Core Plugins? Never Gonna Happen!

This is a guest blog post written by Andrew Rickmann, author of the blog, Fun With WP.

There’s an interesting discussion happening on the wp-hackers mailing list right now about the issue of core plugins. Although it seems to have meandered away from the point slightly, the key suggestion was that WordPress should offer a set of core plugins so that APIs that are not needed by every user are available in a standardized form. This has been taken further to refer to features.

Ever since I started using “The Blog Platform That Shall Not Be Named” I have wanted core plugins for WordPress. TBPTSNBN uses these to great effect and allows it to remain lean while offering a wide set of features. I just don’t see it happening with WordPress though.

Throughout the discussion the posters are talking about content that the core team do not think is right for core. This displays one of the key issues with WordPress: the philosophy up until now has encouraged more and more to be added to the core, and users are consistently asking for more to be built in. The core team act as gatekeepers in a way, preventing content from getting in that isn’t widely useful.

Adding the concept of core plugins will significantly lower the barrier for these ideas. Not good enough for core? perhaps a core plugin then. But someone still needs to develop it; it still needs to be up to the standard expected of WordPress, and with only so much labour to go round something, somewhere, is going to suffer. Eventually it will suffer a lot.

It wouldn’t surprise me if some core plugins started to become out-dated and that would be worse than not having them at all.

Of course it also means a lot more arguing about ideas. It is common for users to demand something that is entirely possible already, but which they don’t know how to do. I don’t mean a feature as such, but displaying certain content in a theme for example. The core team will need to constantly explain why idea x or y is better left in the hands of themers than being a core plugin, so they will have even less time.

It will also open all of the old wounds.

Only a few days ago WPTavern was alight with discussions about the need for a UI to turn off post revisions. If ever there was a feature that was suited to be a core plugin it is this. Most users don’t need it, or want it, but you can’t argue that for taking WordPress seriously as a content management system, it is essential.

How many features will the developers find themselves forced to make into a plugin instead of putting in the core? How much time will be wasted arguing about each one? How much time will be wasted modularizing old features with no progression?

No, it just isn’t going to happen. Partly because of the points above, but mostly because the fundamental philosophy of WordPress doesn’t allow it.

WordPress is a consumer product. It is developed as a finished product into which other people can plug-in, not as a collection of parts that can be put together to create a whole. New features are added to WordPress to address the general needs of the community, and I just don’t see an appetite to spend valuable development time to move WordPress from an install-and-go product to a blog ‘kit’ where optional components can be turned off, or on.

I might be wrong, of course. Perhaps the developers would be up for it. But while I think there are great gains to be made by modularizing, I also think the downsides could be crossing-the-streams bad. I won’t argue too hard for the things I want then, just in case I get them.

15 Comments


  1. But, doesn’t WP.org *already* have “core” plugins?

    can see the benefit/necessity of Akismet. On the other hand, Hello Dolly is an utter waste of resources – and yet it is installed by default.

    So, the question isn’t whether or not WP.org will have core plugins; the question is, what will the devs allow to become core plugins?

    Report


  2. I agree that taking functionality out of core into plugins is a bad approach. However, if Automattic / the gatekeepers do a good job of keeping the core clean, other projects could follow in the footsteps of buddyPress which is essentially a series of social media plugins with WordPress MU + a theme designed to support that added functionality.

    There are some very heavy weight plugins starting to come towards quite functional – PODS comes to mind – which present a more interesting set of questions about what should / should not be in the core in the future. Too much and you get bloated feature creep. Too little and it seems like the foundation has grown stale.

    Report


  3. @Chip, I think Hello Dolly is included to keep things light and as an example of how easy it is to create your own plugin. It sucks that you find it a utter waste of resources.

    Report


  4. Pretty much everyone on the list (myself included) disagrees with what Mike is proposing.

    I very much doubt myself that any new core plugins will be made and included with WP. Automattic has many useful plugins themselves, but they are not included with core, because they aren’t of use to a significant majority of users.

