There were a few good ones I did get to, though.

First was titled “User Experience: The Next Wave” and was Ray Valdes’ take on the core value of user experience and “usability-centered design.” He had some great general principles for how organizations can take advantage of scientific, measurable approaches to usability to get beyond the “I like the blue one” design process to many still follow.

He also pointed to some of the key fallacies about “usability-centered design” (I still prefer user-centered design as a term):

1. Usability testing (“validation”) has to be expensive
2. User-centered design has to explode the project schedule
3. Users like the system we designed the old way, therefore we don’t need to change
4. Having a customer-focused attitude replaces doing formal design

He closed by talking about some of the new technologies and approaches (social software, new interfaces and input modes), and how really the primary challenge (and answers) remain mostly unchanged: solid strategy, best-practices in design, and a constant feedback loop with actual testing.

The second talk I saw was titled “Strengthen Your Governance Strategies for the Wave of Web 2.0 Technologies” and was by presented by L. Frank Kenny.

He talked about all the new kinds of endpoints into the enterprise which characterize web 2.0 – mashups, rogue service endpoints created to connect to outside services, users consuming outdated versions of corporate web services, etc.

Ultimately he argued that the current generation of mashups and syndication feeds probably don’t necessitate new controlling technologies for most enterprises – they can be governed by existing CMS systems, firewalls, filters, and the like.

He suggested that organizations should consider taking advantage of some of the new services which monitor social networks, the blogosphere, wikis, and forums – services like brandimensions, cyveillance, and webclipping.com – what he called “Brand Protection” as an emerging market.

—
Last session of the day was the Yochai Benkler keynote about which I wrote earlier.

I missed, unfortunately, the “It’s the Web, Stupid” presentation by David Mitchell Smith and Gene Phifer – based on the presentation slides (which attendees get access to but I can’t share) it looks like I would have enjoyed it. Here’s how they describe the presentation in the agenda:

Although the Web 2.0 name is popular and represents the Web of today, the world seems hungry for 3.0, whatever that is. While Web 2.0 suffered from being perhaps overly broad, the special interests driving 3.0-mania have the opposite problem – they are often too focused. We’ll look at the future of the Web including the semantic Web, the mobile Web, the virtual world Web and other candidates for “3.0.” Regardless of what the next big buzzword is, the Web will remain one of the major catalysts in technology and one of the major sources of innovation.

Anyone reading this who did see that session care to comment on it?

When you’re at a conference (I’ve been at both Ajax World West and Garnter Open Source / Web Innovation Summits in the last week) or in an airport, electrical outlets are at a premium. There are countless web 2.0 knowledge workers wandering the halls seeking power. (Ampires, or wherevolts).

This little device turns that moment of potential conflict – where you spot an outlet but all the available sockets are in use – into a moment of collaboration. (In case it isn’t possible to tell from my hotel room photograph, this translates a single three-prong outlet into three. Simply approach the user of one of the existing outlets and ask to unplug them for an instant – they get to stay plugged in, you get to plug in, and you get one bonus plug for a third person or a second device.)

It’s “just good enough” – carrying a real powerstrip with fault protection, etc. would be better, from the point of view of protecting your laptop – but hey, you were plugged directly into the socket already, so this doesn’t make things worse.

It’s small enough to put in your computer bag and travel without problems.

It’s cheap enough that if you leave it somewhere by accident you can just go buy another one.

It’s even in RSS orange.

(Full disclosure: Optaros was a sponsor of the Web Innovation Summit).

Unfortunately I got in too late on Tuesday night to see any of the Tuesday evening sessions. I would have enjoyed Anthony Bradley‘s Web 2.0 Basics Tutorial, based on reviewing the slides and seeing Bradley’s other presentations. I like the way he approaches questions about adoption and Enterprise class Web 2.0 applications.

