Out with the old, in with the new.
It seems an age since I posted here an appeal for input on a renewal of the European Parliament's website. Actually, it is an age! I suppose I had little idea then just how huge an undertaking this was going to be. Nevertheless, this week saw an important stage of the process completed. It's blue, it's new and it's online – the redesigned official institutional website of the EU's directly elected democratic lawmaking institution! And, speaking personally, I now know why you don't do this kind of thing too often.
An oil tanker of a website
The renewal of Parliament's website was always going to be a pretty ambitious undertaking. It is, truly, a vast and sprawling website, answering the needs of many different user groups. It provides daily news, live and recorded webstreaming of plenary or up to thirteen parliamentary committees simultaneously, a mass of documentary information about Parliament and how it works, profiles of every MEP, a database of their assistants, another database of registered lobbyists, a register of official documents, every conceivable form of parliamentary document (reports, agendas, working papers, resolutions, amendments, adopted legislation, procedural rules, etc.), sophisticated legislative tracking tools and search engines, details of all committees and delegations, their calendars of meetings, membership, agendas and meeting documents, studies carried out and commissioned by Parliament, a host of applications and online forms, and, as we were somewhat shocked to discover, between seventy and eighty sub-websites on various themes, ranging from the Sakharov human rights prize to the page or the Scientific and Technological Options Assessment panel. (Pause for breath…)
And all of that in 22 languages. (So it's really 22 websites, not one.) No-one has ever done online multilingualism like this.
I was never very keen on those colours on the old site
So it's big. It remains a remarkable undertaking and massive achievement. But it was also a bit long in the tooth and decidedly Web 1.0. The beast dated back in its essentials to 2005, and, in the intervening period most of the online excitement in Parliament had been elsewhere: special temporary sites for elections and hearings, EuroparlTV and all those cool social media platforms we've spoken about so much on this blog. The dear old website got a bit left behind, slowly, moreover, accumulating the clutter and detritus of time: all those mini-websites, random pages for special occasions, outdated content, promotional buttons for things that seemed important at the time…
And let me now finally admit this publicly: I was never very keen on those colours on the old site. (Funnily enough, we heard this same sentiment from a cool techie guy we met today, telling us why he liked the new version.)
A project (with a dodgy name)
In that slick way of public administrations everywhere, we called our project "The Renewal of the Web Presence of the European Parliament". In French, it was called the "Refonte". For once, the French was shorter than the English. Confession: I came up with the long and awkward English formulation, and have lived to regret it. The point was: "this isn't just about the website, it's about everything we do online". Good point, but still simpler to call it the refonte.
Call it what you will, the political masters told us to get on with it in December 2010 – yep, a year ago.
The first delivery was the mobile version of the site in June 2011. Check.
The second was the redesign. "Autumn 2011" we said, and here it now is. Check.
The idea was simple: we needed impetus, things had to move on, be seen to move on. And yes, those colours… So the first big thing on the main website was to get a new look. Of course, that wasn't the whole story. Though we couldn't envisage a major restructuring of the underlying technology in 2011 (budget planning oblige), we could do something to address some of the more pressing user feedback: palliate some of the navigation problems our user feedback was telling us about, re-organise the rather cryptically structured content in the "Parliament" – now "About Parliament" – section, address the fact that EuroparlTV was still an outlying website, introduce some basic social media features and make it much easier to find MEPs. So Phase 2 was in the end a little more than "just" a redesign. Hence the new portal, hence the new EPTV section, hence a great new MEP search tool, hence much more…
But the third phase is the technological, and indeed editorial, heavy lifting: rebuilding the CMS, integrating new features, a top to bottom rethink of the navigation, integrating social tools, real open data facilities, better decentralised publishing mechanisms and contemporary editorial content (multimedia, infographics, applications and so on). Oddly, phase three, though probably more significant in what is delivered to internet users, will be far less visible and contentious than phase two, consisting of new features progressively integrated over time, rather than a one-shot graphical transformation of the entire site, introduced from one day to the next, shocking to the eye, perhaps, but fundamentally an exercise in continuity.
Hands off my website!
Or perhaps, "who moved my cheese?" People grow accustomed to anything of course, develop habits and strategies, they know where they are. So when you change anything, naturally, they are unhappy and complain. If, as is almost certain, the switchover does not go smoothly and, as is 100% certain, there are myriad bugs and errors to sort out, the complaints will gain extra purchase. 'Twas ever thus, and thus it was on the great day, 29 November, when our lovely new site, which had behaved impeccably in pre-production, stubbonly refused to work, with several sections falling foul of inexplicable and grimly persistent server problems. A day spent waiting for news that the site is finally online, punctuated at frequent intervals however by reports that the problem is still proving intractable, is – believe me – a very long day. Most reasonable people, of course, will cut you some slack on days like this, and most did so, but that did not really make it any less nerve-wracking. So when, after thirteen hours of pure daylight angst, the message came through at 7.00 pm that the site was finally online in its entirety, and that the entire IT department had its fingers firmly crossed, the relief was palpable.
