cost of Immersion devices at the edge, new possibilties emerge and the barrier becomes content creation.  That's why I am so fascinated by Tele-Immersion, where the content is created by scanning an existing environment. 

Now I always thought Tele-Immersion was a way off from getting outside the lab.  But considering the deployment of Gigabit Ethernet in Metro Area Networks, recent advances in Grid Computing and the falling cost of potential edge devices -- we may be closer than we think.

1:00:34 PM    comment []

Thursday, November 21, 2002

IBM's Autonomic Computing Products
I.B.M. Releases Self-Fixing Computer Software. Autonomic computing is a major component of IBMs On Demand Computing initiative, the part that keeps systemic costs down.  Their first releases are functions in their DB and App Server products.  Look for similar advances to flow down the stack to support Grid Computing.

``All the IT staff does is define these business rules and the systems will then perform to those rules and make sure all the right things happen,'' said Miles Barel, IBM's director of autonomic computing. In addition to setting itself up and running, autonomic computing includes enabling systems to run in the most efficient manner and stay running, fixing itself when something goes wrong.

7:23:32 AM    comment []

Wednesday, November 20, 2002

The Good, the Bad and the Ugly of Homeland Defense

The Good

The Homeland Defense Act forces institutional change for new organizational behavior and  information sharing becomes systemic.  Provides needed funds to secure new spheres.  Many new government acquisitions are for technology and 23% of new funds go to small businesses, areas of the economy that need a boost.  The resiliency of the American demand for privacy and freedom will eventually counter civil liberty infringements.

The Bad

Infringements of civil liberties. The formation of a new committee instead of providing funds at the level that does the works, or as Sen Byrd calls it, "An irresponsible exercise in political chicanery." Consolidating the 22 agencies takes one year and fails as most corporate mergers do.

The Ugly

Pork City.  The last minute amendments to the act included future and retroactive immunity for Big Pharma for vaccine liability.  If a child dies because of mercury concentration in childhood immunizations, when alternatives are available, the family has no legal recourse.  Why?  Homeland Security.

9:10:13 PM    comment []

Grid-enabled Tele-Immersion

Utilizing a grid computing architecture of distributed processing, UPenn computer scientists have been able to achieve real-time Tele-Immersion [Science Daily].  They can to scan, process and present a room in real-time to enable virtual presence:

When they make their first public demonstration of tele-immersion at this week’s Super Computing 2002 conference in Baltimore, computer scientists will also attain another first: a “network computer” that processes data at a location far removed from either input or output....

“Shifting the computing from 10 processors at Penn to 1,240 parallel machines based in Pittsburgh will speed data processing 75-fold, turning tele-immersion into a true real-time technology,” said Kostas Daniilidis, an assistant professor of computer and information science at Penn. “It now takes our tele-immersion system roughly 15 seconds to scan, process and display the entire volume of a typical room. With help from the Pittsburgh Supercomputing Center, that time will shrink to 200 milliseconds.”

The most far-out crazy idea of Internet 2 is finally becoming a reality, or should I say, a virtual reality.  Distributed Immersion as an application has the potential for changing more than how we communicate -- but how we learn, train and design.

11:53:58 AM    comment []

On Demand Innovation Services

IBM Research to offer consulting services. Big Blue unveils On Demand Innovation Services unit [InfoWorld: Top News]

Initially, the On Demand Innovation Services will be limited to four areas, IBM said: Advanced Analytics, modeling scenarios to solve emerging problems; Business Process Transformation, aligning business strategy with IT investments; Information Integration, and Experimental Economics.

IBM research is the largest R&D unit that files the most patents.  Making R&D customer-facing is more than an attempt to grab more services dollars.  It gives researchers to real world exposure to large scale distributed architectural challenges to advance their grid computing initiative.

9:11:27 AM    comment []

Tuesday, November 19, 2002

Occupational Spam

Kevin Werbach [Werblog] posted an article on Slate about how spam will kill email as we know it.  Its a good summary of how 30% of email is commercial spam, the failures of blacklist approaches and the rise of whitelists. 

Blacklist filtering, no matter how the filter is defined (individually, at the server, serviced or collaboratively) will never be fully effective because new circumvention techniques always arise, funded by the economic of spam that favor spammers.  Whitelists succeed, but change the nature of email -- it looses the characteristics of an address.  The costs of first contact will increase, requiring contact through alternative modes of communication or handshake negotiation that creates personal costs doesn't work with some uses of email.  Other solutions, economic and legal have yet to present viable alternatives.  Spam is 2002's biggest consumer leaky pipe, the arms race is on and polution costs will only escalate.

That said, even if total solutions don't exist, existing ones provide value.  Each solution is more appropriate for different segments. I believe MailFrontier's whole product concept approach may be the best for a spamming company and may be the winner in the latest investment bubble.  Im using Brightmail's private labeled service, Earthlink's Spaminator.  A serviced blacklist approach, its about 95% effective with no false positives, which works for me.  For now.

But let me draw your attention to a bigger problem than commercial spam -- Occupational Spam.  Its the unwanted or uncessary email that clogs corporate mail systems, characterized by excessive CC'ing. Gartner and others estimate Occupational Spam to be 30% of email volume.  The problem grows at pace with email, at 30% CAGR. 

What's worse -- filter-based solutions do not work to solve this problem.  Email sent by an employee to another cannot tolerate false positives and must at least be skimmed.  This creates tremendous time costs relative to commercial spam.

According to Business Professor Catherine Cavanaugh, "It's a management issue, if you're getting 50 or more emails a day, you're spending four hours a day just doing email.  That means it's no longer a productivity tool."  Quantify this with the average fully loaded cost of an employee and multiply it by 10,000 employees -- and you have a really big problem.

Occupational Spam is a larger problem than commercial spam.  Both increasingly create costs that will drive adoption of other modes of communication.  And what's worse is we are all loosing time.


4:27:36 PM    comment []

Leonid Meteor Shower

Amature Astronomy [Corante] night at the Mayfield house.  Got up with my six year old.  If I blogged it, it would have been something like "Ooh! Over there!  I saw one!" x 37.  Which is why somethings are best not blogged live.