Wednesday am, running a few minutes late due to a conference call with Optaros colleagues on the East Coast, I wandered into the opening remarks just in time to hear the speaker (was it Adam Tinkoff?) ask “is jeckman in the room?” – he’d been following me on twitter as I tweeted away about my travel saga. (Planes never arrive on time anymore – it’s really just a question of how late they will be or if you’ll get there at all). Best publicity I’ve had from twitter so far, though I’m not sure my “complaining about travel” tweets are the ones I most want to be known for.

Then, I watched the initial keynote session: “Planning for Five Major Mutually Reinforcing Disruptive Discontinuities,” featuring Tom Austin, Gene Phifer, and David Mitchell Smith. Three analysts, five discontinuities – it was a whirlwind trip.

High level, the five discontinues are:

1. Software as a Service (SaaS)
2. Consumerization
3. Web 2.0
4. [Free and] Open Source Software
5. Global Class Architectures

Although I wish Gartner analysts in general would stop talking about the “hidden costs” and “hidden risks” of open source, since I don’t really think they’re really hidden in either case, most of what they had to say about the five discontinuities made perfect sense to me. If anything, my only critique was that they didn’t seem to me to be telling the audience anything they didn’t already know – but I guess it is difficult to gage the audience’s level of familiarity with these concepts, and the keynote did ground most of the discussions for the following three days in some shared basic concepts.

—

Second session I saw was on Enterprise 2.0, with Anthony Bradley and Tom Austin. (Also covered here in eWeek).

Although they had some logistics issues (the version of the slide deck they had loaded wasn’t, apparently, the one they expected to see) they ran through most of what I’d expect to hear about as Enterprise 2.0. Enterprise 2.0, like all the major overlapping discontinuities, and like open source, was described as unavoidable – the message to enterprise IT organizations being they need to get involved and move beyond skunk works type projects into real projects.

My favorite section was on the myths and urban legends of enterprise 2.0 (paraphrased):

• People will naturally share things on the web
• There is exactly one right way to organize any set of data
• If you have a good culture, other controls aren’t needed
• Social software is for kids (like Kix)
• Enterprise 2.0 is just vendors trying to sell Knowledge Management in a new wrapper

They also offered a nice list of 8 ways to ensure success and 8 practices to avoid. I won’t reprint them all here (not sure what Gartner’s copyright policies are) but they included:

Success: Start small but real. Be open, let emergent structures emerge, and lead by example.

Mistakes to avoid: Don’t ignore accountability; don’t think Web 2.0 is a “fad”; Don’t have a plan for growth.

–

Having been intrigued by a couple of things Bradley said, I then went to the “Plant Seeds: A Model for Community Adoption in the Enterprise” which he also presented.

I really liked the framework he presented, though I’m not terribly fond of long acronyms (all of PLANT SEEDS is an acronym). Focusing on starting not with “experiments” or “proofs of concept” in the Enterprise 2.0 but real solutions to real problems, with scope controlled so as to minimize risk. Larger successes build from small successes, not from experiments.

He described the “legal-precedent” type approach to adding governance to these efforts – you set out high level rules, then as/when examples of borderline behavior (or outright bad behavior) come up, you use reactions to those behaviors to guide future behavior. (Rather than trying in the abstract to determine all the ways people might behave wrongly and explicitly forbid those).

He also described the nature, nuture, or both notion – that some people will naturally want to share, while others will need too see sharing be cultivated and rewarded before they take to it.

Finally, he described the way in which your Enterprise 2.0 efforts need to be integrated, not new silos separate from individuals’ “real jobs” but part of the larger IT ecosystem. The net effect can’t be additional work added on top of a full stack – it needs to replace and ultimately make more convenient the work people are doing – as it will, if the problem is a real candidate for these approaches. (If it doesn’t fit, you may be forcing it as the solution to the wrong problem).

—

The rest of day 1 I spent at Open Source Summit sessions I’ve already blogged about.

Wasserman works with Carnegie Mellon West, and is the Executive Director of the Center for Open Source Investigation.

He covered a lot of the basics of what organizations need to keep in mind as they evaluate open source projects, and some resources (the Business Readiness Rating, for example) they can use to support those adoption plans.

