Radio Userland

Radio Userland, or Radio, as it’s being called, is . . . huge. It’s more than a blog tool—because of the syndication, or “broadcast” features it offers, and the “news” database, which allows you to quickly, easily see headlines from other sites, and link to them.

There’s a reviewof Radio here on WinPlanet which offers a good description of blogs in general, but also suggests they are a Mac phenomenon. I’m not sure why the author sees blogs as a Mac thing–given the dominance of WIntel, I rather doubt it.

On the other hand, Mac users do tend to go for smart technology, and good design. Radio, while showing signs of being “smart technology,” is pretty low on the “smart design” scale. It violates a number of basic, standard Mac interface conventions, and the documentation is appallingly awful, in terms not only of the English but in terms of the lack of a useful instructions and a coherent structure. It’s far too frustrating for the average user— it looks and behaves like an early Beta version, not professionally released software. I wrote tech support a few short, specific queries and am looking forward to a response. Radio has enormous potential, not just for consumers, but in education.

Yet More Blogs—And some Perls

So in my frenzied selfless search for blog tools, in between writing my dissertation, I took a look at Radio Userland. Cool. I think. Yeah, cool, but I think it’s more than it seems. It lets you automatically fetch “headlines” from a list of sites. I chose to look at O’Reilly’s, and found a link to an article on the perils of Perl for the unwary, specifically, finding and downloading .CGIs. There’s a nifty checklist of things to watch out for, worth knowing about. Part of the article is a plug for .nms, a SourceForge project, as a source for solid Perl scripts, but it is nonetheless interesting.

More Open Source for Education

Yes, I’m still looking. I’ve found a couple more sites that are collecting information about open source projects of interest to education. There’s an article in NewsForge about SchoolForge, but the stuff that looks viable to me is mostly at the more traditional sites like Fresh Meat, or CPAN, the mother lode of Perl modules. Which reminds me, there’s now a well organized site at, for learners.

Software doesn’t have to be New to be Good

Lately I’ve been talking to others in IT who are, like me, interested in solid products with good interfaces,and are taking advantage of stable but geeky standard protocols and unix tools and applications but putting a web front end on them.

I’ve been thinking about doing this with an NNTP server. Network News Transport Protocol is what makes UseNet newsgroups work, from a server perspective. Most of the discussion board products are monolingual; UseNet News is not, and readers are designed to support pretty much any language someone would post to UseNet in. News is threaded, and there are a number of ways to integrate HTML front ends. Why not use News as a discussion bard?

There are scads of News servers, but I keep hearing about DNews News Server, which runs on Macs, among other platforms. It’s particularly interesting since user authorization controls allow read and/or post access to be restricted for particular users or newsgroups; NetWin, the developer, also offers dBabble, chat server and webNews, a .cgi for a web front end to a news server. NewsRunner is a neat Mac application you can point to your News Server, and it will convert posts to HTML, text, digest form, email, or database and archive them. We were early adopters of Web Crossing, which supports UseNet news groups via the web or a client, email, private discussion boards with customizable templates, chat, and ssl connections.

CMS Crises, or Opportunity?

Yesterday, my interest in blogs and CMSs got a lot sharper; WebCT announced a deadline for ending licensing for 3.1x, the version we use on E-Campus. We’re going to have to decide what to do. We do have lots of alternatives, but it means my research and personal interest in CMSs and blog tools is suddenly professional and vital. But, you ask, what’s to decide? Why not just go with the flow and upgrade? Ah, well, there’s the question, and it means starting at the beginning, or at least in Spring of 2000, when I wrote a report on our LMS options, including an informal review of the various LMS products. We ended up staying with WebCT, even though I already knew it was deeply flawed, because it was the best option for our users at the time.

The Once and Future E-Campus

A year later in Spring 2001, I wrote an informal proposal, passing it by my manager and programmer, and discussing it with the chair of the faculty advisory committee, and then later, giving the new Assistant Director a copy.

Since I wrote that software review in Spring of 2000, BlackBoard has risen in popularity, WebCT fallen, and numerous small CMS systems have been swallowed by the two giants. Most recently, Promethius, built with ColdFusion, was bought by BlackBoard. Microsoft has a cooperative arrangement with Blackboard, privileging NT. And, like WebCT users, Blackboard users are not happy with licensing changes, judging from the complaints in the Blackboard users’ list.

I, like the software engineer and sys admin I work with, have been following the growth of Open Source popularity, and have even looked at various open source content management systems. I’ve not been impressed with them; sure, they are fine for use with one or two classes, but we host roughly 2000 or so classes, with over 25 thousand accounts. Even WebCT has jumped on the Open Source bandwagon. I’ve followed with interest the MIT/Stanford Open Knowledge Initiative. These collaborative projects have some great potential, though none seem to have much code yet. The Learning Online Network from Michigan looks promising. Frankly, the idea of open, public standards, particularly support for things like Unicode, are crucial. Yet as Blackboard and WebCT move towards databases, they are also moving towards proprietary data storage, including encryption.

The more I think about this, the more I am persuaded that the best solution for our students and faculty is to create our own tools and infrastructure for class site and content generation and management based on our templates and open standards. We need an infrastructure designed to allow us to include modules for specific tasks, for both “Designers” or instructional staff, and students, an infrastructure that will allow us to grow our class site support as our users grow. Such a system would allow us to offer the best of tools, and the features our users want, without being tied to any specific platform or publisher.

The danger of course is that such local systems, systems created by a small group of developers, make the institution deploying the system dependent on that group. This potential problem has already affected other local systems on several campuses. However, none of the local systems I’ve researched have been designed professionally, with real specifications, published APIs (APIs with the standard descriptors, the sort one expects from a developer), with documented and commented code, and properly tested via QA and standard user testing. Moreover, institutions are just as dependent on a commercial developer for support, and given the experience we and other have had with WebCT support, or Blackboard support, or support for the many products purchased by the two giants, I feel a local host system is an opportunity, rather than a risk.