When DIUS launched its Science and Society consultation in July 2008, I took the opportunity to throw the kitchen sink at a consultation, digitally-speaking. Not all of it worked (in fact, hardly any of it did, you could argue), but I learned some useful lessons and the policy team have maintained their appetite for engaging online. My first proper WordPress site long outlived its intended lifespan, continuing as a blog with a bunch of pages for expert groups of scientists to continue their deliberations in public.

But as these things do, the limitations of one-thing-turned-into-another became more apparent over time, and it became clear it was time for a rebuild. We’re putting that live today.

Simon Dickson of Puffbox has done a nice job on the project, cleverly deploying WordPress multi-user to host a set of linked blogs for the groups, with a unifying homepage and RSS feed. It’s a good fit for the job, enabling the secretariats to the groups to manage their own presence, post up minutes and draft reports, and use WordPress widgets to promote special announcements.

But there may be the makings of a more profound point here – which Simon has made to me before with a curious smile – about government web platforms of the future. I’ve been asked a few times over the last month or so ‘How large can a WordPress site be and still be workable?’. I’ve tended to suggest, perhaps, 500 pages as a workable limit, unable to conceive of managing complex content trees and thousands of pages in a tool built for blogging.

But will the government websites of 2015 need to look like the behemoths of today? In departments where policy is key, stakeholders need to be involved and where ‘whats new?’ is the primary question users ask, could the departmental sites of the future be a series of linked blogs, written by enfranchised and enlightened policy officials, engaged with by stakeholders, and summarised on new lifestream-style corporate homepages supported by a meaty document and data library search and solid archiving?

I’m not pretending this site is the future here and now. But it’s an interesting thought.

Oh, and watch that curious blue bar at the top. There’ll be plenty more of that if Neil gets his way.

Update: Simon blogs his part of the story

Update 2: I failed to mention in the original post that Jenny project managed this project, keeping a tricky internal supplier (i.e. me) on course. Thanks Jen.

Get notified of new blog posts by email


They all ready have for many small companies and see no reason why this can’t be expanded to larger sites as you have explained above. Sometimes we all get caught up developing mega rich content sites and forget the lesson less is more, especially if focused and inergrated with social media tools in a coherent strategy.

Hi Steph, I like the sound of your blog based future.

I’ve not been in Govt for as long as many people here, but it seems to me that policy team wishes generally seem to be one of the following:

– A stream of announcements, either new or noteworthy (i.e. your average blog)

– Document store

– Ability to message the audience (most have heard of newsletters but all seem to get RSS, Twitter, Feedburner notifcations when you show them)

– Feedback mechanism (everything from a panel with plain and simple contact details to commentable pages, twitter accounts forums and more)

– Handful of other page types (e.g. speeches, RDFa consultations)

– Interactive features (quiz, polls, commentable docs and whatever you invent next Steph!)

Throw all this together on a cheap and easy to customise platform and you’re on to a winner.

I’ve been thinking about a few ideas on top:

– How can users who’ve happened upon highly specific content such as Society & Society see the bigger picture and navigate to broader themes such as Science Comms > Science > Innovation generally > Growth generally

– Wouldn’t it be nice if urls and docs could stay the same irrespective of MoG changes (give every post, every doc and organisation a joinedup.gov permalink?)

– How great would it be if users had a more seamless experience when navigating from one govt department site to another? Could we throw in a bit of personalised content at the same time?

You’re probably way ahead of me already, I’m just throwing these thoughts in just in case. I’ve been sketching a few bits and pieces and can post them somewhere when they’re a bit more together.

I’m keen to hear more about any joined-up-without-being-centralised initatives.

If Neil gets David’s way, more accurately, but I’ll admit to being a bigger than average fan of joining gov sites up with BBC-esque topbars.

You and I have discussed this policy blog farm vision a couple of times before and the more I think about it, the more I agree it should be the future. It has potential like nothing else to create direct, unnmediated engagement between officials and their audiences (which can only lead to better policy-making), and for a fraction of the cost of the behemoth sites. And how much static content is useful in an increasingly real-time medium? Beyond the basic facts about the organisation and what it’s about, some info on established legal and policy frameworks, arguably not much.