Astronomy is the study of distance over time as Geology is the study of pressure over time.  Is it me, or are there so many celestial events occuring during my lifetime, that only occur once a lifetime, that they seem to be happening all the time?

9:03:41 AM    comment []

Monday, November 18, 2002

Social Network Activism

The value proposition of technology-driven social network analysis (SNA) is just beginning to be explored.  Until SNA is deeply embedded in software systems, the greatest value is when analysis results in individual and collective action. 

Conversely, SNA can be used to understand the value of individual and collective actions.  These actions are most valuable when they plug network holes:

From John Udell's Seeing and Tuning Social Networks

The general principle at work here, [Valdis] Krebs says, was articulated in Ron Burt's Structural Holes: The Social Structure of Competition. It states that networks with "holes" -- that is, unbrokered connections -- present the most opportunity. A successful actor is one with ties to many points in the network who can uniquely fill one or more of those holes. To that end, Krebs -- who is writing a book on his experiences with social networks and business organizations -- plans to mine Amazon, map out the communities of interest relevant to his themes, and tune his presentation to optimally broker among them.

For Krebs, Metcalfe's Law (network value is N2, for N users) and Reed's Law (network value is 2N, for N groups) are great in theory, but of limited practical benefit:

The six-degrees small world is a fallacy. The small world is two or three steps. I, for example, am supposedly six steps from Madonna. But if I want a backstage pass, it's not going to happen. On the other hand, if I know you, and you know Madonna's manager, there's a chance it will. The practical limit is about three hops. After that, information, for the most part, doesn't travel. We can form 2N groups, but I can belong to only so many.

Krebs calls this effect the horizon of observability. The most effective organizations, he says, are the ones with the best reach -- that is, the richest interconnections within that horizon. He cites a project for IBM that studied fifteen client companies following major changes (mergers, downsizing, etc.), and scored them on how well they handled the change. The winners had two things in common: high reach, and emergent leaders (sometimes formal, sometimes not) who were plugging structural holes.

Structural holes are created by market failures -- in inter-enterprise commercial markets or intra-enterprise knowledge marketplaces [on knowledge marketplaces, see T. Davenport and L. Prusak, Working Knowledge: How Organizations Manage What They Know].

Perhaps the greatest market failure of our time is that of tech employment.  Not just that talented and qualified people cannot find jobs, but so many people are searching and submitting resumes it creates information overload for corporate recruiters.  The role for an information broker in this space has never been so high, however, the buy side of the market (employers) is less willing to pay recruiting fees.  The sell side of the market (employees) are not in a position to subsidize information brokerage to plug this hole.  Or is it?  Enter social network activism.

Take for example the Software Product Marketing eGroup, "The Ultimate Resource for those Seeking Software and Networking Marketing Professionals." Founded by Cynthia Typaldos, who provides the emergent leadership.  Over 1800 marketing job seekers have banded together in a volunteer organization to create a highly targeted channel for recruiters -- for free.  They created an information broker function using Yahoo Groups and developing a website. And by self-subsidizing the cost for its development, have done so in spite of the market failure.  Now when times get tough, you go with what you know.  Product managers within the group developed a shared database.  All that marketing talent has gone further to develop its own marketing, PR and business development initiatives.  The resulting service is a zero-cost highly targeted channel for recruiters and a channel to specialized employeers for job seekers.

For full disclosure, I am a SPM volunteer.  But the point is that this collective action occured due to market failure, activists plugged the structural hole and leveraged convenient social networking technologies.

This social network activism is also what is happening with the formation of a Ryze Tribe for Bloggers.  A structural hole existed between two networks, easily patched with a little activism and convenient technology.  The result, as one Ryzer put it, "suddenly this is beginning to feel like a community."

So look around you.  The greatest lesson anyone can gain is to be aware of the networks around us, structural holes exist and understand that a little activism creates tremendous value.

9:50:31 AM    comment []

Total Information Awareness project
A Snooper's Dream. Washington's efforts to create sophisticated surveillance systems for identifying potential terrorists cannot be allowed to undermine civil liberties. [New York Times: Opinion].  Points out that the man behind the Total Information Awareness project is John Poindexter, who obstensibly claims it could protect privacy.  Reccomends congress shut it down for review.
8:19:57 AM    comment []

Saturday, November 16, 2002

GoogleShare -- Another Googletrinket

Calculating GoogleShare. Steven Johnson came up with the notion of GoogleShare: it's the proportion of pages containing some phrase (i.e., "Boing Boing") that also contain your name. Rael has whipped up an automated GoogleShare calculator. [Boing Boing Blog]

Requires a Google API Key.  Seems I have a 80% share for Googlevote, big whoop. 

You can also calculate GoogleShare for brands instead of names.  Neat.

6:35:37 PM    comment []

Hedge Your House

[eFinance Insider] Protecting the Value of your Home. "Tom Skinner, and his Washington, D.C. based firm, Real Liquidity LLC, is working to address the risk of price declines in residential real estate by developing innovative financial products that allow homeowners to "hedge" out some or all of their real estate market risk with "puts". [Web Voice]

Problem with this model its a single company offering an insurance policy with little backing (high risk of default) and values are according to an index that is open to manipulation through methodology.  If you want a hedge for your house, go to the liquid market, REITs.

9:56:38 AM    comment []

Friday, November 15, 2002

Morn the Seed, Plant the Seed

Must-read Fortune article on venture capital in 2002. [Scripting News]

What really worries [Stanford business professor Hau] Lee, however, is not washouts and wipeouts but the fact that a fresh, new generation of entrepreneurs is finding it nearly impossible to get venture backing. "The idea that this is hurting entrepreneurism is certainly true," he says. "I have a number of students with interesting ideas. In the past they wouldn't have hesitated (to start a new company) and try them out, but now they've opted to go work for a conventional company."

I morn the seed.  Unfortunately, all the Angels have gone to heaven, VC firms have their own problems and the Seed stage market barely exists.  VCs typically lag behind the NASDAQ by about 1 year, so expect more of the same. 

The prolonged contraction has exausted most entreprenuer's abilities to bootstrap.  And bootstrapping is further held back by buyers of technology who could potentially seed by providing cash flow.  Primarily because buyers have shifted to be more conservative in the technology adoption lifecycle[see whitepaper].  This will have a tremendous downstream impact both in terms of economic growth and innovation.   