His basic principles for evaluating software:

• Does the software do what I need it to do?
• Are there good sources of documentation and support?
• Is the software being maintained and updated?
• What do others think about the quality and performance of the software?

Good advice for open and closed source alike. People often get caught up in the details and intricacies of licensing options and miss the basics. Not that you don’t need to think about licensing, but you can’t let a focus on the fact that you’re looking at open source software distract you from the core questions you already know how to evaluate.

—

Another panel I saw was “Commercial Open Source: Beginning of the End or End of the Beginning?” by Brian Prentice.

He put the recent controversies this summer of SugarCRM’s attribution license and the CPAL in the context of a longer term divide between competing interests within the open source world – pointing to VC’s funding commercial open source companies, who hope to control the costs of sales and marketing by using open source as a distribution model but feeling the need to hold back some intellectual property to create a sellable asset.

He described the challenges inherent in the “functionally delineated” model, where there is a community edition which is free and an enterprise edition which is not. Users and organizations adopting this style of commercial open source must be careful to recognize the details of what is and is not included in the solution they’ve adopted. (Just as in a functionally delineated closed source model with different versions of a product each version must be clearly differentiated).

Alfresco, on the other hand, was signaled out as a counter-example, or at least another way of doing commercial open source, since the community and enterprise editions are functionally identical, with the difference being support and services. (Disclosure: Optaros is an Alfresco Platinum Partner).

I suppose you could say that what we’re seeing is a period of experimentation as companies which would otherwise have been traditional proprietary companies trying to learn from and benefit from the open source ecosystem. It’s neither the end of the beginning nor the beginning of the end, just another chapter in the ongoing saga.

Many of the points Tiemann made about the efficacy of open source as a development methofology as compared to closed source were reported here in eWeek: “Is Open Source the Best Way to Unlock the Value of IT?” (which ironically enough had a big Microsoft VisualStudio ad in the middle of it when I read it).

Tiemann covered the approval this summer of the Common Public Attribution License (CPAL), noting that the safe harbor provision – enabling a variety of mechanisms for handling the attribution – was the fundamental change that made him support approval of the license, and giving props to Ross Mayfield and SocialText for choosing to submit themselves to that process – very few CEOs would request validation from an external body over which they have no control and inside of which are many competing ideologies and interests.

He also discussed the GPLv3, talking about it as a small incremental upgrade, and pointing out that in the closed source software world, every minor point release may come with new licensing terms, where the GPL has only had three versions in nearly 20 years.

(See also “OSI Calls for Major Revisions to Microsoft Permissive License” for other discussions with Tiemann from the conference).

–

Again due to the overlap between the Open Source and Web Innovation Summits I missed the opportunity to see Ray Valdes talk about “Web 2.0: The Open Source Connection.”

I’ve often made the argument (to anyone who will listen) that the real engine behind Web 2.0 innovation is open source – not that every important web 2.0 property is itself open source or even uses open source, but that the explosion of new approaches, new techniques, and new properties would not have been possible without the mature open source stack, which let startups create real functional applications without such a high barrier of entry, and without the “success tax” of increased licensing based on user adoption.

The opening session was Wednesday afternoon with Mark Driver : Gartner’s Open Source Scenario for 2007: Risks and Rewards for Mainstream IT.

This was the session which led to this Network World article and corresponding Slashdot flame-fest. But both missed what I thought was a perfectly rational set of statements:

1. that commercial software vendors cannot ignore open source as a disruptive innovation
2. that commercial software vendors are increasingly incorporating open source in a non-trivial fashion, and
3. that this trend will continue to deepen over the next four years.

Driver walked through some basic definitions and argued that we’re in a Third-Wave of Open Source reactions from Corporate IT: whereas many people in Enterprise IT departments first reacted to open source with some mixture of indifference and irrational emotion (that is both pro and con), the current phase is one characterized by “realism” – which will lead ultimately to “leverage.”

I suppose one could argue it is because of the company I keep, but I’d argue a large number of commercial enterprises passed realism and have been enjoying leverage for some time – but otherwise I think the model is accurate enough in describing the process many organizations go through in learning about open source.

