As the three people who regularly read this blog know, I work at Penn State. Those of who know Penn State know it is big. Somewhere in the 80,000 student range across all the campuses. There over 40,000 here at our University Park campus alone. We are starting to talk about giving blogs (as James Farmer would say). Clearly we’ve thought about why it would be a good thing and are now moving into the land of what we would do and how would we do it. Our IT people are very smart and understand scale in a way I do not, so I am not going to argue with them when they discuss things like server load and security. I am lucky enough to be a guiding member of the team, so I am taking my thoughts public … I usually stay away from large-scale work projects here, but I thought since this is a blog project what a better place to start than to solicit feedback from the blogosphere. So please help us think about this!
What follows are some general thoughts about where we are after a bunch of general conversations and one meeting … oh and a lot of this is my thinking along with one other colleague — in other words, this doesn’t represent Penn State’s position on any of this. Please feel free to leave comments or email me ideas. First the basic assumptions:
Blog installation/activation managed centrally (vs. local installations) a lot like blogger.com to provide:
- activation via an easy to use control panel
- no access to underlying code
- blogs are published into individuals’ personal web space via a static publishing model (a lot like MoveableType or remote Blogger publishing). At PSU we provide faculty, staff, and students with 1 GB of hosted web space.
Why would we do this? Well, for one reason we can maintain a single code base for all blogs … so when things need upgraded we don’t have to do 40,000 updates at once. It also facilitates integration, ensures security, professional look and feel can be maintained on blogs (esp. important if student blog being used like ePortfolio), and we absolutely don’t want to end up with scattered, disjointed collection of blogs, and have no way to fully leverage this service.
Some of the Features
- Group blogs with multiple authors
- RSS 2.0 with support for enclosures to allow for podcasting
- Control Panel model
- Allow individuals to create and post to multiple blogs
- Access controls to enable public/private blogs as well as public/private posts on blogs
- XMLRPC – if it can be secured
- Tagging at some level — to be discussed further
- Categories – critical feature for the creation of custom URLs and custom RSS feeds
- Themes so end users can skin their blogs
- Blog Rolls – manage a group of links in multiple link categories
- Track backs
- Allow local search on a per blog basis
- Spam protection
- Allow for comments
- Comment controls — approval, edit, delete, etc.
- Text formatting — do we use a WYSIWYG editor for posting?
- Plug-in architecture?
So as we are going forward we have many questions and are in the early stages but are very excited about what is going on … we still have questions and are very open to suggestions. I would be curious to hear about others are doing at their schools to enable blogging in a quasi-controlled environment.
It’s going to be quite a challenge pulling this off right – i.e. getting good adoption while making the solution educationally valuable. I’ve expanded on the idea of adding social networking features as a strategy to increase adoption, at my blog: http://chrismillet.com/?q=node/5.
(The trackback module for Drupal 4.7 doesn’t appear to be working yet, so this will have to do…)
I think adding the features you are discussing would be a major plus, but how do we encourage them to create meaningful connections? As you mentioned the FB doesn’t really promote all that meaningful links between users. I continue to wonder if students will ever use anything we build them.
Cole – you’re right, adding social networking in and of itself isn’t going to guarantee meaningful connections. My point is that users need to have the means to easily create the initial connections, and do some simple things to reinforce those connections. The problem with something like FB in an educational context is that the system actually constricts the depth of the interaction, based upon what the interface allows you to do. Blogs a) have substantially fewer restrictions on the types and sophistication of possible interaction (esp. if you consider podcasting and vodcasting), and b) they’re not contained in a walled garden, so a lot of incidental interactions can occur.
As for will they ever use anything we build them… Some of the initial findings of my blogs in the classroom study are showing that student attitudes towards blogs become more negative throughout the semester, presumably because they start looking at blogging as just more work. Reducing barriers to entry by making this service act like something they’re already comfortable with and complelled to use can only serve open the doors a bit for something educationally valuable to happen.
Heya Cole – I would have commented earlier, but you already know everything I’m doing in this space 🙂
We’re using Drupal quasi-officially here at UCalgary, for weblogs.ucalgary.ca – it provides all of the functionality we need, and performs quite well. With some decent infrastructure, it should scale to 100,000 users easily – Big Sites like TheOnion.com run on it, so performance doesn’t have to be a problem.
Drupal’s strengths here are the community/connection possibilites – self-forming groups, blogs, podcasting, etc…
Also, Elgg has these same strengths, packaged in a different way.
Cole – I have taken the approach that students are blogging before they get here or once they get here they ‘hooked up’ with their friends. I would very interested to know the uptake of this project but my gut tells me it will be low _if_ you have a lot of students that already have their own blogs.
It is similar to uni email. How many students actually *want* to use it? Here it is very very few.
Jesse … we have similar indicators here … students that show up with blogs tend to stick to those blogs. University email is where we measure our users in a lot of cases. It is the most heavily used technology we currently have. More and more students are showing up with gmail for example and are sticking to it … but, they are still forwarding all their University mail into that space. Blogging is really a platform for all kinds of things and if they end up getting that we’ll see good adoption (IMHO). If we cannot convince faculty that it is the right thing to push for portfolios, note taking, intellectual development tracking, and whatever else , it will fail as a large scale program.
Darcy, we are close to committing to the static publish model — that could change, but it takes some very compelling solutions out of the mix.
NVU is a multiplatform stand-alone HTML editor that evolved out of the old Netscape Composer: http://www.nvu.com/index.php. For what it is (a low-end Dreamweaver alternative), it’s quite nice.