The ray of hope eminates from greed.  At some point, technology buyers will risk adoption to gain competitive advantage.  At some point, VCs will invest for their own competitive advantage.  At some point, acquisitions to enter new markets and increase share in existing ones will offer valuable exists.

"I don't believe these guys can effectively manage the dollars they have under their belt," she [Ashley Read at Blueprint Ventures] says of the biggest Bubble Funds. They have so much money, she says, that they can't invest several million bucks in a small startup anymore--they wouldn't be able to manage that many investments. So they're stuck investing larger sums in later-stage ventures or seeking other investment alternatives. At that point "it's no longer a venture capital firm," she says. "What is it? I don't know. An investment company? What it is is an accident waiting to happen."

Ashley hints at a greater point.  While some say that the current environment is simply like the way Venture Capital used to be, I beg to differ.  The bubble spawned too many herdlike trend chasers and its forgotten what venture capital used to be.  Great thinkers.  Great risk takers.  There are some VCs who still hold this title.  At some point, from smaller funds or not, a thoughtful contrarian will envision a new company, take risk and gain competitive advantage -- the way it used to be.

1:52:19 PM    comment []

Sun springs for software maker. The company says it will acquire Terraspring to further its N1 initiative to provide businesses with increasingly autonomous computer systems. [CNET]
11:22:59 AM    comment []

Explicit Knowledge Categories

Jim McGee offers a must read 4 categories of explicit knowledge:

  1. Sharing Answers
  2. Sharing Questions
  3. Sharing Practices
  4. Discovery/Innovation

This kind of segmentation is invaluable.  Besides defining the categories, he identifies their key management issues.

It might be a little out there, but I can't help but offer a 5th, maybe as a category and at least as lens for viewing potential use of the categories -- Revealing Conflicts.  Some of the best tacit knowledge is made explicit through debate, the socratic process of people taking sides on an issue.  When I have a decision to make, lets say on how to do something like define an incentive compensation issue, access to answers helps offer options.  Questions offer some considerations, but are likely not in context with the answers I am considering.  Practices, as Jim points out, lack context as well.

Information and knowledge has no value until it informs a decision.  Until then its just an option.  You exercise both the option of consuming the knowledge at the cost, or premium, of time to read it and hard costs of acquiring it.  The problem arises if you base your decision on categories of explicit knowledge that are "Pros,"  you skew your mental probability distribution of the outcomes of a decision to one side.  If, however, "Pros" are balanced by "Cons" the distribution will be more balanced.  Pros and Cons together also define the context for an issue at one point in time.

But the problem with using conflict as an asset is that many cultures purposely avoid it, will not document it and will even bury it.  Unfortunate, because the best answers, lessons and context come from healthy debate.


11:20:13 AM    comment []

Thursday, November 14, 2002

Greatest Infringement of Civil Liberties our History

If we aren't already, we will soon experience the greatest infringement of civil liberties US history.  The executive branch has already empowered itself with increased surveillance, the ability to hold suspects indeterminably and try them in military courts.  Now the GOP is claiming a mandate from a rather narrow mid-term election to pass the Homeland Security Act.  But its not that war and security was what won the election, its the absence of critical mass on other issues.  What is deeply concerning is we are giving up not only our liberties, but our privacy for uncertain benefit.

Boing Boing quotes a William Safire op-ed:

Every purchase you make with a credit card, every magazine subscription you buy and medical prescription you fill, every Web site you visit and e-mail you send or receive, every academic grade you receive, every bank deposit you make, every trip you book and every event you attend — all these transactions and communications will go into what the Defense Department describes as "a virtual, centralized grand database."

Now I don't want to make this a poltical blog, but if there is an issue we should all blog about its this.  Decidedly absent from recent moves by our government has been real debate on the infringement of civil liberties, whether its effective or even constitutional.  Mitch Ratcliff quotes best: 

"They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." - Benjamin Franklin

Mitch goes on source and summarize that "From the Homeland Security Act text, the government has the right to any information it wants to examine given it perceives a "credible threat," a term not defined by the bill -- and that which is not defined by the law can be abused by the government to an unlimited degree"

We are creating our own problems here.  Once this data is being tracked and captured it will take a significant backlash to stop or erase.  And if our government can't provide a link between Sadaam & terrorism (weapons of mass destruction come at great cost and provide Iraq strategic parity, why would they give them away to others who have different objectives?), imagine the links they can make between you and god knows what.

3:48:23 PM    comment []

The Inverse of Googlism. GooglePeople will take a "who is" question (like Who is the founder of Google?) and give you back a list of potential people. Fun!... [Google Weblog]
3:21:08 PM    comment []

Wednesday, November 13, 2002

Dissecting PageRank

"You see, most blokes will be playing at 10. You’re on 10, all the way up, all the way up...
Where can you go from there? Nowhere. What we do, is if we
need that extra push over the cliff. 11. One louder."
- This is Spinal Tap

SearchNerd dissects Google's PageRank display. 

pos.gif value

The true Googlenerd will also appreciate how Google gives itself an 11 on a scale of 10.

8:17:04 PM    comment []

Enterprise Search Market Consolidates

Verity Buys Inktomi Enterprise Search. In an unexpected move, Verity has announced that it is purchasing Inktomi's enterprise search software business (formerly known as "Ultraseek") for $25 million in cash and assumption of some debts. [SearchTools News for 2002]

Why? The Google Search Appliance of course.

2:51:51 PM    comment []

Marketing Cycles

One of the interesting things about living through an extreme business cycle is extreme changes are more noticable.  The marketing function itself, shifts wildly in its activties as I commented on with a focus on ROI.  But there are larger shifts at play as well.

Organizational Structure

  • The boom period was marked by two organizational trends for marketing: centralization of some functions and customer-focused organization.  Advertising, New Product Development, Market Research, Sponsorship and Events and Major Promotions were centralized to realize economies of scale.  Merchandising, Local Promotions and Channel Management remained decentralized, falling under the Business Unit VP of Marketing.  For more on these trend see a Marketing Leadership Council survey
  • But the bust that followed the boom has led to increased spending in decentralized activities, as the value of existing customer relationships is at a premium and these methods are more effective.  Centralized organizational structures will likely remain, but their primary role is in financial controls.  The payments, and to some degree the bits -- leaving the atoms to be managed in the periphery [Using Mohan Sawhney's language].  Now budgets are power, power is politics and organizational structure is highly influenced by politics -- so don't be suprised to see a decentralization trend.


