Enterprise Drupal Revisited: What A Difference Two Years Makes

This week I submitted a proposal for a panel talk at DrupalCon Paris called Continuous Testing and Drupal:

This talk will briefly explain the concept and value Continuous Integration (CI), but focus primarily on the nuts and bolts challenges and techniques of getting a CI environment working with Drupal. The touchstone for our presentation is a case-study of the infrastructure developed by The Economist Group (economist.com) to utilize four teams in five timezones from London to California to effectively collaborate on developing a complex, high performance Drupal application.

That work has been a lot of fun, and it's inspired me to resurrect an old theme here on the Chapter Three blog, a tag I like to call "Enterprise Drupal." It's been a while since I first posted on this topic after being inspired by a 2007 OSCMS session featuring Ken Rickard and Jeff Robbins, and with some lively audience participation (IIRC) by Ivan Labra, Jonathan Lambert, and Robert Douglass. Those were the days.

Since then a lot has happened. Drupal 6.0 hit the big time, and has continued driving the amazing growth in community size, application downloads and the Drupal services market. Fearless founder Dries Butyart helped launch Acquia, a company predicated at least in part on a belief that the answer to the question "Is Drupal and Enterprise Solution?" is "Yes!" My own business venture did a lot of great work and grew from a three-man partnership to a fully-staffed Drupal consultancy.

And so I return to this question about Drupal in large-scale institutions -- "enterprise" isn't just about big companies, but governments, universities, and more -- inspired now not just by my belief in the code and community, but by extensive first-hand experience with these sorts of implementations and their technical, cultural, and organizational challenges. Once again, this is the first in a series of posts.

The Lay of the Land:
At a high level there are still many challenges ahead, but the door to pervasive Enterprise Adoption is more wide-open than ever. In the spirit of my colleague Zack's SXSW-inspired post on The Future From Texas, here's what I see happening:

• Increasingly, proprietary CMSs are not just costing and arm and a leg in licensing, but also bottlenecking innovation and driving up TCO. Enterprise site owners have digested the lessons of "Web 2.0" and want to integrate their properties with other things on the net, but are held up by clunky architecture and outmoded/arcane stacks.
• The commoditization and maturity of next-generation hosting resources and technologies such as Amazon's EC2 and memcached have made it possible for open-source/LAMP applications to go from pilot to full-scale production in rapid fashion, and with amazingly little sunk cost.
• Methods for managing and maintaining an iterative development environment and practice are similarly entering the mainstream, including within the enterprise. Nobody wants to be stressed-out all the time, especially when there are better alternatives, which happen to be more profitable and less risky than traditional large-scale development/management methods.
• Even as IT in general faces cuts, nearly two-thirds of large-scale site owners plan to expand their content management bugets in 2009, more than half of them in the name of improved user experience.

In our own neighborhood, Druapl 6.0 moved the ball forward significantly on many fronts. It's cloud-friendly, supports the jQuery UI library and has good support for simpletest in contrib space. As compared to Drupal 5.0 (or 4.7, which was where we were when I made my first post on this topic), the current release of Drupal is an order of magnitude more "enterprise ready," if not more, something we can see clearly in the continued expansion at the high-end of the market.

People wonder whether or not the growth of Drupal can continue, and while it clearly can't go on for ever, I'm of the opinion that it's a long way to the top. Other tools and services will rise and drive hot new internet waves, but the core application of Content Management has a vast audience worldwide, and Drupal is uniquely positioned to mature into the go-to open source solution over the next two to three years. There are vast numbers of developers around the world who are only beginning to tune in; tens of millions of active sites is actually a conservative estimate as to the total potential install base.

And when it comes to the marketplace, there's a huge amount of greenfield at the top-end as well as in an ever expanding base of small to medium-scale sites. We're working on a research whitepaper to demonstrate this, but take to heart that even as we have conferences attended by thousands and shops landing seven-figure development contracts, there's still a whole lot of headroom in the space. Just to give you some idea, Open Text, the 800lb gorilla of CMS that just ate Vignette, had a gross income of around $750M last year. And that's just one company running off an increasingly antiquated and arcane stack of tools (.NET, tcl, J2EE). The total market is well into the billions, and the open-source/LAMP share of it is only going to grow.

The trends are clear, especially with the coming release of Drupal 7 which will be well equipped for high-performance architecture out of the box, support rigorous unit testing in core, and sport an impressive array of usability improvements. Add that to a maturing marketplace for services at all levels, increasing availability of high quality documentation and training, a new drupal.org, and amazing contributed modules bringing more and more eye-popping functionality to the masses, and I'd say there's another Perfect Storm brewing. Attention Drupal Planet: prepare another surge in popularity, community, and market-share.