Serious barriers to going open

While the title of this post sounds gloomy and might make it appear that I don’t favour opening up the web as much as humanly possible, in truth it’s the exact opposite. However, I like to think I’m also a realist and as I read and think about the move toward more open and accessible content and services on the web, there are significant issues in our way en route to achieving that goal.

It’s not just about cost

If, at its heart, the evolution of the web is dependent on social computing, web platforms, mashable services and bottom-up, user-generated content, then we’ve got a ways to go yet. For example, RRW laments web platform performance, and rightly points out that services without a solid revenue play built in will not succeed:

Our culture of sensation and free makes it much harder for platforms to think deeply and be disciplined. Google felt they had to come out with something to stop Facebook’s momentum. Facebook rushed to create a completely open infrastructure; and it backfired both for users and developers. Having been burnt by Facebook, small and large companies alike will now think twice before investing in a presence on platforms. This is a shame, for we need platforms and we need them to work well.

That’s only part of the story and misses a very Canadian perspective on it. While RRW also links back to a good article on the risks of using free services, we’ve got a little problem with trust called the Patriot Act, which goes way beyond the concerns of a service being free. It’s all about trust – service longevity, data security, ownership and more – I’ve worked extensively in the public sector, and it’s probably their number one issue.

Connecting through open standards

What about connecting people through applications and services, and then in turn, connecting those services to each other? Single sign-on for every web application you use would seem to be a start. OpenID should be a no brainer. Sign up for one account, then any time you want an account on another web application, you should be able to login with something like your OpenID and have access. All your credentials are stored in your OpenID and you simply provide that ID URL on any OpenID-enabled app, confirming your password when you log in. I can probably count on one hand the number of places I can actually use my OpenID. It simply hasn’t caught on yet.

The recently announced Windows implementation of OpenID is much like other web giants, Yahoo! and Google, in that it’s a one-way move aimed at still protecting their walled garden. If you have an ID on one of these sites, then you can use it elsewhere, not the other way around:

Although Open ID is becoming a viable option for independent sites, particularly after this announcement, there’s still a great deal of friction preventing serious uptake. Major Internet hubs that offer Open ID are generally providers only, so a Windows Live user is still going to need a Yahoo account to use any Yahoo service. As long as this is the case, the utility of Open ID will be constrained.

Leading web services are typically providing an open API to allow programmers to extend those services in new ways. It’s another of the key underpinnings of the social web, but means there are as many ways to mash together new applications, as there are applications themselves. Google’s Opensocial promised to change all that by delivering one API for all social applications to use so that they could easily talk to each other. Once you’d built one service on Opensocial, then all other services would work the same way. Much like user IDs, many services see a great deal of value, financial and otherwise, in setting the rules in how you play with them.

Education goes open

Given the purpose of open access, education is probably the most natural fit of all. There are plenty of calls from within the educational sector to open things up for learners and instructors alike. I’m not sure if the move to learner-directed credentials could render traditional credentials worthless, but with the growth of Open and Distance Learning and things like the Open Courseware movement, open and social concepts in learning become more important every day.

For example, MIT is embracing the OCW movement, along with several other institutions worldwide (including Canada’s own Capilano U and Athabasca U), while UCLA is bringing their videos to the web.

There has been a significant push in recent years to look at new licensing models to make textbooks more affordable. This is particularly an issue when a couple chapters change marginally and then university courses prescribe the updated releases each time they come out, complete with a new $100+ pricetag. News that Google has settled with book publishers over their online book indexing should be a model for all publishing, and not just older or out-of-print books. It wouldn’t be hard to embrace the realities of online distribution and develop reflective licensing approaches, but is not likely to happen when it threatens the bread-and-butter of the scam that is the textbook industry:

The agreement doesn’t cover books that are currently in print and therefore making money for publishers. Those books can still be marketed through the book search via Google’s Partners program. Books that have had their copyrights expire are likewise unaffected. What it does cover is what Google’s Chief Legal Counsel, David Drummond, calls “the vast majority of books in existence”: those that are in copyright, but out-of-print.

It’s not just textbooks, but broad access to all content and the flexibility of applications. Boy is Blackboard
going about it all wrong
. They know systems like Moodle are giving them a run for their money in the LMS space, so they let their users connect to external and open source tools from one interface. Huh?

Blackboard is still controlling the environment and implementing the features they want. It’s the LMS that needs to be open and flexible in the first place. By providing a secure, flexible tool with easy API extensibility, I’d imagine a better route to choose would be to build upon Moodle to act as the connection framework, allowing users to integrate their tools and academic databases of choice from within the framework. This kind of thing reeks of desperation ploy by Blackboard.

Now, imagine building that LMS on top of solid, ubiquitous authentication and API standards. You want Skype or YouTube or WikiSpaces with that? Just enter your account and your LMS does the rest. Or, how about building a script within Moodle along the lines of Fantastico and implement an instance of your favourite OS tool right with Moodle, with adoption of your CSS styling intact. We’re talking about a lot of coding here, but it’s not hard to imagine a platform that really rocks and serves learners and instructors well.

A better way to go open

Right now, OpenID is all over the place with proprietary services allowing your username to be an OpenID. That’s a start, but think about issues like domain registration, Web Standards or look at more granular technical protocols like HTTP and FTP.

Opensocial is a great concept and Google isn’t going anywhere in the near future, but I’m not sure we want a completely commercial (and obscenely dominant) entity holding the keys and protocol for how a large part of the web might work together.

Eventually ideas catch on, they become ingrained, with broad-based collective management and protection. Standards are the rules by which all developers and providers must play to ensure a better experience for everyone. While we need to stop the walled garden approach to hoarding content and extorting huge license fees for it, what we do need is centrally managed approaches to application integration and user authentication. Things are moving in the right direction, but sort of kicking and screaming. If Microsoft and Blackboard won’t do it the right way, they should just get out of the way.