Last week the long-awaited WordPress.org plugin directory was rolled out. While this re-design has been in an open beta for months. Many users and developers felt taken off guard by the launch, which happened without warning.
In a comment on the announcement post, Kevin Hoffman summed up much of the community’s frustration about what was supposed to be a user data and feedback driven re-design he said, “In many of the discussions leading up to launch, user data and feedback were promised to be the decision-makers, although it did not exist at the time. Much of the early feedback (regarding tabs and Read More links) was framed as “developer feedback” and not necessarily indicative of the 80%. The fact that those decisions proposed in July 2016 are still present in yesterday’s launch suggests that some type of user data must have validated those decisions. Otherwise a lot of folks who offered that early feedback will feel it was ignored.”
This has been a long, community driven project that is only at version one. I spoke with several members of the community about what they think about the plugin directory refresh, what they hope will come next, and how those who want to can best get involved in shaping the future of this important part of WordPress.org.
Who Does This Help?
The new plugin directory has two major components: a new search algorithm and a new design for individual plugin pages. For the most part, everyone seems to be fairly happy with the plugin search. That’s a super important feature in terms of helping users find the right plugin to fit their needs. While not perfect in version 1, I think it’s a great improvement that will help all stakeholders.
In regards to the new plugin page layout, I’m not sure who this new design helps. Placing all of the content on one single page doesn’t make it more readable. Removing the prominent display of contributors doesn’t help developers build their name or users develop trust with developers they know and trust.
Removing links to older versions of the plugins doesn’t help users maintain their site or aid in figuring out when bugs were introduced. Removing stats from the page can’t help anything.
One new feature is showing the last two reviews, in full on the main plugin page. These reviews are not selected based on usefulness or reputation of the submitter. Allowing any random person on the internet to insert content into the main page for a plugin doesn’t provide any real help for other users, while it could be easily abused.
The rest of this article is thoughts from me and others in the community and I think it’s important to frame this discussion of what is wrong and what can improve. We need to look at “who benefits and who is hurt by these changes.”
My perspective is from someone who develops WordPress plugins, attends numerous WordCamps, teaches WordPress, contributes to WordPress core, has 12 WordPress-branded items on his desk, and more. So I’m not a typical user.
I spoke with Mark Uraine, a WordPress community member who works at Automattic as a designer that worked on the design of the directory, and he reminded me about the needs of less vocal users. These users, those “who generally use WordPress, but may not contribute back to the open source project. We don’t hear from them too often in the more obvious channels, so we created user tests on usertesting.com and sat down with users at WordCamps, etc.”
This Is Just The Start
Before I go on, I want to point out that one of the major reasons for this change was to make it easier to iterate on the system. According to Samuel “Otto” Wood, in a comment in the AdvancedWP Facebook group “The directory is now in WordPress itself and not in an old hacked up version of bbPress. That’s a big deal. That means we can actually make changes. Also, that notably means that the code for the directory is open-source and you can see it and even make patches for it to be added.”
Obviously, the immediate focus is on fixing any bugs that the roll out caused. As anyone who has ever released a major re-write of a site or piece of software knows, these things can’t always be found without going live and letting user fix the bugs. This is a frustrating process, but it’s probably the only way.
Hopefully, once initial issues are sorted out, bigger improvements will be incorporated and more community contributions will be included.
Uraine assured me that community feedback is being taken into account. He said that there is “a strong desire to remove the ‘read more’ accordion links and I’m working on alternatives right now.” He pointed out that the accordion system was used for more parts of the layout originally.
Who Are We Building For?
I’ll be the first to admit that my perspective is based on my goal for the WordPress plugin repo – acquiring free users for my freemium plugin Caldera Forms. I don’t apologize for that. If people don’t buy the paid add-ons for my plugin I can’t keep building it or spending my company’s money on traveling to WordCamps to give talks, or using my company’s time to contribute to WordPress core.
The plugin repository isn’t Github, it’s not really a place for developers to share and discuss amongst developers. It’s a place for site builders, implementors, and programmers to find an important component for their site.
Of course, the WordPress plugin directory is for users, but I think that can’t be at the expense of developers. Improving the developer experience to me means making it easier for plugin authors to serve their users.
For this reason, I am focusing this article on the user/developer relationship, and not the design aspects.
Was Feedback Accepted?
In the first community meeting on the plugin directory after the launch Matt Mullenweg said, “FWIW, I feel like my feedback was ignored as well. I hope we can do another major iteration on the directory because I’m not really a fan of the new one perhaps the WP backend will make it easier to make incremental improvements in the future.”