    Features should never be moved from the core into “core plugins”, but rather into normal plugins hosted on the WordPress.org Plugin Repository.

    @Chip Bennett – You’re mistaken in saying that Hello Dolly is a waste of resources and that it is installed by default. If you’re concerned about a 2.17kb file, then you really need a better host. Additionally, it acts as an example file to developers on how to develop plugins.

    Report


  5. I don’t see Akismet as a core plugin. It is bundled, sure, but it is a one off, not the product of a philosophy to do this sort of thing.

    Ryan, I think there is value in some cases of producing core features as plugins, and if the repository had a way of distinguishing more official plugins then I would agree. I do think that if plugins were developed that were intended to form part of a WordPress release then it would be important to clearly identify them so they were more connected with official releases.

    I don’t see the value in using the core, or official plugins, simply for the purposes of standardising back-end code which I think is really what is being suggested on the list. Standardisation is very easy, just come up with the most extensible option that fits everyone elses needs and they will use it.

    Report


  6. @Ryan: So, Hello Dolly is no longer included in a fresh install of WP.org? If true, I’m glad to see they removed it. It has been there every time I’ve installed from latest.zip (as opposed to upgrading e.g. via WPAU).

    For my purposes (and I daresay, for *most* WP users), Hello Dolly *is* useless. It adds nothing meaningful whatsoever to my blogging experience. That it is only 2.17 Kb is irrelevant; those 2.17 Kb are used needlessly – thus, they are wasted.

    The claim that Hello Dolly acts as an example plugin for developers is also a specious argument with respect to justifying its inclusion in the default install. The vast majority of WP.org users are *not* developers, and have no need of an “example” plugin.

    All that said, my main point remains: certain plugins *are* included in the default install of WP.org. Such plugins are, for all intents and purposes, “core” plugins.

    Report


  7. I couldn’t agree more on Hello Dolly – it adds no useful functionality and, for the vast majority of users, simply adds another step to the installation: having to remove it.

    It’s not about its size (actually 4kb on the disk) but the unnecessary cluttering of the plugin page. If developers need “an example file on how to develop plugins”, I’d suggest they download it along with the necessary documentation.

    I appreciate a light-hearted gag as much as anyone, but this joke isn’t funny anymore.

    Report


  8. @Lloyd: How many (better: what percentage of) WP.org users will ever develop their own plugins? I’ll be generous, and say 1% (not sure of the aggregate number of users to which that percentage equates).

    Of the remaining 99% of WP.org users, how many/what percentage find the *functionality* of Hello Dolly useful/beneficial? I would find it hard to believe that the number would even be measurable.

    Nonetheless, let’s be really, really generous, and say that 5% of all WP.org users would find Hello Dolly beneficial, either as a developer template or in its function to end users.

    Now, I have to ask:

    What *other* plugins are useful to *more* than 5% of all WP.org users?

    Please note, I really don’t have a dog in this hunt. I’m comfortable with installing/uninstalling plugins, so whether some functionality gets moved into core, or some plugins become designated as “core” plugins, or the plugin situation remains exactly as it is now – no matter what happens, I can make my blog work the way I want it to work.

    Report


  9. Hello and thanks for the post and the comments to the idea of core plugins. I have write a post over this point, actually only in german – later in english. I write you and i hope youcan read the nice comments on my blog over this idea, maybe you can translate with a service?

    Report



  10. I´m actually against core plugins. keep the extra functionality to the community. If it´s really a good plugin that is needed badly, someone will pick it up and continue to develop it even if the original author stops development. It already happened many times.

    WordPress IS an end user orientated platform and you should never underestimate the unknowingness of the end user. Some people just don´t know things that are basics for people taking part in those discussions. If it doesn´t work out of the box like intended, they will stop using it.

    Report


Comments are closed.