Since I became converted to WordPress a couple of years ago, finding it easier to build new websites with it than using the Windows / .asp / vbscript-based content management system I’d actually lovingly handcrafted myself from scratch five or six years ago, I’ve never ceased to be amazed by its power and versatility – power and versatility which has only increased during the last two years, and for me increased exponentially since I learned how to create my own themes (and most recently, started learning how to create plugins).
I’ve persuaded Birmingham to use WordPress in-house for our own microsites (with apologies to my friends in the local multimedia agency community…) with reasonably successful results – our weather disruption site helped us keep people informed during the bad weather (though I’m still narked that you-know-what failed to notice it when they did their annual survey…), and our commercial property management division is now using a WordPress-based site in order to try to shift its offices and shops to let (sorry, I know the ‘browse properties’ menu has its usability faults – I’m planning on trying to fix that next week!). Our Birmingham Newsroom website has been so successful it’s won awards – real ones! Other councils (Lincoln and Shropshire councils were in the room represented, and the inspirational Staffordshire Hoard website is also running WordPress) have also used WordPress for microsites, for its flexibility, speed of deployment, and agility.
But could WordPress run an entire council website, with thousands of pages of content using a devolved authorship and publishing model of a community of a couple hundred web editors?
Regardless of what you might think of the content strategy of such a website, it’s a question which would need resolving before any council webteam or IT infrastructure team could remotely seriously contemplate taking such a bold leap. The obvious answer of ‘well wordpress.com runs off one single instance of the same WordPress multisite that you install yourself at home’ doesn’t really answer the question adequately – for one, we don’t know what kind of massive server farm wordpress.com is running on compared to what any given council’s site is running on, and for two we don’t know what extra hardcore security measures wordpress.com has applied to it which haven’t been – for good reason – open-sourced.
In the session we had a free-flowing discussion, with plenty of points made for and against. Although as a session we didn’t come right out with a complete agreement that WordPress could run a whole council’s website, we were clear that as far as we could tell as the hive mind, there’s nothing definite saying it couldn’t.
Things which occur to me to keep in mind if one is going to seriously consider such an endeavour include:
- Be clear about WordPress’ limitations before you start; specifically, accept that if you’re going to go down this line, you will need to fundamentally rethink your existing content model. If this is a good thing, go ahead! If that’s not such a good idea for your situation, then Jadu is still a very fine content management system, by all accounts.
- If you are going to proceed, then make yourself fully aware of how the WordPress content model works, and plan your content strategy around it – learn all you can about WordPress tagging, categories, ‘static’ pages, posts, etc; you’ll almost certainly want to make full use of the little-understood WordPress custom taxonomy feature which lets you assign multiple groups of tags to your pages / posts / articles / whatever you’re going to call them, sorted and separated sensibly rather than serving up one big bowl of useless tag soup.
- Make a decision early about whether you intent to go with one single WordPress site, or a network of sites using the WordPress Multisite feature, and plan accordingly before you actually start migrating content. It’ll only be a major pain in the bum if you want to change your mind later.
- Take advantage of everything WordPress offers in terms of agile development; rather than jumping straight in to designing your theme with all of its graphics etc in place, design your layout as a working wireframe theme first, testing and consulting, testing and consulting. Extend your wireframe theme upwards with its graphical elements after the layout has been finalised. Don’t forget to also design a much simplified theme which can be switched over in an instant if a major disaster occurs in your town putting extra load on the server.
- Share your development process along the way – start off internally, then share it with a few trusted external people (selected according to user profile criteria), gradually widen the net, maybe even run the two sites in parallel before doing the big switchover. That way you and your IT team should be about as confident as it’s possible to be that part of the answer to this post’s question has been found.
I don’t want to trot out a list of plugins to install on this post, but here’s a few things which have caught my eye just in the last couple days and, indeed, during the hour I’ve spent writing this post:
- How To Run A News Site And Newspaper Using WordPress And Google Docs – case study of how a newspaper went to a web-first workflow publishing model, very akin to many councils’ workflows, and information about and links to download the specific plugins they’ve used to adapt WordPress to work with that model.
- Network Theme for WordPress and Buddypress – a premium WordPress theme which will be invaluable to take a look at and rip the guts out of (disclaimer – I’ve not looked at the licensing situation for the theme so it may not be possible to rip the guts out of it itself, only ‘learn from it’) if you’re committing to the network of sites model, in order see how to share menus and content from one site to another.
- Site lister function – a quick little function shared by Steph Gray which can be used to list a network’s sites (and from there, feeds from sites and content, if your .php skills are hard enough).
Enjoy your flight!