Things like accessibility, copyright, writing effective web copy, legal deposit of publications, propriety, parliamentary privilege, consistent messaging [and so on] would still need to be managed somehow, but oughtn’t scupper the idea of delegated, direct communication. In fact, I rather suspect it can’t be stopped.

@Will – great ideas as ever.

Great stuff – I tend to think of a content cloud which includes everything we do digitally and its then a matter of working out how to get users to interact with it. My big issue at the moment is its fine to create stuff but whats the point if no one reads it (is that enough anyway – change behaviour, start a discussion…), or does anything with it – pdfs anyone.

I imagine that over time users will have their own dashboard (igoogle type) and pull or let the relevant content be fired at them and that the sort of static sites we have been used to in the past have a limited past, i.e. agree with above – of course. Makes me wonder therefore about the future of the supersites….

I’ve finally managed to get my own slightly more tech-centric up on my own blog:


We had such fine plans to synchronise blog posts, too. Sorry I let you down, Steph. 🙂

Interesting suggestion from Will C about maintaining links regardless of departmental changes. One idea to toss in: there is an open-source, PHP-based, WordPress-friendly ‘short URL’ engine which might be the basis for this. You’d still be looking at some manual intervention and some crafty find-and-replacing if/when sites changed; but if it’s all in one place, maybe that’s bearable?

Great work, and I’m all for more use of MU for projects like this. Any news on whether/when MU and standard WP will merge?

This comment was originally posted on Puffbox.com

@Robin – you’re literally too kind. It’s an idea which seems to make sense, but I’ll withhold judgement on its awesomeness for a while yet.

@Neil – the merge is definitely happening in the v3.0 release: it’s ‘mainly completed’ in the latest development builds, and is said to be ‘going well’. The project schedule currently proposes a Release Candidate on 29 March, and a final release on 13 April.

This comment was originally posted on Puffbox.com

It’s far too early to declare our Science & Society project a success, either as a site in itself or as a model for building a larger website from a collection of smaller blogs. But it’s certainly an interesting idea, and we’ll only learn about its true potential by trying it. Hopefully that’ll then make it easier for others.

Your core point is absolutely right though. And in my mind, it’s further reinforcement for the open-source, lightweight and (dare I say it) WordPress way. If big expensive CMS solutions can’t do what you want, and small cheap CMS solutions can’t either, then the latter wins on cost and manageability. And if it’s open source, then you have the right to build whatever add-on you feel is required. Bingo.

This comment was originally posted on backpass.org

[…] The stakes are pretty high, if you ask me, with the reputations of individual digital teams and the profession as a whole at the mercy of what their chosen system will let them do. (“That cool thing you saw on that website you like? Sorry boss, we can’t do that with our CMS.”) Vendors should be mindful of the power they yield, for as long as they still yield it. […]

Having initiated a full-scale procurement exercise for a local government CMS implementation, as so many other web professionals in local authorities, one of the biggest barriers to ultimate success of the selected product is invariably the vendor’s claims as to the product’s capabilities versus the reality experienced by the web team when trying to deploy it in line with their original aspirations.

With the benefit of hindsight I would have much rather preferred the vendor to have admitted that the product couldn’t do all that we aspired to achieve with our selection, and was better prepared for selling into the local government marketplace.

This comment was originally posted on Mission Creep | Neil Williams

Can I float a few more ideas Neil? Here’s a few of the things most Govt webbies want (well Corp ones anyway):

1) MoG proof and no link breakages ever

Links break when we move pages and files between depts. So why move them? How about uploading a page or file and it stays at the same hmg.gov.uk/… location forever. You could use bit.ly style unique urls hmg.gov.uk/13t4m4 or something more friendly like hmg.gov.uk/joinedup.

If you want it to be available on a particular dept’s site, tag it. If that dept changes just rename the tag or add a completely new one. Re-tagging en masse is way easier than shipping files around or conjuring up hundreds of page to page redirects.

2) Show the big picture

If all files and pages are tagged with a common taxonomy (needs to be much more granular, layered and rooted in plain English than IPSV) then it’s far easier to show the relationships, including those cross government.

