Dries Buytaert

Speed matters

More proof that speed as perceived by the end user matters. This time from a Google Research paper (PDF). Google's experiments demonstrate that increasing web search latency 100 to 400 ms reduces the daily number of searches per user by 0.2% to 0.6%. Furthermore -- and this is where it gets really interesting -- users do fewer searches the longer they are exposed. In other words, the cost of slower performance increases over time and persists.

To use Peter Van Dijck’s words:

In other words, if your website is a little slower, users will use it less (we knew that), but they'll also use it less and less over time, and when it speeds up again, they’ll still use it less than before the slowdown.

Based on their observations, Google suggests site builders to think twice about adding a feature that hurts performance if the benefit of the feature is unproven.

State of Drupal (September 2009)

Two weeks ago at DrupalCon Paris, I gave my traditional state of Drupal presentation. The video of the presentation is available from archive.org, and you can download a copy of my slides (PDF, 8 MB) as well.

I don't want to give away the spoiler, but essentially, the state of Drupal is strong. :) We should be really proud of what we have accomplished with Drupal 6, and what we're about to accomplish with Drupal 7. In the presentation, I also talk about what it means for Drupal to grow up, and what the next phase of our life will most likely look like.

Drupal can help pay for your rent

The demand for Drupal talent continues to exceed the supply! What will we do about it?

Drupal 6 growth

The graph above is made based on the project usage statistics collected on drupal.org. As ever with statistics of this sort, they don't tell the whole story. This is because only sites running the update status module report data back to drupal.org. This module is part of Drupal 6 and the installer prompts the user to enable the module when Drupal is first installed. It is not required to enable this module. People upgrading from Drupal 5 aren't even prompted to enable it. Plus, many Drupal sites are hidden behind corporate firewalls. As a result, we don't really know how many Drupal 6 sites there are.

Either way, based on the growth data that we do have available, we can predict that we will near 240,000 Drupal 6 sites by January 2010. See the black trendline on the graph. The R-square represents the variability in the data set that is accounted for by the prediction model. Its value indicates how likely the predicted values are -- the closer to 1.0, the better. If Drupal 6 continues to grow like it did the past 9 months, our prediction should be pretty accurate. It would mean that the number of Drupal 6 sites will double over the next 9 months. Not bad.

I always believed that the best way to grow Drupal is to make the software better, and that is why we continue to work hard on Drupal 7. But until Drupal 7 is released, there are a lot of things that we can do to help people get started with Drupal 6 -- from offering .zip-files instead of .tar-files, to launching the drupal.org redesign, to sharing more Drupal 6 success stories, and more. There are a lot of barriers that could be removed and that would result in faster growth.

Once in a while, it is good to make lists. You're all invited to share your list of "things you think we should do" in the comments of this post. I recommend that you prioritize your list so the most important item is at the top. If you are actively working on any of the items on your list, let us know too. Bonus points for ideas that have high impact, require minimal effort and benefit the Drupal community at large. Penalty points for people that recommend Drupal 7 features based on self-interest. Ready, steady, go!

Faster is better

Through a presentation from Nicole Sullivan, a former member of Yahoo’s Exceptional Performance Team and co-author of O’Reilly's upcoming book on performance optimization, I came across the following data points:

• Amazon: 100 ms of extra load time caused a 1% drop in sales. (Source: Greg Linden, Amazon)
• Google: 500 ms of extra load time caused 20% fewer searches. (Source: Marrissa Mayer, Google)
• Google: trimming page size by 30% resulted in 30% more map requests. (Source: Marrissa Mayer, Google)
• Yahoo!: 400 ms of extra load time caused a 5 to 9% increase in the number of people that clicked "back" before the page even loaded. (Source: Nicole Sullivan, Yahoo!)

While we all knew this was true (and while I'd like more detail on these tests), it is nice to have some quantitative data from different sources. Long story short: even the smallest delay kills user satisfaction. Let's make Drupal even snappier! (Hat tip: Peter Van Dijck)

State of Drupal presentation (August 2008)

Last week at DrupalCon Szeged I gave my traditional state of Drupal presentation. The video of the presentation is provided below, and you can download a copy of my slides (PDF, 11MB) as well.

The presentation discusses the results of the recent survey that I conducted; the survey ran for 30+ days and collected more than 1300 responses so it should provide a good idea of the community's current thinking. I'll provide more color and details about the survey results in a number of follow-up posts.

Google insights on Drupal

Recently, Google launched Google Insights. Like with Google Trends, you can just type in a search term to see search volume patterns over time, as well as the top related and rising searches. You’ll also have the ability to compare search volume trends across multiple search terms, categories (commonly referred to as verticals), geographic regions, or specific time ranges. Great for marketing people.

Below are some examples specific to Drupal ...