What happens downstream?: Scott Mattoon's blog

Sun's annual Customer Engineering Conference wrapped up yesterday.  Officially, the conference themes were Green Computing and "Redshift" - the Internet growth trajectory that is under-served by Moore's law.  There was plenty of great content on these themes, but the overwhelming message I walked away with after three days was that Sun has finally risen above the Not_Invented_Here syndrome that has plagued it's market growth for years.

Todd Fast said it best in a post There is no chicken. There is no egg:

"To most people at Sun, "community" is a walled garden cut off from the real conversation, a case of we know better, shut up and listen. Sun is hardly alone in holding this attitude, but particularly prone to it because of its technology-centric perspective. From that perspective, community is not something that is self-modifying nor self-sustaining, rather it is something to be moderated, controlled, and ultimately held at arm's length. We talk about gathering community in order to drive usage, but we're actually grasping at a shadow of the thing we should want instead—culture."

Courageous and controversial for a circa August 2007 assessment, but accurate to my mind.  Less than two months and a few tipping-point innovations later it appears that Sun has truly transcended (ascended?) this navel gazing mindset. Phew.

The transcendence was no more evident than in Tim Bray's talk, "Web 2.0: The Nitty Gritty".  He wasted no time getting to the point: Sun underestimated the priority of Time to Market for the Web 2.0 developer.  He cited Time to Market as one of four hot developer issues:

• Scalability
• Time to Market
• Maintainability
• Integration

Of the four, Sun placed disproportionate emphasis on Scalability.  But new scaling models were emerging and acceptance of perpetual beta service levels was increasing, both of which served to disrupt the market dynamics that propelled Sun to precarious heights in 1999 and 2000.  Tim went on to espouse all the advantages and adoption patterns of Ruby on Rails and PHP, which play well in a Time to Market driven business model, while Java was akward for a Test First style of programming.   He then went on to say "SOA is bad," sending gasps through the audience.  But his statements were really not that controversial to this crowd, whose collective experience suggests the developer marketplace will not wait for Sun.  People were smart enough to not take Tim too literally on this latter point, but everyone can make sense of Atom and RSS, while not so for SOA - how many developers have the time or will to wade through WS-* specifications in order to integrate a handful of apps?  

The working definition of Choice for Sun engineers of yore (and I count myself among them,) was, "noun, Any color you want as long as it's Indigo".  Tim bid this notion goodbye brilliantly.

We're getting on with the business of competing on merit and mindshare.  Evidence of transcendence is popping all over the place;  heard on an internal conference call today from a software product lead, "We should not constrain ourselves for religious reasons." 

You want to do Flash, Flex, AJAX, JavaFX.  No problem.  You write code in PHP, Ruby, Java?  No Problem.  You deploy on Linux, Windows, Solaris?  No problem.  Choice is the new mantra at Sun.  There's no more room for NIH fussing anymore.

As Jonathan Schwartz said in his keynote on Tuesday,  "Revenue is a lagging indicator of how well you're driving adoption." Given recent growth in Sun's revenue, especially the double digits in software revenue, I'd say we're finally on track with this adoption thing.  No factor has been more crucial to the upward trend than the embrace of technologies that did not originate at Sun.

Start Up style enthusiasm was in no short supply at the second CommunityNext conference held at the Plug And Play Tech Center in Sunnyvale last Saturday.  Several companies represented were not even online as of the first CommunityNext conference held last February, yet many have since built thriving communities with millions of users.  The secret to their success, and the theme of this gathering, was viral marketing.

While February's day long event featured lessons on How to Tap The Wisdom of Crowds, Saturday's teachings might best be described as Getting Inside The Teenage Brain.  The scope of possibilities seemed to have devolved in the intervening six months to a level more concerned with how 14 year old girls will place a widget on their MySpace page than how the network effect can improve the lives of millions.  I left the event feeling like the social networking party had moved to the trailer park and the Anchor Steam on Draught had been supplanted by Pabst in cans.   But I don't spend much time on MySpace, so take my sentiments with a grain of salt (and a lime).

Amid the inanity of Profile Bling and Breakup Alert best practices there were sensible exchanges about building net communities the viral way.  Some insights from the viral front lines :

• Metrics matter - measuring effectiveness against goals are as important to running a widget based viral marketing campaign as any Madison Avenue ad campaign.  
  
• It turns out that wikis are for everyone - given a wiki as easy as making a peanut butter sandwich, teachers will use it to develop curriculum and engage their classes, as PBwiki discovered.
  
• The Facebook platform was deemed by many widget developers here as the API sine qua non.
  

