A number of projects in my department are web-based: iKneer, globalhub, cleerhub. If you go there and can't find out what they are, what they do, how to use them, or who you'd share them with and why, you're probably not alone -- having talked with some of the researchers behind them, they are awesome groups that really want to transform engineering education, and the resources they're creating have tons of potential, but it's hard to see it from the outside.

But that just means there are a lot of opportunities for opening things up -- which is fantastic. I'd much rather help something awesome that should get out to a bigger audience go reach that audience than a bunch of people who want to reach a lot of people "just because," but have nothing to give them. Since the term "opening up" is still a bit vague, here are a couple common recommendations I've made:

Whenever possible, don't require people to register. They shouldn't need to just to see the data that you have. Allow users without logins to access as much of the functionality of your site as possible. If I can't find out whether the site is worth my time, I will decide it isn't worth my time, and I will walk away. Simply requiring registration makes traffic plummet.

Only require registration if it is clear that you functionally need my information to
provide me with a service I want. For instance, it is arguably difficult to send me email
updates on new features on the site without my email address. Mozilla add-ons is a good example of a site that does it right. Let me quote their registration policy:

Registration on [this site] is not required if you simply want to download and install public add-ons.

You only need to register if:

  • You want to submit reviews for add-ons
  • You want to keep track of your favorite add-on collections or create one yourself
  • You are an add-on developer and want to upload your add-on for hosting on AMO

This makes sense. By the time I'm thinking about (for instance) submitting a review, I've probably already used an add-on for which I want to write a review; I'm already invested in your project and you're going to help me deepen an investment that I want to make. Good.

Try to let people reuse what they already have. Login with (insert your favorite popular web service). Instead of making people make yet another profile page, just let them link to one they have (their faculty page, their personal blog, their Twitter/identi.ca account, whatever).

Make it really, really easy for people to take your work and data and displays out to the places they already have. This almost requires allowing public data viewing, because otherwise people can't even link to your work. If you want to go one step further, allow embedding -- think about how you can include a YouTube video on your website... what's your website's equivalent for other people? In effect, you're letting your users do your marketing for you on their sites, which likely reach an audience you can't (or maybe didn't even think of).

Provide tutorials. Now that you have an unexpected audience you may not be able to see, you won't always be there to hold their hand and answer questions when they arrive... so make sure that have some way to get those anwers.

Now for some common responses.

"But we need to track impact statistics! How can we get funding unless we know who we're affecting?"

Open it up. Does Google require registration to perform a simple web search? No. Is anyone going to say Google's search page is not having an impact on the internet? I doubt it. Be creative - there are many other ways to measure impact (hits, pageranks, clicks, links) that don't rely on registration. Web startups do it all the time.

If you don't do something unless you can measure it, you're limiting your impact to what you can measure, which is always going to be a tiny, tiny subset of the impact you could have. Instead, try to have the biggest impact possible, then measure what you can, and argue that there's plenty more that's not measurable ("so if we can measure this much, think of how much more is out there!")

"That works for the internet, Mel. Academia is different."

Right, and your academic project is a website. The laws of human behavior don't suddenly go wild just because something comes out of a university. (Facebook and Google and Yahoo came out of universities.) If you want to see evidence that this applies to metrics that researchers care about, there are plenty of studies showing that open access increases impact. The more accessible you make your scholarly work and the easier you make it for people to use your research... the more they'll use it. (This shouldn't be surprising!) And yes, this has been verified as likely to be a causal effect.

Anyhow. I am available to help with these sorts of things, if projects are interested -- applying open source principles to academia and education is a big interest of mine (it's why I'm in grad school in the first place). If you are interested in working something out, poke me -  I'm always up for interesting conversations!