  • During the boom, a premium was placed on the ability to communicate to generate new customers, or outbound marketing.  This was driven by the percieved value of market share (a common belief that there was more than 100% of the market to be had), lifetime new customer valuation (a common belief irrespective of churn), and the low cost of capital for use in promotion to achieve share. 
  • During the bust, despite the decline in advertising prices, the ROI of promotion declines relative to product, place and price -- the domain of inbound marketing.  There are exceptions, specifically direct marketing activities continue to achieve relative ROI.  But the relative cost of the Inbound function (people making smart decisions based on investments to get good data) and return from optimizing the product line outperforms when you have to get more out of what you have got.  The good news is technology companies are beginning to invest in R&D again, which will drive the creation of product management and product marketing jobs.  The cycle continues to benefit MarCom.


  • During the boom, customers have more to spend and are more receptive to how companies define their product.  Investment in advertising, events and major promotion to directly influence customers makes sense.
  • During the bust, the wallet thins and transactions scrutinized.  The particularly excessive vendor claims during the boom now have customers doubting the competitive advantage of technology.  Customers believe in themselves first, 3rd parties second and vendors last.  Today vendors must focus on 3rd parties, influencing the influencers -- through PR, customer references (Don't get Siebeled!), word of mouth.

1:22:44 PM    comment []

New Marketing Fix

Rick Bruner, John Engler, Steve Hall, Robert Loch and Oliver Travers,  have launched Marketing Fix -- a collaborative weblog on e-Marketing news.  Seems to be B2C and Ad oriented. It fills a much-needed niche and is now one of my daily reads. 

Robert and Oliver also collaborate on End of the Free (a must read).  The majority of them are also at

11:08:40 AM    comment []

S+B's Best Business Books

Great expert reviews summarize Strategy + Business' top business books (53):

Strategy by Chuck Lucier and Jan Dyer
Management by David K. Hurst
Ethics by Frances Cairncross
The Biggest Management Book Ever by Kenneth Roman
Leadership by Bruce A. Pasternack and James O’Toole
Women Leaders by Kate Jennings 
Globalization by Rob Norton
New Europe by David Newkirk
Management's Renaissance Man by Charles Handy
Networks by Michael Schrage
Best Business Books 2001-2002

So much to read.

10:37:12 AM    comment []

Tuesday, November 12, 2002

WiFi Workplace ROI

This article in Fortune magazine highlights WiFi installations at Novell and talks about the value of WiFi at work [original pointer from: Windley's Enterprise Computing Weblog].

"Today, you don't sell technology--you sell ROI," says Paul Fulton, former general manager of 3Com's wireless-LAN business and now executive-in-residence at venture capital firm Mayfield [no relation to Ross]. "You have to say this is going to raise revenue or lower cost."

Here are the ROI highlights from the article:

Hard Cost Substitute Benefits -- ...a wireless LAN deployment for a new office of between 50 and 100 people costs $20,000, vs. $250,000 to get a wired LAN up and running. What's more, he says, "you never know when you're going to move an office, so all the cabling you put in could be wasted."

Increased Productivity -- Employees spend an average of 3 hours a day in meetings, and some users claim increased meeting productivity.  An Intel study of its own 800-person deployment found that WLANs deliver 23 minutes of additional productivity per day. Most employees interviewed for this story said that Wi-Fi "saved" them between a half-hour and 90 minutes per day.

Decreased IT Administration Costs -- Mostly through elimination of LAN administration tasks.

Employee Satisfaction -- In a Cisco-commissioned study of more than 300 organizations, 87% of WLAN users said their overall quality of life had improved. Improving employee satisfaction is the cornerstone of the service-profit chain: employee satisfaction --> quality & service --> customer satisfaction --> loyalty & profit.

New Potential Costs -- "We have parallel conversations running at my meetings, and I'm hoping that they are finding solutions, but they are probably complaining about how unreasonable I am," quips Intel CIO Doug Busch. "When there are laptops everywhere, culturally it can create a whole lot of problems," says Novell CIO Debra Anderson. "We're finding the balance between Wi-Fi as an intrusion and as a powerful, productive tool."

Intangibles -- The article doesn't touch on this too much except to highlight the concept of convenience.  An opportunity exists to measure benefits according to opportunity costs.

10:13:47 AM    comment []

Klau's Klog Trial
Rick Klau shares results from his company klog pilot of Radio.  He chose a small set of users from various divisions which resulted in various uses.  Takeaways include a general satisfaction, potentially greater value if deployed to address specific problems and adoption issues common to new modes of communication.  Seems like if deployed to a larger set of users, with consumate packaging and introductory training, social support and social feedback on use could resolve any of these issues.  Really insightful, worth a read.
8:37:22 AM    comment []

Monday, November 11, 2002

Numbers Games

Google's Top 100. What are the most linked-to pages? If you do a search for HTTP (as in, http://) you get something pretty close to the answer. It's probably not exact (Google and Yahoo are reversed from their Google Directory rankings, for example) but it's pretty interesting. How close are you to one of the big names? I'm two hops from number 21.... [Google Weblog]

What would be really interesting to know is the ratio imbalance of links to traffic.

Phil Wolff recently suggested a need for a census of blogspace.  Why not Google by blog vendornames seperated by "or" to get a rough estimate of the number of blog pages?

UPDATE: [Source: Scripting News] A Google developer who asked not to be named says that yesterday's theory that searching for "http" gets you a list of sites in order of PageRank is incorrect. It gets you a list of sites that are most-linked-to with http in the link, like this: He says "A very small fraction of links on the web look like that, but the more popular sites have more of them."

10:06:37 PM    comment []

Measuring Marketing ROI

Never before has there been as much scrutiny on the ROI of marketing activities or changes in the activities themselves.  In the downturn of a business cycle, marketing is often the first department to suffer cuts.  Within this cyclical process activities and campaigns are rationalized with greater focus on measurable cost/benefit.  The difficulty, of course, is that the greatest returns from marketing invesments are often long term or intangible. 