Perhaps most relevant to a field architect like me was the implicit adoption pattern threaded throughout the presentations: widgets get combined with other widgets to make new and interesting platforms that are essentially loosely coupled composites of fine grained apps.  This seemingly chaotic trend toward Widget-dom foreshadows an adoption pattern that runs orthogonal to the SOA model so many enterprises are pursuing, where heavy duty governance is critical and service discovery with its attendant infrastructure represent costly overhead.  Corporations are spending millions planning multi-year SOA initiatives.  Meanwhile the Facebook platform allows developers to build composite applications quickly, all the governance is essentially embedded in the client libraries, and services are discovered virally - registries and WSDL are not critical to the ecology of a Widget World.  Granted, student social calendars and virtual food fights are a far cry from a CRM/ERP/BI mashup, and identity management and access control through a RockYou widget would make any CIO cringe, but with the addition of JSON support to platforms like Facebook I expect we'll see more complex integrations emerge soon, and WS-XACML shows promise for protecting data exchange between loosely coupled apps according to some rich policy.  How quickly will this hosted RESTful approach displace enterprise owned and operated SOA infrastructure is hard to predict.  No doubt the transformation is driven by many of the same factors driving the redshift market transformation Sun is betting on.

The explosive growth experienced by many of the companies at CommunityNext reaffirms Sun's focus on designing for network services at scale.  The results of that focus, such as Project Blackbox, the Niagra processor, and the Sun Grid, ought to figure heavily in the future of many of these start ups.  Sun's SOA technology, which looks more and more RESTful by the day, could be the best bridge for Enterprise IT to cross into Widget-dom, and a good platform for social networking platforms to adopt in order to penetrate the Enterprise market.

My cryptic and incomplete notes from this Sun sponsored event are below.

---

The People Formerly Known As The Audience landed a piece on Wired.com today about the Open Architecture Network (OAN).  

The story was part of a pro-am journalistic experiment orchestrated through Assignment Zero that produced 80 distinct articles dealing with crowdsourcing, of which 12 earned a spot on Wired's home page.  As one of the OAN story contributors, I bagged my first byline on Wired.com, one in which I was decidedly on the am side of pro-am.  While my contribution was relatively small, it was enough to give me a taste of the post-edit blues - most of my copy was either red-lined, or reduced to the point of inaccuracy.   My original reporting expanded on the technology choices and process of developing a collaboration site using the open source CMS Drupal.   The final piece gives a nod to Drupal and Sun Microsystems, but leaves the wrong impression:

"Even the software powering the site -- designed by Sun Microsystems -- is open source: the Drupal content management system chosen by thousands of nonprofits for its ease of use."

Sun did not design Drupal, and while ease of use is one of the virtues of Drupal it's an oversimplified view of why so many nonprofits use it.  Granted, the focus of the story was not so much on technology as the potential for open source design, but one of the points red-lined from my copy was perhaps one of the most relevant given the context of this crowdsourcing experiment:

"Organizations that use Drupal for their online communities include... Assignment Zero."

Drupal is everywhere.   Drupal's integral role in the explosion of collaboratives born of open sharing and grass roots participation is worthy of an entire Wired issue, if not more than just a mention buried deep in one story.  Even among citizen journalism sites, it's a dominant software platform - witness The Witness Project, for example.

Still, good to see the momentum for Architecture for Humanity continuing in the media - in addition to the Wired piece, another story about the OAN appeared this week in Business Week, and AFH's founder, Cameron Sinclair, was added to the distinguished list of "Thinkers of Tech" for Fortune's iMeme conference next week in San Francisco.

We're still the audience, we just have something to say now. 

 

On Saturday February 10 I attended a great event on the Web 2.0 phenomena known as social networking (or should I say the social networking phenom known as Web 2.0?): CommunityNext at Stanford. Great speakers, good schwag, and a large dose of optimism about the potential for networks of people to affect enormous positive change in the world. See my notes from the event posted over on downstream.org.

Two orgs represented at the event I found especially compelling:
Kiva.org and Cambrian House.

Kiva.org has enabled the infusion of capital into the 80% of the world's population that is not banked through an innovative microfinance model that has gained much praise and attention from the media, philanthropists and financiers. My friend Tracey Pettengill, Founder of MicroPlace has partnered with Kiva's founder, Premal Shah, to bolster microfinance in Silicon Valley hosting a periodic event with the Silicon Valley Microfinance Network. Their next event features guest speaker Bob Annibale, Citigroup's Global Director of Microfinance. Should be an interesting discussion.

Cambrian House has an innovative model for incubating and propelling new business. Leveraging Crowdsourcing, the model facilitates those with a great idea for a product or service to connect with funders, technical expertise, legal support, guidance and moral support to get the idea off the ground. I'll let you know how my idea for the next big thing flourishes on Cambrian House.

If ever there was a group of entrepreneurs who are invested in the neutrality of the net's telecom infrastructure, it's the participants at this event. Perhaps better than any other business demographic, they demonstrate just how precious net neutrality is. No word on the next CommunityNext event, but I'll track it and post here.