I visited BarCampNYC2 over the weekend (check out the Technorati Tags for others discussing the subject). I wasn’t a real active participant, and I was only there for a few hours, but it was a remarkable experience. Lots of collaboration, cooperation, discussion, debate, and other fun academic exercises with a geeky twist.

One session I attended that I want to highlight in this post is the “Future of Web Apps” session. It was a bunch of techies, mainly programmers, discussing where they thought web apps would be mainly in the next few years and even as far as the next twenty years. It was more conclusive evidence to me that people really have no idea what direction this whole people-centric web is going.

One of the Iridesco founders (I forget if it was Danny or Shawn…) mentioned that he currently sees the pendulum of scope swinging towards splintered, decentralized web apps, which I think everyone would agree with. However, he then said he expects the pendulum to swing back soon, such that the all-in-one web app solutions like Google or Yahoo will be the services with the best experience and most significant adoption.

I disagree with his forecast; all-in-one services can’t offer a sensible, intuitive UI mainly because the scope of their goal doesn’t allow for simplicity. It’s not a problem that will be solved in time; it’s an inherent contradiction in the mission. Therefore, smaller niche services will be the source of the easiest UIs, with the smallest barriers to adoption.

Furthermore, as an all-in-one provider grows large, they become scared of changing their interface significantly because any change (good or bad) is viewed as something different, which is bad for an existing user base with deeply entrenched expectations and UI muscle memory. This inability to innovate opens the door for smaller companies to be disruptive and creative in ways that big all-in-one giants never have the opportunity to do. Thus, the fracturing of web services into smaller niche markets will continue.

However, our disagreement is rather subjective, and neither of us has evidence to back up our respective points.

Similarly, the question of identity came up. A Microsoft rep at BarCamp mentioned that Microsoft Card Spaces (soon to be released) was going to solve the problem of identity on the web… it sounds a lot like Passport 2.0, but more open, and it’s going to be built into the .NET framework in Vista for ease-of-implementation of developers.

Here, I disagreed because I don’t see identity as portable from site-to-site on the web. Andrew Parker at the Facebook is very different from Andrew Parker at LinkedIn, and, while I don’t mind if people know that those profiles are the same person, MANY of my peers do mind, a lot. They’re all the same people that are fighting against opening the Facebook network and the release of the news feed. Having a virtual identity card that one can carry from site to site might solve the problem of having to sign up for an account when you want to try out a new service, but it opens up so many new problems by tying together people’s identities across sites that I don’t think it is a reasonable solution.

But, again, neither the Microsoft Rep nor I could point to any specific evidence to prove we were right. Sure, we could talk about the past failures or successes of other products, but many great ideas on the net fail multiple times before they get adopted, so net history is weak rhetoric. We are relying on our subjective opinions and experiences.

That’s why I am so exciting about my job and this time on the web. No one knows what’s going on or where we are going, and it’s anyone’s opportunity to shape our path.

To come full-circle on the subject of BarCampNYC2, congrats to Amit for a terrific event. You’re a rockstar.