Marketing is undergoing a siesmic change, primarily driven by new technologies and customer shrewdness.  New modes of communication allow new opportunities to interact with customers.  Campaigns can be tracked and even automated.  There has never been so much data available on customer interactions, preferences and perception.  However, customers know they are being tracked and benefit from increased market transparency.  Saavy customers have driven changes in what activities are effective, which presents complications for measuring ROI. 

Here are the recent shifts in marketing activity and how they relate to measuring value:

  • The Shift From Advertising to PR: Consumer sensitivity has decreased the effectiveness of advertising, especially relative to the third-party validation that PR provides.  Advertising is easier to track, especially interactive forms or those designed with lead generation feedback, and reasonable statistical inference can project returns on investments.  PR, on the other hand, is notoriously difficult to project.  There is greater variability on if a journalist will pick up a story pitched by your PR firm, let alone slant it in your favor.  With the greater returns for PR come greater risks.
  • Direct marketing & Telemarketing on the rise: In the bottom of the business cycle, direct marketing rises to the top of activities -- primarily because it is easy to project and is ROI driven.  However, customer sensitization has reduced response rates.
  • Event marketing suffers: When the economy's growth isn't rapidly creating new customers to meet, the return on event marketing declines.  Smart event marketers are moving to smaller seminar formats over broad conferences, and emphaizing lead generation over expensive brand placements.
  • Product marketing becomes paramount: Relative to the costs/benefit MarCom, having the right product for the right customer segment, differentiated from competitors at the right price yields the greatest return.  Problem is, these are somewhat internal activities and many companies dont invest enough in the right people to support this as a process.  Tracking returns is actually relatively easy, especially through giving product managers P&L responsibility for the products they manage.
  • Word of Mouth & Thought Leadership:  In serious decline because benefits are hard to measure in the short term.   Getting creative on how data can be gathered and applied to an existing ROI methodology to support these campaigns could yield differentiated returns.
  • Channel Marketing: Biz Dev died as an occupation because of the lack of focus on ROI in activities.  Gone are the partnerships for press-release sake, to the dustbin of illusiory perception.  Now the focus is on channels with direct customer relationships generating revenues.  Easy to track, sales based, and in a methodology that should be agreed upon with the partner from the outset.  Problem arises when opportunties for whole product partnerships are missed in exchange for shorter term and smaller returns.

The best marketing activities are unique, requiring creativity in design and execution, and dont necessarily fall into the above buckets.  Developing performance metrics and getting agreement on an ROI methodology is a key aspect of designing activities that is even more challenging and critical in today's market

1:07:41 PM    comment []

Sunday, November 10, 2002

The Performance of Stock Options

When Options Rise to Top, Guess Who Pays.  By Gretchen Morgenson. [New York Times: Business]

Joseph R. Blasi and Douglas L. Kruse, professors of human resource management at Rutgers, examined stock option grants and shareholder returns at the 1,500 largest American companies from 1992 to 2001. They found that companies dispensing significantly larger-than-average option grants to their top five executives produced decidedly lower total returns to shareholders over the period than those dispensing far fewer options...

The 375 that gave the most to their top executives — more than 40.8 percent of all options — performed worst, returning 22.5 percent over all to shareholders through 2001. The 375 companies that gave the fewest options to their senior executives — less than 19 percent — fared the best, giving investors a 31.3 percent return, on average.

Options are the only proven compensation incentive to generate long-term performance.  The problem is that in practice, this study shows, the distribution of options is heavily skewed towards executives over employees.  What's worse, its a pattern of systemic complicity that will be hard to reform:

"The assumption that the system is better for everybody by giving most of the pie to the top of the hierarchy is an assumption that is widely accepted by lawyers, accountants, Wall Street investment bankers and even by many academics," Mr. Blasi said. "But when you compare companies against each other, the more you increase the option grant to the top five executives above the mean, the worse your shareholder return gets."

The tech industry has a history of distributing a greater share to the rank-and-file.  Unfortunately the industry is loosing the options accounting battle.  Options will be expensed.  The valuation game that remains, of how options are priced, is yet to be determined.  But the larger corporate governance issue looms, relying on institutional investors to pressure executives to distribute options to the very people that need motivation.  Lets hope the baby doen't get thrown out with the bathwater.

9:14:50 AM    comment []

Friday, November 08, 2002

Blogging to Prevent Email Overload

You know klogging has a future when smart people consider blogs as a solution to the pain of information overload.  A product manager at a major software company posted an informal RFI to a discussion board:

Here's the situation: I'm managing four

> products that are sold by about 300 sales engineers and 300 sales

> people worldwide. That translates into a lot of sales-related

> questions sent via email. There is some degree of overlap to these

> questions. Currently I'm answering these queries by responding

> individually to each email, and this takes a big chunk of time...

> Given this situation, I'm thinking of setting up a weblog (aka blog)

> to post answers to questions likely to be of interest to a

> significant chunk of the sales team.

This validates two threads of mine:

1) The time cost of email is driving adoption of other modes of communication like blogs and online communities like Ryze.

2) Blogging is at the early adopter phase.  Today the majority of users are programmers or writers. For blogging to cross the proverbial chasm, the whole product needs to support the needs of business users with leaky pipes in information management. 

The great thing is that some customers realize they can both fix the pipe and scale plumbing expertise.

7:23:25 PM    comment []

Thursday, November 07, 2002

Why Ryze?

Launching a tribe on Ryze for Blogs and Bloggers needs a little justification.  We are in the early adopter phase of both weblogs and explicitly open social communities and commercialization concerns pervade.  Ryze is a business and blogs have arm merchants...that and the users are why they have evolved to be so useful.

Ryze provides something other places don't -- explicit social identity.  When you combine that with link transparency and continual user friendly integration with other modes of communication, it yields powerful network effects.  What a business.  What a tool.  Look for Ryze identity to continue to extend.  Blogs are more content than identity centric. 

