The Gartner hype cycle and Drupal

I'd like to discuss some observations the technology analysis company Gartner Group has made about product lifecycles -- and how their model affects Drupal release management.

The Gartner hype cycle is a model that posits five market stages for all emerging technologies. It looks like this:

Development cycle

Here is how I think the hype cycle's stages played out for Drupal 7's development cycle.

  1. Technology Trigger. As Gartner puts it, this is where “A potential technology breakthrough kicks things off". For Drupal 7 that occurred when we started development on 4 February, 2008.
  2. Peak of Inflated Expectations. This is a period when excitement about the forthcoming release makes people dream big. We entered this phase shortly before Drupal 7's release, as news of all the wonderful things we were doing entered the consciousness of early adopters.
  3. Trough of Disillusionment. This stage came after Drupal 7's release, as people examined it closely to decide whether it would work for their new projects. Some discovered that a necessary module wasn't available yet, or that they had to learn some new APIs. Many chose “the devil they know" in Drupal 6 while cursing Drupal 7's differences. That's understandable, and it will pass.
  4. Slope of Enlightenment. I believe this is where Drupal 7 is now. It is where modules are being upgraded, bugs are being fixed, and the whole ecosystem is becoming more stable overall. Some people adopt it, while others still wait.
  5. Plateau of Productivity. This is where we can breathe easy. It occurs when developing in Drupal 6 seems pointless, because the Drupal 7 ecosystem exhibits more promise.

That last stage is where a product really shines. I think we reached the Plateau of Productivity for Drupal 6 sometime in late 2009, about 18 months after its release. At that point there was no advantage to developing in Drupal 5, and Drupal 7 was still a long way off.

I have my own opinions about Drupal 7's path through the hype cycle, but would like to hear your thoughts. When do you think Drupal 7 will reach the Plateau of Productivity?

Comments

Tom (not verified):

This is an interesting exercise, but are you sure that it is appropriate to apply the Gartner hype cycle specifically to Drupal 7? I feel that it might be more appropriate to apply it to Drupal as a whole, instead of to a specific release. Or even to CMS systems as a whole, just as you would apply it to cloud computing, mobile internet (remember WAP?), etc.

Raf (not verified):

I guess you could apply it to both: Maintaining both a macro- and a microview. It's actually pretty typical to developments. You don't go around coding one huge chunk, either. You cut it down to small chunks and focus on those, while often taking a peek at the bigger picture.

As to the topic at hand, I guess that productivity platform'll be reached at least next january -- a year after its launch. There're significant API and theming changes. There seems to still be some movement in the theming changes (see http://drupal.org/node/455844 and http://drupal.org/node/1158090)

Because of these movements, and with the "core" modules (OG, Views, ...) still being in the works to be fully D7-ported, and the vast amount of smaller modules that haven't been ported yet at all, but are needed for development and installation profiles, the earliest I see everything stabilizing and getting widely adopted, is about a year after launch.

I think that, once the dust's really settled, not many people'll stick to D6. Personally, Entities is enough for me to not stick to D6 once everything else's stabilized. No more need to install a small army of modules to get some cross-data (user - nodes - taxonomy - comments) to play nice. It's one module to rule them all!

Rene (not verified):

Reaching the "Plateau of Productivity" depends on the level of users. For module developers, I think it has a lot to do with figuring out entities. From the programmers I've talk to, coming to grips with this is a major obstacle with module releases

For non-programmers who are more UI Centric, it revolves entirely on contrib modules getting their act together (hence the first point). For all the modules that made the so called pledge, it's amazing how many are still in dev, beta or simply have no releases for 7.

Senu (not verified):

As less and less shared hosting users use Drupal 7, and the internet now being rather than a net of individual sites, crystalizes to the monopoly of Facebook, Apple, Google there will be less of small individual site owners running their own sites - the typical mass who popularized the very initial versions of Drupal. Thus plenty of sites using D5 and D continue to run those and will continue to run those Drupal versions, which means they are out of this cycle and will be so for indefinite time, as long as their php based shared host support the PHP version that lets Drupal 5 and 6 run without a problem.