One interesting point Driver made was that open source tends to create “Investment Protection” where proprietary software creates / ensures Intellectual Property protection. In the open source world, the investment the user or adopting organization makes gets preserved, because there is real vendor independence. In the commercial world there is real protection for the investments of the producing organization.

In addition Driver showed Gartner research which demonstrated that many organizations are using open source in “mission-critical” applications – that the percentage of open source software used in a mission critical application was almost the same as the percentage of internally developed or commercially purchased (non-OSS) software used in mission critical applications.

Driver argued that the adoption prioties are changing as open source moves further into the adoption curve and becomes more maintstream or is adopted b more conservative adopters. Where earlier adopters (“technology aggressive adopters”) focused on open source because it provided flexibility and independence, later adopters will be more focused on cost savings and risk mitigation. (All four motivators are important to both audiences – in Driver’s argument it is just their relative priority which changes).

Driver talked about the possibility of an increasing bifurcation within the open source community between “community class open source” projects versus “business class open source” – differentiated not some much by their features or specific license but by the goals, aims, and cultures of project governance. For conservative adopters whose focus is cost and risk avoidance, community class open source may not be a viable option, whereas for technology aggressive adopters the business class open source may be too slow moving or non-innovative. Additionally, he described the emergence of “gated source” options, which lie somewhere between the open source and proprietary models,

Driver listed four factors enterprises should consider in planning open source adoption:

• Fitness of Purpose (does the software do what you need it to do, well)
• Maturity (is the software project well governed, and capable of reliably producing quality?)
• Your technology adoption profile (is your organization an early, mainstream, or late adopter of new innovations?)
• Deployment scenario (how will the app be used, in the context of the organization’s mission? Is it mission critical?)

He closed by noting that “ignoring open source is not a viable option” and that the days of “skunk works” adoption are over. Enterprises should be planning adoption strategies, just as they have corporate management strategies around procuring commercial / proprietary / closed source software.

—

Unfortunately, given the overlap of the Web Innovation and Open Source summits I didn’t get to attend Nikos Drakos‘ session on “Open Source in the Workplace: What it Promises and What it Delivers,” but based on the ppt from the session I think I would have enjoyed it . He covered the growth of open source outside the “infrastructure and development tools” categories – into areas like content management, collaboration, and customer-facing communications. He also went into the leverage of open source collaboration principles in other contexts – perfect lead up to Yochai Benkler’s Keynote on Thursday.

Luckily, despite some overlapping sessions, the keynote by Yochai Benkler was shared across summits and I was able to attend.

If you’re not familiar with Prof. Benkler, you should be. His book The Wealth of Networks: How Social Production Transforms Markets and Freedom is the treatise on /study of commons-based peer production. (It’s available in many formats including free versions under a Creative Commons Noncommercial Attribution Share-Alike License).

He’s also the author of “Coase’s Penguin, or Linux and the Nature of the Firm,” in which he argues that:

while free software is highly visible, it is in fact only one example of a much broader social-economic phenomenon. I suggest that we are seeing is the broad and deep emergence of a new, third mode of production in the digitally networked environment. I call this mode “commons-based peer-production,” to distinguish it from the property- and contract-based models of firms and markets. Its central characteristic is that groups of individuals successfully collaborate on large-scale projects following a diverse cluster of motivational drives and social signals, rather than either market prices or managerial commands.

What follows are my rough outline notes of his talk. Benkler’s the kind of speaker where the notes or even the slides don’t do justice to seeing him speak – but at least I’ve got some of the highlights and examples down.

Benkler:

We now live in a world in which:

• The most important inputs into the world’s core economic activities are widely distributed (the ability for globally distributed populations to create information and culture)
• Behaviors once on the periphery of economic concern are moving to the core (social relationships, friendships, concerns about decency and fairness)

Example: The Encyclopedia – used to be thousands of dollars to get a 24 volume set of bound encyclopedias. That pressure drove the price of the Brittanica down to $500 in 1989. That was then followed by Encarta for $59.95 in 2000. Finally, wikipedia which is free.