Unlike Ryze, where identities are built through linkable interests and inter-personal ties -- blogs build identity primarily through content and context.  An anonoblogger with much to say and reference is practically an identity, but the ties are based upon less trust.  Bi-directional nested feedback loops would certainly complement the blogging medium.  But the beauty of blogging is such trust links are negotiated in one-off fashion, evolving organically.

Both Ryze and Blogs require you to share part of yourself in return for exchange.  They serve different functions, that of identity based relationships and content driven expresssion.  What is facinating from a social network perspective is the intersection that is the tribe and the potential convergence of convenience.



10:42:59 PM    comment []

hung up: strangely the post I made this morning didn't go up until I later edited end reposted tonight.  radio news aggregator wasnt working too. a little like speaking on a cell phone and not realizing you have been disconnnected until you are finished with your diatribe.

10:23:53 PM    comment []

Chasm Group on Product Management

Last night I attended a talk by Paul Wiefels of the Chasm Group, the author of The Chasm Companion [amazon] for the Silicon Valley Product Management Association.  Paul's book is a field guide that complements the Crossing the Chasm [amazon] series by Geoff Moore. 

The book is a must read for us technology adoption lifecycle geeks, but also for anyone dealing with the challenge of selling and marketing discontinous technology today.  Filled with post-bubble updates, frameworks, strategic checklists, practical insights and worksheets.

I won't repeat much of what is available in the book, but highlight other insights with pointers to additional content.

Buyers don't believe that IT provides competitive advantage [another survey showed this too], so drop it from your lexicon.  Today's market is characterized by:

  1. An overhang in technology capacity relative to other sectors.
  2. Sellers are competing for marginal budgets, there is no line item left for you, they have all been blurred.  Further, 40-50% of budgets are for maintaince [see MS CIO survey post]
  3. "Some Assembly Required" is driving a customer revolt
  4. Increasing focus on the Investment portion of ROI.  Key measure is payback period, followed by bottom line (EBIT) monthly impact.  TCO is overhyped, remember that the product with the best TCO doesn't necessarily yield a return. [see whitepaper for another take on this subject]
  5. Pain-based messages resonate.  Focus on fixing leaky pipes (broken critical processes).

Product managers must:

  1. Perform a strategic assessment (are we vulnerable to market change, what is our category fitness and trend, market share, viable strategy, ability to understand these issues and execute),
  2. Recognize that discontinous innovations require infrastructure to be adopted and montetized.  When a technology is completely diffused, it is the adoption of a new infrastructure/platform/standard.  Make a list of who is for and against the new standard.
  3. Focus activities according to the technology adoption lifecycle (e.g. when and how to fulfill the whole product concept)

Two most common false assumptions of the technology adoption lifecycle/diffusion theory:

  1. The distribution is not a timeline
  2. There are chasms between segments

In a question, he confirmed my view on how the business cycle effects the technology adoption lifecycle.  Buyers have become more conservative, a population shift in the psychodemographic segments of the lifecycle.

10:43:46 AM    comment []

Wednesday, November 06, 2002

Blog Tribe on Ryze

One of the best real world applications of social network theory is Ryze, whose purpose, it turns out, is business networking.

Last night I created a Tribe on Ryze for Blogs and Bloggers.  In the first 12 hours, these bloggers have joined: Peter Kaminski Matt Mower Roland Tanglao Oliver Travers Jerry Michalski Edward Vielmetti Zack Lynch George Nimeh Mitch Ratcliffe

It will be interesting to see how this cluster evolves and how it connects two communities.

11:18:01 AM    comment []

Clarification: Mapping Trust and Other Social Networks

The seminal article I mentioned in my post on Trust Networks that lays out 3 networks (Trust, Advice, Communication) is Informal Networks: The Company Behind the Chart by David Krackhardt and Jeffery Hanson, HBR 1993. 

Karen Stephenson lays out 6 networks which overlap with the original 3.  Her Work Network and Learning Network are similar to their Communication Networks.  Her Social Network and Strategic Network are similar to their Trust Networks.  Her Expert Knowledge Network and Innovation Network are similar to their Advice Networks.

Her identification of different modes of inter-personal relationships is invaluable, and based on significant experience, but of course increases the complexity of analysis.  The cost of social network analysis is also furthered by reliance on surveys for data, which are one-off rather than continous or automatic.  Benefits of social network analysis are also hampered by the current means of presenting results, usually in the form of a report or consultation with select managers.

The real value in social network analysis will be realized when the process is automated to realize the speed required for map these dynamic networks and continous to realize nested feedback loops.

11:11:46 AM    comment []

Googlevote worked


Apparently, the Googlevote methodology accurately predicted the Republican Party's success in issue framing of war vs. economy.

10:49:42 AM    comment []

Tuesday, November 05, 2002


Its Election Day and you haven't been paying attention.  Well that's okay, its only midterm elections and you can make it up on the final.  But just in case you haven't formulated your own opinions, you can easily Googlevote your decisions with the following steps.

First, Googlefight the two candidates.  In Gray Davis vs. Bill Simon for California's governor, it picks Bill Simon as the hands down winner.  But if you think this is simply a case of "no such thing as bad press," skip to step two.  If you can't remember the name of the other candidate, say facing Jeb Bush in Florida, Googlefight him against "the other guy."  It's clear you should vote for "the other guy."  This first step will ensure you don't give your vote away to a less popular candidate, however, to really understand what people are saying about the candidates, skip to step two.

Second, Googlism each candidate's name.  Here you will learn such critical matters such as "Gray Davis is sweet and speed" and such conspiracy theories as "Gray Davis is busily throwing your tax dollars away on his secret electricity."  You will also learn that Bill Simon is "just the man" and "the Gordon Gekko of his generation."  But if the answers don't satisfy you, you can always Googlism "the other guy."  If you are still unsatisfied, go to step three.

Third, Google Images the candidates.  Perception matters and you don't want to elect the ugly candidate.

Fourth, Google News  the candidates and research the issues.  Maybe your opinions actually matter.  In this election, your vote counts more than a member of the electoral college.  And the issues matter -- especially if you live in Nevada.  There you can vote to legalize maraijuana and spark a revolution in property rights.  The property rights issue especially pertains to those enriched Silicon Valley tech execs who chose political exit to Tahoe for tax avoidance, where a referendum would make local governments compensate property owners for environmentally friendly zoning laws that decrease their property value (nevermind the total social costs, I guess).