3) Global design language

I think we all want greater visual consistency (BBC example at http://twitpic.com/13t4m4) compared with what we currently have (http://twitpic.com/13taa2). Sites don’t all have to look the same, but they shouldn’t require users to re-learn navigation as they pass from one Govt site to another

4) Common templates

A speech is a speech is a speech irrespective of the Minister that made it or the dept that he or she is part of. Nuff said.

5) Common faceted search

See results from This site/ Related sites/ All Govt sites across the entire network and provide the ability to filter in a logical and usable way.


For me this points to a common approach to building corp sites across government – taxonomies, templates and perhaps even the CMS technology that’s used. One CMS even?

But rather than level down, any system should be open enough to allow technical contributions from any quarter. Each organisation ought to be able to make customisable local versions of the core stuff if needs be.

I don’t massively care if it’s WordPress or something else. What Corp webbies want is something that’s cost effective, stable, easy to use and customisable with code snippets and widgets. It could be a hotrodded version of an existing CMS or built on a framework – Symfony anyone?

It won’t work if we get bogged down in change control boards and endless paperwork. It should work if we all want best practice and are happy to chip in to reach a common goal.

Just think…

Re-launches could be a thing of the past. Who needs them when the platform(s) and widgets are constantly upgrading?

Convergence headaches would magically vanish too. You don’t want to converge with us? Well, how about converging with hmg.gov.uk. In fact, why not converge all sub sites with hmg.gov.uk – it’s MoG proof.

The good doctor talked about the tyranny of moving content from one CMS to another following a MoG change. Well, what if that move involved leaving the pages and files where they are and simply re-tagging a few sections. We could re-tag a policy section in 5 mins.

If this is tyranny, bring it on. Less time shovelling links, more time concentrating on effective communications.

Before anyone says it, I’m not advocating one monolithic dinosaur of a CMS but rather common or one tech to save us all a lot of time. It needs decent data in/ out as I’m sure off platform work will continue to thrive – exactly as it needs to.

Hope this makes sense and hasn’t put people’s backs up too much.There are some sketches to go with this if I ever finish them.

This comment was originally posted on Mission Creep | Neil Williams

That’s an interesting list of ideas, some of which I’d like to see in non-gov systems as well. When scoping a CMS, how hard is it to get past the vendors claims and get real-world feedback from technical and non-technical users? I’d love to see some transparency in user case scenarios rather than a list of capabilities.

This comment was originally posted on Mission Creep | Neil Williams

As ever a provocative and well informed post. I do not know much about CMS and I rather frightened of these beasts. My ideal CMS would have something of the spirit of Tumblr and Posterous. A CMS should enable you to get content out with minimum hassle. Also CMS should not only publish content to websites but also to ’social media’. In a small way Hootsuite kind of acts like a CMS for social media.

This comment was originally posted on Mission Creep | Neil Williams

Dare I suggest that some of these features are (or at least should be) met by the government super-sites? And that consequently the way to meet your requirements is to converge onto a single CMS platform rather than attempt to standardise multiple products? Some degree of tongue in cheek there, clearly…
I think a lot still boils down to needing a better communications about what WCMS are meant to do and a better way to procure and implement them. As my post suggests, that’s not just a problem for government.

This comment was originally posted on Mission Creep | Neil Williams

I just voted up the idea of auto-suggesting tags.

Glad to see IPSV is mentioned, but the real power in IPSV is in its contained “non-preferred terms” ( or “plain english to IPSV mappings” as I prefer to think of them).


There are links from my post to the iks-project.eu site, where they are trying to orchestrate and formally document requirements and eventually solutions to problems such as “semantic lifting” (term extraction).

I also concur with the remark that Posterous and Tumblr-like tools could have some role to play, but then again I am bit of an extremist in that I think in principle all your images should be on Flickr – all your CMS should do is manage the relationship with your user(s) and Flickr.

Is there a place for such a “Federated content management” tool, it all boils down to what you understand as being content.

I send an email to posterous, and it ends up on my website as a new article.
I take a photo with my phone and it gets sent to Flickr (via posterous, haven’t tested this personally)
I tweet something with a #tag and it ends up on my website as a flash-info news ticker.
I enter something into my Tumblr account and it ends up on my website as a new FAQ.

