Is Drupal the Right Tool for the Job?
Yesterday, at a campus-wide meeting of web editors and developers, UC Berkeley announced a major new hosting initiative for departmental sites running the Drupal content management system. The service is a partnership with Pantheon, who provide turnkey hosting for Drupal sites. I've got mixed feelings about the announcement.
These observations are written in the context of university hosting, but many of them also apply to news organizations trying to decide on their next CMS.
Many campus departments budget for a "web editor" but not for actual developer resources. As a result, a shocking percentage of departments have no content management system at all. These departments desperately need a system that can help take their static sites to the next level, with a minimum of programmer intervention. The UC budget crunch is real, and directly affects what departments are willing to spend. If they are going to need a developer's help to make the CMS leap, they need to be able to hire those developers at reasonable rates. The initiative provides infrastructure and developer resources affordably.
I was impressed by the Pantheon system we saw demonstrated. It's young, but appears to be extremely well-engineered. It provides Development, Testing, and Production environments, full integration with the git version control system, and looks like the kind of system users will be able to get going on with minimal training. Bravo to the team at Pantheon for building the "just-right" solution for Drupal hosting, and bravo to the University for selecting what appears to be an excellent provider.
Between the UC Berkeley Graduate School of Journalism and the hosting/consulting business I run on the side, I'm involved with the construction and/or hosting of almost 200 sites. In my role with the Knight Digital Media Center, I'm talking to journalists constantly about their CMS situations. Over the past decade, I've come to the conclusion that the vast majority of sites currently deployed in Drupal could have been built more quickly, more easily, and with less frustration, if they had been created with WordPress instead. Yes, Drupal is more sophisticated than WordPress, but there's a very high cost to pay for that additional sophistication, and the vast majority of news and departmental sites simply don't need the few features that Drupal still offers over WordPress (especially since WordPress gained its "custom post types" functionality).
I speak to journalists and site owners frequently who regret having gone down the Drupal road. The usual scenario is along the lines of "It started out great, but became increasingly difficult to manage as we started to add features, and upgrading has been a costly, frustrating nightmare."
Part of the problem is that so many people still vastly underestimate the power of WordPress, and think it's not sophisticated enough for their needs. This statement by Pantheon CEO Zach Rosen is characteristic:
“WordPress is wonderful if you’re a blog or have a small business site with a limited number of pages, but it isn’t suited for big enterprise sites,” Rosen said.
Unfortunately, this characterization of WordPress simply isn't accurate. WordPress' capabilities go far beyond the needs of blogs and small businesses, and there are thousands of major publications, companies, and university sites running on WordPress. Many of those sites includes tens of thousands of pages, many disparate content types, and sophisticated features beyond what you'd expect from a "blogging" platform. For example, take a look at the site for Bates College, done entirely in WordPress. Or check out the constellation of WP-driven university sites at bu.edu. Still think WordPress is a mere blogging platform?
Looking hard at the "real needs" question can potentially save site owners lots of time and money, and can streamline the future evolution of those sites (for more technical details on what I see as Drupal's limitations, see my Drupal or Django: A Guide for Decision Makers).
Of course I acknowledge that there is a level of complexity beyond which WordPress is no longer ideal, and that a more sophisticated system is needed. For those sites, Drupal may be the right choice, though I'd still suggest that if you're in that position, you'll be better served by a modern framework like Ruby on Rails or Django.
The important point is this: As I surf through UC Berkeley's roster of departmental web sites, I see almost nothing that couldn't be reproduced quickly and easily with WordPress.
By doing so, those departments would gain substantial advantages in simplicity of development and maintenance over and above what they'll get with Drupal.
At yesterday's unveiling, it was made clear that the Drupal decision was reactive to the fact that there's already a lot of Drupal activity on campus, and that a similar WordPress solution is a possible future consideration. We have no idea how long it will take the university to get there, if ever. In the meantime, Drupal is seemingly being encouraged as a de facto standard, since it will become the only official CMS solution provided by the university. Dozens of departments will "accept the defaults" and go with Drupal, buying into a level of complexity that far exceeds their actual needs. In my opinion, this is a shame, and not a great way to spend scarce university funds.
Here's a sampling of responses I received from webmasters at other universities after I tweeted about the announcement yesterday:
[Drupal] has a massive cloud of hype around it. That said, I lead our [similar] initiative and would've kept my damned mouth shut if I could do over!
I'm gradually moving away from [Drupal] after paying many pounds of flesh in the university where I used to work.
Again, Pantheon's done a great job on the platform, and the University should be commended for rising to fill a long-overdue need. I just fear they've chosen a non-optimal tool for the job relative to the needs of the actual departments on campus who need it.
See also: 17 reasons to pick WordPress over Drupal