On the otherhand, Googlefighting the issues already reveals that the Republicans may have won the election so you can go on being your apathetic self:

(49 100 000 results
(16 700 000 results)

The winner is:    war

8:59:50 AM    comment []

Monday, November 04, 2002

Trust Networks

Strategy + Business features Social Network Theorist Karen Stephenson [registration required], emphasizing her work on trust networks:

Professor Stephenson’s quantum theory of trust holds great potential as a diagnostic method for the unquantifiable aspects of business. Imagine that at any given moment, you could analyze the health of an organization’s networks. For instance, a company might have a healthy work network (with a great deal of open information flow about processes and very little workaholism), a medium-grade social network (with little real contact but also little pressure), and a low-quality network for what Professor Stephenson calls “continuous improvement” — the ability to innovate new processes easily. Any organization can be stunted in one of these areas and bountiful in another.

Professor Stephenson suggests that most organizations do not remain static. Their network health profiles continually change. An organization’s path from one network health profile to another not only is predictable, she says, it can be influenced. There are archetypal patterns that repeat, over and over, and, depending on the prevalent pattern, make it possible for one company to thrive where another fails. A startup technology company might begin with a low work/high social/medium improvement profile, as people first get to know each other. Then, as venture capital and deadlines kick in, the profile would move to high social/high work/medium improvement. And then there might be a betrayal by one of the senior executives. At this moment, the fate of the company’s networks hangs in the balance. Does its improvement capability, for instance, go up or down? Does its social capability flatten to the point where people leave the company? Or can the strength of the networks, fortified by the trust people feel for one another, override the crisis?

She also expands upon the original 3 networks (trust, advice and communication) in her surveys and analysis:

Six Varieties of Knowledge Networks

In any culture, says Karen Stephenson, there are at least six core layers of knowledge, each with its own informal network of people exchanging conversation. Everybody moves in all the networks, but different people play different roles in each; a hub in one may be a gatekeeper in another. The questions listed here are not the precise questions used in surveys. These vary on the basis of the needs of each workplace and other research considerations (“Don’t try this at home,” says Professor Stephenson), but they show the basic building blocks of an organization’s cultural makeup.

1. The Work Network. (With whom do you exchange information as part of your daily work routines?) The everyday contacts of routinized operations represent the habitual, mundane “resting pulse” of a culture. “The functions and dysfunctions; the favors and flaws always become evident here,” says Professor Stephenson.

2. The Social Network. (With whom do you “check in,” inside and outside the office, to find out what is going on?) This is important primarily as an indicator of the trust within a culture. Healthy organizations are those whose numbers fall within a normative range, with enough social “tensile strength” to withstand stress and uncertainty, but not so much that they are overdemanding of people’s personal time and invested social capital.

3. The Innovation Network. (With whom do you collaborate or kick around new ideas?) There is a guilelessness and childlike wonderment to conversations conducted in this network, as people talk openly about their perceptions, ideas, and experiments. For instance, “Why do we use four separate assembly lines where three would do?” Or, “Hey, let’s try it and see what happens!” Key people in this network take a dim view of tradition and may clash with the keepers of corporate lore and expertise, dismissing them as relics.

4. The Expert Knowledge Network. (To whom do you turn for expertise or advice?) Organizations have core networks whose key members hold the critical and established, yet tacit, knowledge of the enterprise. Like the Coca-Cola formula, this kind of knowledge is frequently kept secret. Key people in this network are often threatened by innovation; they’re likely to clash with innovators and think of them as “undisciplined.”

5. The Career Guidance or Strategic Network. (Whom do you go to for advice about the future?) If people tend to rely on others in the same company for mentoring and career guidance, then that in itself indicates a high level of trust. This network often directly influences corporate strategy; decisions about careers and strategic moves, after all, are both focused on the future.

6. The Learning Network. (Whom do you work with to improve existing processes or methods?) Key people in this network may end up as bridges between hubs in the expert and innovation networks, translating between the old guard and the new. Since most people are afraid of genuine change, this network tends to lie dormant until the change awakens a renewed sense of trust. “It takes a tough kind of love,” says Professor Stephenson, “to entrust people to tell you what they know about your established habits, rules, and practices.”


3:19:51 PM    comment []

Morgan Stanley CIO Survey

Morgan Stanley's Chuck Phillips discusses CIO priorities in reference to their September 2002 CIO Survey in Optimize Magazine.  Survey highlights:

  • One third of the 225 CIOs polled have cut their IT budgets since January, but one in five still plans to do some shopping this quarter. That1s fewer than the 50% who traditionally go on a spending spree in the last quarter to use up their budget allocations. Nevertheless, Morgan Stanley sees it as an encouraging sign that sellers will see a 3mild budget flush2 this month and next.
  • CIOs are much more optimistic about an economic recovery than they were in June. Almost all of the respondents (95%) say they expect an economic recovery in 2003Ðup from just half of the CIOs in the June survey. Still, the number who have a positive feeling about their own companies1 recovery fell slightly, to 56% in September from 60% in June.
  • Half of the companies surveyed say their hardware spending will likely rise in 2003; a third say they1ll increase network-equipment and server expenditures. Application integration, security, and Windows XP upgrades will be continuing IT priorities.

ERP has found its footing while wireless, portals, document management, and database software are advancing fast
Sept. 2002
July 2002
Jan. 2002
Net change
Application integration 1 1 1 (tie) --
Security software 2 2 1 (tie) --
Windows 2000/XP
upgrade, desktop
3 4 4 +1
ERP software or upgrade 4 3 8 +4
Wireless initiatives 5 17 11 +6
E-commerce initiatives 6 5 3 -3
Employee/enterprise portal 7 14 15 +8
Windows 2000 upgrade, server 8 8 7 -1
Document-management software 9 14 22 +13
Microsoft Office upgrade 10 (tie) 6 16 +6
Network equipment 10 (tie) 11 9 -1
Database software 12 24 29 +17
DATA: 225 respondents to Morgan Stanley CIO Survey, September 2002

Prevalent in these priorities is applciations with time saving value propositions (Portals, Document Management) for the increasingly overburdened worker, the WiFi trend, and maintainance and upgrades in ERP, Databases and Microsoft (effect of their new licensing scheme?).   Chuck makes another important point:

"The 10 top software companies represent 109% of the profit and 54% of the revenue. And if the economy doesn't turn soon, the spoils are likely to be even more concentrated."

