I've just come off a whirlwind conference tour that started in San Francisco last week with Web 2.0 Expo and ended with the Web 2.0 Kongress yesterday in Frankfurt. I was fortunate enough to be able to speak at both conferences and it was fascinating to see the differences in focus between the two events, as well as some of the apparent trends they had in common.
Given the estimates of the size of the crowd at Web 2.0 Expo, anywhere from 10,000 to 16,000 people depending on who you talk to, there's little doubt it was one of the leading events this year around the next generation of the Web. In contrast, the Web 2.0 Kongress was a smaller and much more business focused affair with a lot of focus on integration and SOA. Yet it was abundantly clear at both, based on my conversations with numerous attendees, that we're now well clear of the early hype of Web 2.0 and much more on how to exploit the opportunities that it maps out for us.
Another key trend I saw was the attendance of mainstream business people who were very much in evidence at both events, something that I've noticed has been increasing at Web 2.0 events lately in general. I met attendees from major corporations, federal and state government, and many others from medium to small size businesses. And a good percentage of them were business people and not from the technical side of things. This doesn't come so much as a surprise if we take into account indicators such as the McKinsey global survey on Web 2.0 which I covered in my last post.
The boundaries of the Web are blurring
The other hot trend, besides of course of just about anything social to do with the Web, that was explored at both events was something that still doesn't yet have a good name yet. And that's the increasing dissolution of clear boundaries between sites, content, and even functionality on the Web. Sometimes referred to as the "atomization" of the Internet, I prefer to call it the syndication of everything -- software and information alike -- via simple Web technologies and ad hoc techniques.
Part of the cooperate, don't control mindset of Web 2.0, the advent of widespread Web parts and services such as widgets and badges and open APIs is an attempt to take advantage of an important concept that has only recently been adequately appreciated, sometimes known as Jakob's Law. This law states that online products and services are best designed when they take advantage of the fact the users spend most of their time on sites other than yours. This realization isn't new and is one important reason that online content syndication through things such as RSS and ATOM has grown so important.
But syndication of content is only the first step in fully exploiting the intrinsic power of the Web as a networked application platform. Access to a site's functionality, either via a Web services/open API as well as widgets and badges that make it easy to others to take portions of a site's capabilities and host them where they wish are the next logical step in this progression. Being everywhere on the Web -- instead of just on your site -- has been a brilliant strategy for sites like Google who realized that to build the largest possible business, you had to scale your products to the size of the Web and their AdWords widget is arguably far and away the most potent example of exploiting Jakob's Law yet created. Not to mention that it has a money making business model built deeply into it.
It's this last part, ensuring that you have a business model baked into your site's open platform strategy that is as important as anything. I recently had a discussion with John Musser, whose ProgrammableWeb.com site is probably the best way to see what's actually happening out there today in terms of mashups and APIs, where he indicated his research strongly suggests that making sure your company's business model is woven deeply into an API is how to best make it commercially successful.
Of course, that's not to say that consumer won't be using Web parts and services (primarily syndication via RSS/ATOM of course) by the millions for their own reasons, and are doing so already. But giving them a good reason to help spread the parts of your site far and wide can be important for them picking your site's widgets and badges over others.
But this is the Web parts story in the large; making a play to be as much of the foundation or the ornament of as many other sites as possible. On the other site of the story is the mashup story, which got some especially good coverage by IBM at the Web 2.0 Expo and about the same at the Web 2.0 Kongress.
I've discussed this story before in numerous posts here on this blog, but I continue to be encouraged by the enormous amount of effort that many are putting into filling the wide divide that currently exists between the simple model of copy-and-paste widgets and badges we have today primarily for consumers and the detailed programmatic wiring currently required of those with programming skills to connect them together into real software.
Even though this latter part is getting easier all the time -- and radical ease-of-consumption seems to be required to get significant widget uptake at all -- the nirvana continues to be to enable a whole new generation of power users to literally create the software that they need on the fly, akin to the way people use spreadsheets and personal database software to create the solution they need now in the desktop software world.
But the hallmark of all of these trends is emergent techniques and methods that have been proven by success on the consumer Web. Google's Adwords widget and YouTube video badge are just two big examples of this working successfully and I've covered in the past how sites like WidgetBox are making it easy to find what you need to host them in your blog and wiki or create whole new sites from the great Web parts elsewhere on the Internet. Furthermore, I'm very encouraged by products such as QEDWiki and Itensil in particular for applying existing Web 2.0 and Enterprise 2.0 platforms, wikis in particular, as an effective and easy-to-use end-user development platform.
Getting a better sense of the picture: The Mashup Ecosystem Summit in May
I was able to talk with IBM's Rod Smith at the Web 2.0 Expo and he told me about The Mashup Ecosystem Summit that IBM is organizing in early May. I've been talking in private with a number of folks about creating a definition of a mashup stack. This seems vital since the mashup application development model seems to be on track to become a major new way of creating software and we don't yet have a clear picture of the whole field yet. This is further reinforced by the sense that mashups are essentially naturally emergent composite applications made from Web-oriented service-oriented architectures (SOAs) that are done in a way that actually works and that people really use on a widespread scale. This summit will hopefully largely serve the purpose of defining the key moving parts in the mashups and I'm hoping to make it there on May 7th. There may be some limited room left for participants, and if so, please contact me and I'll send a note to Rod Smith to see if there's any more available.
Recent surveys report that 21% of executives globally will be investing in mashup development this year. Will you?