I reached out to the team at Yoast SEO for their feelings on the community involvement in the process. CEO Joost de Valk said that “We’ve had some feedback and gave that during the weekly meetings on Slack, most of that was acted upon almost immediately.”
When I spoke about this article with Matt Cromwell of WordImpress, makers of Give, he disagreed that the community was excluded from the process. Instead, he argued that the Community didn’t show up to the weekly meetings where many of the important decisions on implementations were made. Cromwell regularly attended the open meetings about the redesign and said: “I felt like I was representing the community during the meetings as much as I could and I did speak my mind, and I was heard and I did help make change happen.”
Of course, having an opinion and seeing your advice being taken, is not the same thing. As Matt said, “Did I get my way on everything? Nope. But I contributed.” The plugin repo is huge, with almost 50,000 plugins and so many people relying on it that there is no way everyone will be happy.
Is This An Improvement?
Personally, I’m not in love with the look of the plugin pages. The old system had tabs, the new version has one long page. The sections are broken up into accordions that often cut off paragraphs, and opening an accordion section with a ton of text is a pain.
I like that it’s all one page load, but I wish that the tabs were actual tabs, switching out content. I asked Brian Hogg, a WordPress plugin developer and creator of the Making Pro Plugins course, if he thought the new design was an improvement:
Hiding the majority of the full description behind a ‘read more’ is by far my biggest issue with the new design, which hinders the ability for users to easily see what the plugin author wants them to. This includes not only how to use this plugin and additional resources, but also how the plugin might fit together with other plugins or with related add-ons, and tips on how to make the most of it. If plugin authors want to make the description long and push the inline plugin screenshots and FAQ sections further down, they should have that choice.
While I suspect the design will improve, I also wanted to know if others felt like this would help or hurt the process of growing and supporting a freemium plugin business. I think this is important because without healthy business behind these plugins the innovation and continued support of these plugins will not continue at the same pace.
Joost felt like it would. He told me “If anything, the improved search means more users will be able to find a plugin that’s useful to them with good ratings and support. So we fully expect we’ll see user growth go up after this.”
While the plugin repo was in beta, there was a feedback survey. That data and analytics data has not been shared publicly. It’s unclear what data exists and how it was used.
If one thing is clear, it’s that this is just a start. All of the code is now open-sourced and anyone can create a ticket or submit a patch. Just like WordPress core, opening a ticket or submitting a patch doesn’t guarantee a change will be made, but it’s how discussion and change starts.
Joost told me that he felt like this was just the start and that his team has “already opened a ticket to improve the stats given to plugin developers and we’re preparing a ticket to allow for a minimum PHP version.
Brian Hogg told me, “Overall it would be nice if there was more of a priority towards the needs of those offering a free solution with paid upgrade and support options. Without the freemium model, a large amount of the plugins that millions rely on wouldn’t exist.”
I agree, but historically the plugin repo has not been designed with plugin business in mind and I doubt that will change. That said, given the importance of the WordPress plugin repository, I do think it’s fair to let plugin company’s control this important landing page for their business.
Hogg suggests that, “If plugin authors want to make the description long and push the inline plugin screenshots and FAQ sections further down, they should have that choice.” I agree with him on that and removing the last two reviews from the plugin page. Allowing any random person on the internet to insert arbitrary content into the middle of a plugin page seems like a bad idea, and I hope that capability will be removed before it is abused.
It’s Open Source, Contribute
Since the new design launched, there have been a lot of feedback. A lot of it negative. But those who are involved keep reminding us that this was the first big step and it provides a system that is easier to build on. That system is open-source, and the old system no longer needs to be maintained, so I think we can expect some major improvements, which is exciting.
There is a tool on Github for building the open-sourced parts of WordPress.org, including the plugins directory inside of VVV. You can use this tool to contribute code to the project just like you can with WordPress core.
Uraine stressed that he was excited to incorporate feedback. He told me, “Visuals are super helpful! Some people would fork my code pens and send me their thoughts visually. I loved those!” The code pens for his original mockups are in this post and the most recent mockup of the single plugin page can be found here.
What Do You Think?
So far the new plugin directory has mixed reviews. I’m not a fan, but I appreciate the strategy of open-sourcing it and shipping a version one to attract more feedback.
This is a BIG change for WordPress and everyone should feel free to submit a ticket on meta trac, attend the weekly meetings or contribute a patch. Let’s keep it respectful of course, and work together to take this from a version one that needs work to something very awesome that better suits the needs of developers and end-users alike.