9:20:24 AM    comment []

Sunday, November 03, 2002

Intel's Communications Effect

A great article in Fortune on Intel's investments into greater manufacturing capacity and Digital Signal Processors (DSPs).

"...the telecom industry has never really benefited from Moore's Law because most of those proprietary network processors weren't made in large enough volumes to warrant leading-edge manufacturing technology. A standard part made with state-of-the-art manufacturing technology, coupled with a set of standard programming tools that are derived from PC software tools, could finally bring Moore's Law to telecom," said Sean Maloney, executive vice president for the Intel Communications Group.

Actually, Telecom has achieved Moore's Law in optical transport; price/performance gains below Layer 3 that accellerated the supposed bandwidth glut.  This helped foster a new traffic engineering paradigm, achiving service quality through over-provisioning capacity (vs. centralized intelligent management of channelized capacity).

If you assume Intel brings Moore's Law to Layer 3 and up, it would change the way traffic is engineered.  It might bring down the hop-tax for traffic.  It might allow scalable intelligence to be applied at edge devices without the cost of complexity.  It might allow quality to be guaranteed through techniques that have yet to be widely deployed like MPLS.  All while decreasing the cost of boxes that are the vast majority of carrier CapEx.

The problem is telecom CapEx is that the markets are not going to allow capacity in excess.  Smart and faster builds according to pre-sales will be required.  And over-provisioning capacity to assure quality will be called into question.  If Moore's Law brings down the cost of boxes and systems houses like Cisco enable more granular upgrade paths, a more variable cost structure will emerge that can be accepted by the markets.

But the problem with the service provider business model isn't just CapEx, its OpEx, namely people and billing.  SG&A costs that average at 25% of revenues, the highest of any industry.  And billing costs at 20%.  If telecom is to become sustainable, software solutions to this problem are the key.

Perhaps looking to the IT industry's experience with Moore's Law provides insight into the changes to come.  When Moore's Law drives the physical commoditization of Datacommodities (storage and processing in IT), first as hardware solutions (PCs) and then as services (utility computing), players change roles.  Intel increasingly takes on the functions of system houses (IBM, Sun, HP).  Systems houses move into software (data center virtualization, utility computing, autonomic computing).  Software providers move into services (IBM's acquisition of PWC consulting, HP/Compaq merger driven by services).

If Moore's Law drives the physical commoditization of the other Datacommodity, bandwidth, it will accelerate a similar shift.  Intel will increasing fabricate the engineering functions previously provided by system houses like Cisco, Lucent and Nortel.  As network elements are built upon a defacto standard set of components, they too will trend towards commoditization.  Increasing intelligence in network elements will allow greater virtualization for management.  As margins errode in their core business, systems houses will further standardize element interface and open their operating systems to eliminate their barrier to entry OSS (proprietary element management) [admittedly this is the weakest link in the scenario].  Systems houses will be well positioned to offer new network management platforms and move up the value chain.  Virtualization and autonomic techniques will drive down cost of people and billing.  Software providers will increasingly move into services (similar to what CDN & VPN providers have done). 

Regardless of the value chain shift, the move towards increasingly intelligent edge devices at lower cost would enable new revenue models.  MPLS engineered traffic that is service and application specific with QoS guarantees between service providers (autonomically negotiated and settled)  could offer a yield management driven revenue model.  Making these network services available at the edge, for access and web services, enable growth at the edge.

Other blogs are highlighting how this effects wireless broadband: Steve Stroh, John Robb, Kevin Werbach.  And Brad Delong says "In the future it sees, it destroys several industries--and creates entire new ones

10:25:33 AM    comment []

Friday, November 01, 2002

Profiting in a Datacommodity Industry: I


While I am working on a whitepaper on the larger subject, a comment on my post of IBM's announcement [also see their whitepaper: Living in an On Demand World] prompts me to elaborate: 


Recently prominent IT companies (IBM, Sun, HP, Microsoft) have announced developments in utility computing, grid computing, virtualized data centers and web services.  The importance of these new models cannot be overstated – they are enabling a new phase of networked computing for efficient management of Datacommodities.


Storage, Processing and Bandwidth will trend towards commoditization while Software and Services will trend away from it.  The technical standardization of Mbs, MIPS, Mbps, & MHz demanded by customers to reduce their technology risk and for vendors to realize production economies create the characteristics of physical commodities. Storage, Processing and Bandwidth can be considered Datacommodities.  The turbulence of the business environment and consumer tastes, on the other hand, requires constant change. Software trends towards commoditization as well, specifically towards standard protocols and components.  But the custom assembly of these components, adapting systems to this environmental change is the domain of software and services, a trend away from commoditization.


The Tech Industry has always resisted commoditization, by creating new overlays, bundles and marketing tactics.  But a fundamental change has occurred that compels a new strategy of embracing commoditization.  A critical mass of Datacommodities have been standardized as fungible physical commodities and made accessible through communications infrastructure and standard protocols.  Since this mass exists, there is no going back.  This also comes at a time where customers have greater power over vendors to unbundled solutions than before.  And new network, data center and web services management systems that virtualize underlying commodity components allow them to aggregate bundles according to their requirements at any given time.  With component commodities accessible and bundling shifted to customers, this customer-driven commoditization is a reality the industry must face.


Although it seems counter-intuitive, there is the potential for even greater profit in the Tech Industry than before, but it requires change.  Providers of technologies that are trending towards commoditization need to adapt their business models to that of lean infrastructure providers while providing new value added services such as risk management.  Providers of software and services need to adapt to leverage underlying commodities while changing how they serve their customers. 

12:45:24 PM    comment []

Click here to visit the Radio UserLand website. © Copyright 2002 Ross Mayfield.
Last update: 12/20/2002; 5:58:17 PM.
This theme is based on the SoundWaves (blue) Manila theme.
November 2002
Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
Oct   Dec