Benkler mentioned the Nature study on the quality of Wikipedia entries, and Britannica’s response (PDF) to it. (Nature‘s since responded to the Britannica objections).

The reality is that most hands on practicing scientists felt both were equally lousy. (Never ask a deep expert to evaluate a paragraph level summary of a complex topic – they always find it lacking). But that this was even a serious question to be tacked – that Wikipedia could be said by a reasonable person as potentially comparable in quality to Brianicca – is Benkler’s point.

“Information Production” is now the critical economic activity – at the same time that our ways of producing information are shifting to commons based production.

Benkler outlined a number of concepts (and drew distinctions between them) related to Commons Based Production:

1. Peer Production
2. Shared Resource Utilization (things like SETI @Home
3. Free/Open Source Software

Examples included (I added links):

• The search for Steve Fossett
• Craters outlined by volunteers for NASA
• The BBC Help Us Make News
• Threadless
• Learning to Love You More
• Kaitura
• Porkbusters and the Secret Holder
• The Sunlight Foundation Earmark Map
• Black Box Voting and the campaign to decertify certain electronic voting machines
• Media Volunteer (as I’m writing this their site seems to be down – asking for authentication for public pages)
• Cambia BioForge

This is really a new kind of production in that it is not market driven and it is not centralized. We’ve had market-driven, decentralized production (standard firms in the US), we’ve had market-driven, centralized production (large corporations), we’ve had non-market, centralized production (governments and NGOs, non-profits). What we have not had is non-market, decentralized production. (This echoes Clary Shirky’s assertions about Perl being an act of love).

Benkler showed a typology of different ways peer production works, in terms of the types of inputs people are asked to make and the types of organizational strategies they use, as well as the kinds of motivations (extrinsic and intrinsic) driving them. The more creativity and knowledge necessary in the types of contributions people are asked to make, the more you have to move to a many to many type collective form of organization. The major examples here are things like Google and Digg, where the effort required by the user is low (making links on the web means helping Google’s algorithm but you don’t think of it that way, digging something is a single click activity); on the other hand Free/Open Source Software requires much more complex structures. (Not sure if he’s overestimating the “volunteer” nature of open source here given the number of developers on may open source projects who are employed and do this contribution as part of their job).

The key question isn’t whether peer production is a fad – it clearly is here to stay – but how it operates and how we can design to encourage the right kinds of collaboration.

Too much of the theories of cooperation has classically depended on “rational self-interest” but newer explorations in a number of fields (sociology, economics, psychology, evolutionary biology) has started to move beyond that.

Benkler’s argument is that people respond in ways which are not always or first self-interested: people resond in ways which are predictably cooperative under certain conditions:

• Communication
• Humanization
• Trust Construction
• Explicit Norm Creation
• Monitoring / Peer Review / Discipline
• Transparency in Governance
• Fairness (in context – concepts of fairness vary widely)
• Self-Selection (as opposed to assignment to tasks)
• Group Identiity and Investment
• Leadership (older sibling style, not parent)

Benkler made a great point about being wary of introducing extrinstic motivators (ie, money) in systems which have been driven by intrinsic motivators. For example, systems which try to introduce shared ad revenue in the user-contributed-video context may alienate existing users who were motivated by other factors. You try to match love with money and some folks end up not wanting the money and no longer wanting to work for love.

Benkler closed with some of the political impacts of social production – ways in which social production is changing the political reality of people all over the world and ways in which industries, governments, and corporations threatened by social production have tried to push back – the DCMA, Trusted Systems, etc. (Unfortunately by this point he was trying to wrap up very quickly and I didn’t get a good list from his last few slides).

Because Benkler’s operating at a high level of abstraction – thinking about the impacts of peer production at a global and historical scale – it can be hard sometimes to connect his concepts to what companies are trying to do in the “Web 2.0″ space – but his elaborations should help us understand the real fundamental shifts underlying what otherwise might look like a “fad.”