This is me perhaps making the incorrect assumption that in this discussion by CMS you mean a WCMS.

I personally dont like the idea but it might be that a Federated Web CMS is a layer that can be constructed atop of your existing CMS with the right hooks.

This comment was originally posted on Mission Creep | Neil Williams

I’ve just voted for ‘User friendly publishing’ feature that should include an ability to add the unlimited number of pages, quick image upload, ability to write meta data, alt and title for images
+ convenient WYSIWYG editor
+ preview and publish on the fly

I’m not sure if you mentioned “Infomodule” in your Feature list, but it would be a nice addition to a website. Your visitors will be well informed as soon as they enter your website.

This comment was originally posted on Mission Creep | Neil Williams

(If you’re averse to my usual WordPress propaganda, look away now.)

I left a comment on David Pullinger’s piece: ‘ask not what open source can do for you, ask what you can do for open source.’ The real beauty of the open source concept is that if you can’t find a tool that’s perfect, you’re entirely within your rights to take the one that comes closest, and built upon it, to make it your perfect tool.

We collectively, and BIS specifically, have built considerable momentum behind WordPress. I’d suggest it’s the platform which comes closest to what most people would want. Sure, it isn’t designed to be a large-scale content management system. But it is designed to be enhanced, forked and cloned beyond recognition, if you so desire. And the GPL licence says you’re free to do so.

Just because you haven’t seen something done on WordPress, just because there isn’t an ‘off the shelf’ plugin already in existence, doesn’t mean it isn’t easily done: the last couple of versions have added significant ‘under the hood’ functionality with huge potential, and there’s more due in a month or two. You won’t have seen these possibilities if you’re just writing blog posts into a fairly basic installation. But they’re there, and they’re often remarkably easy to activate and build upon.

There aren’t many of your killer features which couldn’t be done relatively easily, relatively quickly and relatively cheaply – with a custom theme framework, a few plugins, a bit of experimentation, and the right people.

This comment was originally posted on Mission Creep | Neil Williams

@Will and @Simon – makes absolute sense to me. It needn’t be *just* WordPress (or just Puffbox!) – I’m all for choice in the marketplace. But I’m also all for tools that are easy to work with, build upon, feed back into the pot so there’s less duplication and reinvention; and with more coherence and consistency than we have now for both design and content. Making that happen goes some way above my pay grade, of course, but suffice to say the idea of it excites me personally.

@Will, definitely keen to see your sketches and remember you’re more than welcome to guest blog them here!

@John, Andy – ah yes, the sales folk. Not that they’re dishonest: most platforms *can* do all the things you want; it’s the amount of effort involved in implementation that’s the issue. And usability, usability, usability. Too easy to focus on features and forget the poor folk who have to use the back end.

@paulG – yes I mean WCMS throughout. Agree re IPSV. Not sure I agree about Flickr being all that important! But get the point about thinking differently about content.

@Phillippe – actually, this sort of consolidation is happening now through the website reviews (aka rationalisation and convergence) led by @digigov. Great strides being made there, by no means baby steps.

Thanks everyone for the comments, votes and ideas. Keep them coming.

This comment was originally posted on Mission Creep | Neil Williams

Hi Simon,

Great blog post. I wonder if for the WP fans you might mention a couple of things which tripped us up when playing with an MU site. 1) the switching between dashboards function is a bit hidden -you have to go to ‘My blogs’ and then to an individual dashboard – it’s ace when you know how but we were all stumped at first. 2) That Scrib’d/Youtube/Flickr embed thing where you just put the url on its own line is a thing of joy and should be shared more widely

Well done with the site.


This comment was originally posted on Puffbox.com


Yep I did see your post and have been keeping an eye on the Uservoice – done a little bit of voting as well. I’m not sure I have many ideas that aren’t covered there at the moment – my big bugbear is usability on the publishing side of things – if it was easier for ‘normal’ people to handle day-to-day editing and publishing I think that would free up enough time to solve alot of the other issues!

This comment was originally posted on backpass.org