Parasolx (not verified):

For me, the time for D7 reach Plateau of Productivity when most module contributors wouldn't support anymore for D6.

I more prefer work with D6 for developing system compare to D7 even it provide much more flexibility to core. But, I will change my decisions when most of D6 module stop in their support.

And of course, it is the start of life cycle for D8.

Morningtime (not verified):

I had the same thought as Tom. I think Drupal 7 is the peak of inflated expectations. Drupal 8 will be the trough of disillusionment. Drupal 9 the slope of enlightenment and "Drupal X" the plateau of productivity.

After Drupal X, you need to watch out for becoming an "open source MicroSoft" - a dying dinosaur. But by that time CMS will be very different and "websites" as we know now will have disappeared. Maybe something will replace Drupal in the future, like Facebook replaced MySpace, or Drupal may become the Yoda from the past, like Unix.

Whatever happens, it's like evolution: if the best idea is too expensive, then the next generation will adopt a more imperfect but cheaper solution.

;-)

Wim Mostrey (not verified):

Cross-posting from Chapter Three's excellent post on Training in Drupal 6 vs Drupal 7.

I do believe that Drupal 7 is currently on the slope of enlightenment. There are two key parts to challenge: knowledge and guidance.

People might be hesitant to start building sites in Drupal 7 because it is currently the path less traveled. We need guides to show people the way, to tell them where to mind their steps and where the shortcuts are. Training plays an important factor in gaining this knowledge. Continuing to provide Drupal 6 trainings in itself gives the perception that Drupal 7 is not ready yet, which will lead to slower adoption, and it keeps the path less traveled. We should lead the way and travel the path.

David Eads (not verified):

I agree with Tom in questioning how accurate this is as a model of Drupal 7's evolution, and I'd question the model itself (is productivity a "mood"?).

But instead of using this generic model to describe the impact of Drupal 7 to various audiences, I'd rather see a "hype chart" based on Drupal 7's particular history. And I'd like to talk about whose mood and behavior we're talking about. Different segments of Drupal's developer and user community will visit these peaks and valleys at different times, or not at all.

Personally, I'll be in the trough of disillusionment and an overlapping but slowing eroding plateau of productivity for at least another 6 months as I have to balance the tradeoffs between Drupal 6's maturity and limitations vs. Drupal 7's remarkable improvements and significant regressions.

Agnar Ødegård (not verified):

The nice thing about D6 is that if you hit a snag, there's a fair chance you can find a tutorial or guide somewhere. Resources on D6 are ubiquitous.

If you're currently dependent on the occasional tutorial to get D6 working, why would you use D7 where you might not find solutions to your problems? It's a risk many are uncomfortable running today.

WillyK (not verified):

While it's interesting to apply the cycle to D7, I think that we can't look at the D7 technology/product in isolation.

A key to the success of D6, apart from some great technology is the evolution of the Drupal ecosystem as a whole -- i.e., the momentum and growth of contrib, Drupalcon, companies with Drupal expertise, adoption by enterprise clients, and example of industry leading projects, and most importantly a strong roadmap for the future. It was the combination of these factors that lead me and my clients to adopt Drupal, something we passed on with D4 & D5 as it was not crystal clear until D6 was rolling. Further, if you expand the curve, there is ideally an overall upward trend from one version's cycle to the next. That sort of continued upward momentum will challenging to maintain as an incumbent with legacy issues to support and new nimble competitors emerge (particularly those without the browser/desktop legacy that our community must maintain). While continued good core/contrib technology are going to be essential to D7 and future versions, I think that our fate (D7 and 8/onward) are going to depend more on whether our ecosystem as a whole evolves to carry us up the slope, across the plateau and to the peak of the next cycles.

While D6/7 success were/are crucial, I think the real test of time will be emerging from the D8 trough at which time the online/wired world as we know it is going to be a very different world (as Andressen noted at D9 we'll *finally* be doing the things that people talked about 15 years ago and which created the dot-com bubble, http://on.wsj.com/ln8sYx).