For a year or more now we've been talking about portals at work. For some reason... I think it has something to do with dick-waving rights at the administrative levels... it's been decided that the University neeeeeds a portal. The central problem, of course, is that no one really knows what a portal is. Hell. I don't know what a portal is. If you look around, though, you can certainly find plenty of things called portals. In general they seem to be database-driven websites that consolidate all of the web applications that a given site's users care about. Seems simple enough.

I think, in the beginning, the concept of a portal was even simpler than that. If I remember correctly netscape coined the term to refer to it's site which was, as the word implies, a jumping-off point for a user's web experience.

At TU, though, we have a much grander VISION. Our portal must be, to paraphrase my direct superviser, infinitely flexible and extensible. Huh? Basically, what he wants is an infrastructure that lets us delegate management of portions of our site to organizations or business units. It must also provide a simple interface and a number of prebuilt modules that will let these managers build neat sub-sites that integrate seamlessly with the whole.

Ok. We can work with that. In fact that's a damn good idea.... in theory.

For me, as a system administrator, having a well-conceived web infrastructure usefully solves a lot of concrete problems. It's very useful to be able to develop web applications in an environment where many of the fundamental problems like authentication, state management, and database access have already been solved. Trying to view the situation in any other light gives me accute heartburn, however.

The issue of publishing content in an easily accessible online fashion has been well solved. HTML is brilliant in its simplicity. Anyone can learn it and use it. I think, however, that we have this idea that a "portal solution" is going to solve the broader, non-technical problem of interesting people in publishing their content. In the end departments and organizations on campus that wish to have sites are doing so... and adding a layer of complexity and the perception that we are are taking some of their control away is going to make them very unlikely to wish to participate. On the other hand, the entities that are not publishing most likely aren't going to do so for a variety of reasons... and I don't think we can alleviate any of the problems they face. In truth how much easier can web publishing get than it was back in 1994?

Ironically, I think that applications like front page and golive have just cluttered the world of web publishing with uneccesary obstacles. If you can learn one of those tools you can learn HTML. Period. Portals, as some people here seem to envision them, would just further that sort of obfuscation.

In the end if we could just come up with or buy a framework that lets us (my fellow admins and I) solve the problems we care about and provides the users with easy ways to add things like authentication and authorization and simple web applications to their preexisting sites we would be ahead. To allow the sort of marketing bullshit that the word "portal" bundles to obscure our thinking about web publishing is a mistake.

Anyway.... just trying to clear my head because this has been a troubling and frustrating issue for me.