State of Drupal presentation (May 2016)

DrupalCon New Orleans comes at an important time in the history of Drupal. Now that Drupal 8 has launched, we have a lot of work to do to accelerate Drupal 8's adoption as well as plan what is next.

In my keynote presentation, I shared my thoughts on where we should focus our efforts in order for Drupal to continue its path to become the leading platform for assembling the world's best digital experiences.

Based on recent survey data, I proposed key initiatives for Drupal, as well as shared my vision for building cross-channel customer experiences that span various devices, including conversational technologies like Amazon Echo.

You can watch a recording of my keynote (starting at 3:43) or download a copy of my slides (162 MB).

There is much more data hidden in the raw survey results, so if you'd like to do your own analysis, you can download a copy of the raw survey results (CSV format or XLS format) and look at the raw data yourself. I anonymized the data by removing the names, e-mail addresses and IP address information. If you decide to analyze the raw data, consider sharing your findings with all of us.

Take a look, and as always feel free to leave your opinions in the comments!

Comments

Ivan Stegic (not verified):

Thank you, Dries for continuing to lead with compassion, empathy whilst trying to build consensus for what our community truly needs. I spoke with some folks after the keynote who were offended with your desire to increase reach and richness and took offense with the Apple iOS comparison. I tried to reason that you were *not* saying that the Drupal community should be like Apple and iOS, but rather, that we should be aspiring to greater reach, greater richness of experience for *all* users, no matter their technical background. That we should be focused on getting better and polishing our product even more. I think that there is a part of the community that is concerned that they will be left behind and that they won't be able to do with Drupal what they have always been able to: use it with small non-profits to do what they need to. They are concerned with being a walled garden.

Any thoughts or clarifications?

May 12, 2016
Dries:

The iPhone was just an example of a product that has both a lot of richness and reach. The point is that we should aspire to both greater reach and greater richness so we can optimize our impact on the world.

May 12, 2016
Ruby Sinreich (not verified):

I would really prefer to look toward products and communities that are more aligned with Drupal's open source values. Apple seems antithetical to this. But my greater concern is about what we could lose if we compromise Drupal's strengths in order to emulate products with entirely different goals.

Generally software that is easier to use sacrifices a lot of control in service of usability. Microsoft Word can be used by nearly anyone because it has removed many decisions that would otherwise get in their way. Less technical users often prefer iPhones to Android for this reason as well, or really anyone who values ease and aethetics over control. One could argue that simplicity of user experience has been a strength for Wordpress as well.

One key factor that distinguishes Drupal is the power and flexibility of the platform. (I say this is primarily a site builder, not a coder.) It seems that this could be at odds with the simplicity that would enable a more iPhone-like experience. How can we retain Drupal's advantage while trying to be everything to everybody?

May 17, 2016
J.R. Maroney (not verified):

I was inspired by the keynote. I've always been impressed by Drupal because of it's ability to do more, whether that be built into to core or contributed. And I agree that Drupal will survive through innovation. If we don't continue to innovate at a rapid pace, well funded DIY platforms will surely eat our lunch. I'm looking at D8 for much more than content management, that is, document management, CRM, media management, and more. Lead the way Drupal!

May 20, 2016
Shannon (not verified):

Loved the keynote, specifically, I enjoyed getting information about the survey, your analysis of it, and the translation of that expressed desire into proposed & active initiatives. I think this is powerful. Getting to hear directly from the community to influence where the technology should go is like knowing what minerals & vitamins your body needs, then giving you a plate of food that matches that nutritional deficit. Now, if we can just get people to eat it, that is, contribute to those initiatives, we will come full-circle :)

I was also very impressed with the futuristic solutions that our community is churning out - the alexa business and nike IOT blew my mind. There is real disruption there, and it's possible NOW. Drupal could be shaping the future of these web experiences! How exciting?!!!!

Anyway, thanks for another great keynote! I'll be thinking about which initiative(s) I want to hop on :)

May 26, 2016
Anonymous (not verified):

Is there a reason this data is not yet published so other people can have a look at it?

There are obvious statistical problems on those slides, like for example the misleading normalization to 100% on a multicheckbox survey.

On top of that one could also see interesting results like correlation between answers etc.

June 30, 2016
Dries:

I don't agree the normalization is misleading. Anyway, I'm still planning to share the data so people can do their own analysis and draw their own conclusions. Stay tuned.

June 30, 2016
Anonymous (not verified):

Thank you for the will to share the data.

> I don't agree the normalization is misleading.

I respect your opinion, but I prefer to actually use points, not just faith.

You have seen all the data, so your brain knows maybe the actual data. When changing the data, which is presented to users, you maybe not loose this knowledge, but others will pick up the bias right from the start.

There are multiple problems. The most obvious one is, when everybody votes on F1 (Given there is F1 to F5). As result of this "normalization" you get a number like, let's say, 30% for F1. Its not obvious, when you show this number to people, that actually everybody voted on it. One major problem is that you cannot distinct this case from the case, in which there are less total amount of votes, but F1 was not voted by everybody.

PS: One common tool to optimize for maximum marketshare in multi-response surveys is called TURF analysis [1]. It helps you to figure out similarities between answers, and by that, avoid traps in privatizations.

It would be interesting to see whether this groups/collapses a lot of the similar answers of the survey into one group.

[1] http://www.econ.upf.edu/docs/papers/downloads/1197.pdf Introduction page

July 01, 2016
Dries:

I exported the data from SurveyMonkey and uploaded it to my site: you can download a PDF summary or the detailed results as a Microsoft Excel file. I anonymized the data by removing the names, e-mail addresses and IP address information but otherwise left things in tact. I've also updated the blog post with this information so people can find it more easily.

As mentioned, I'm planning to analyze the data some more myself and will share the results on my blog. That shouldn't stop you from analyzing it though. We all benefit from more people looking at the data. If you decide to analyze the raw data, consider sharing your findings with all of us.

July 01, 2016