Of course, the end of the major problems (the crossed fingers worked) simply opened the season on the avalanche of complaints from people who "couldn't find something any more", felt aggrieved about their content supposedly being less "visible" or who perhaps didn't like the new colours. As the wise words have it: you can keep most people happy most of the time, but never will all of the people be happy all of the time. Of course, there are genuine bugs and gremlins. They will be sorted. Possibly, experience will show that some great ideas will ultimately be shown to have been less than great. But meanwhile, the job is to hold out, correct the genuine bugs and wait for people to get accustomed to the new look and feel.
Don't get me wrong: most feedback was extremely positive, especially that which came from outside. Here finally, said some, was an EU institutional website which didn't look out of place in the second decade of the twenty-first century: here too, observed others, was a site which set out to make things easier to find for a "normal person". Now you're talking!
The long road ahead
This was, yes, the easy bit
So, satisfaction? Yes. But though many may see this as the big transformation, we know it is not. This was, yes, the easy bit, even if all the under-appreciated developers, graphic designers, integrators and editors who sweated blood and burned vast quantities of midnight (and weekend) oil to achieve this result may not see it that way. However, this was our high-visibility moment, the point at which users will have the strongest emotional reactions to our work, when a angry reaction from a powerful individual could consign months of work to the trash. We're not out of the woods yet maybe, but my hunch is that the balance of reaction is positive enough to carry us through.
Of which I'm glad, because having done the easy bit, now we can really start to deliver the even better stuff.
Just wondered whether you saw this <a href="http://www.kosmopolito.org/2011/12/11/the-new-european-parliament-website-a-journey-of-discovery/ ">post by Kosmopolito</a>? I *think* he's being a little sarcastic, but I can't be sure …
Hi, Having read a few reactions to the new site I do feel for the team who have taken on the mammoth task of improving it but as someone who needs the site for their work, I do feel a bit hard done by. Ultimately we have to ask why a site is created – what is the purpose? To help people access the EU. There are two categories; EU citizens coming to the site afresh and those who know exactly what info they need. The latter would be a great source of inspiration for web designers. How much consultation was done and will be done with those relying on the site? When it was up and there were problems, how much were people kept informed of how long the disruption would last. How interactive is the site – I tweeted my frustration but had no-one to tweet to? Why choose a busy EP week to launch the site rather than a constituency week? My main issue however, is access to documents. At the moment the very documents that we need to gain democratic access to the EP are the hardest to find. As an example, the Committee docs are hidden several clicks in and when you do finally get to the draft agendas or amendment pages, they are not easy to sift through. Docs are still uploaded late. For those of us who do not work daily in Brussels, watching Committee sessions via webstream is vital but I find I can't watch them at all on my mac. I am pleased to hear that there will be continuous improvements to the site and would like to know how regular users will be included to ensure that we can get the best website possible.
A redesign always raises remarks. Like any other website, we read them, we like them.
Here are some thoughts we (the design team) gathered answering some relevant points from readers.
(by Fred)
First of all: thanks a lot for your input. Negative or positive we like it because it pushes us to reflect, and because it also means you share our ultimate end of enhancing the site… which means you care a bit for it.
We know: «Every time you redesign, God kills a kitten» (Lou Rosenfeld).
Well… we prefer tigers.
We’ll answer on what matters most for the graphical team: the visual aspect.
1- The «big» blue header.
It is a «parti pris» that we took since the beginning.
Thanks to “CE” for his (her?) proposal of sparing 78px height on the portal page, but that implies we have to spare it.
Well, we don’t need to :)
“CE” noticed that 85% of the users have a resolution higher than 1024 in Jan. 2011? OK, but we are already in 2012, what resolution will it be next year? and the year after? We are are not only making the website for today (even if we think the site fits well 1080+ screen resolution), we prepare it for the next elections.. and beyond.
But still, this kind of header is more and more common on the web.
We are a bit shy about saying so, but you could compare it to big portal pages like the White House, ABC, RAI, Channel4, Arte just to name some. Draw a line on your screen, you’ll see :)
But even before just being «trendy» we wanted to answer the user testing we made on the previous website(s). When they arrived, every single user (specialists, citizens, etc…) felt overwhelmed with information. They did not even know where to click.
OK, we took a 180° turn on it, and maybe for some it is a big change. We are not only thinking office PC screens (just for the sake of telling a story: just today, we’ve discovered that some of our primary users, officials, did not even know that they can change the resolution of their screen… and when we showed them they can switch to 1280 resolution it was like we opened the gates of Heaven).
The star of 2011 Christmas gifts? PC (MAC) Tablets. Here also, the blue banner works well, and what do users do on a tablet? They «finger» scroll. Like hell.
Scrolling: that’s also what we push the users to do: user testing proved us they did not know we had content lower down. Now, we’re actively provoking them to find it.
We wanted the page to have space, space, space, and to be clear and neat.
Once again: It is a «parti pris».
RSS feed and Print CSS: Thank you for noticing it. It is in the pipeline.
2 – Why fixed width / no progressive design?
First of all: We have to deal with some constraints.
It is the only website in 22 languages. In fact IT IS 22 websites each time. And what is good for Latvian is terrible in German (etc.). Believe us: we worked on the mobile version with no fixed size (without naming the «surprise» of each mobile platform that did not react the same way to CSS even in the group of the same phone family).
Plus, we receive content from various CMS. Yeah, we know: IT IS BAD. But we have to live with it for now and that means: different picture sizes, different title length, blank paragraphs, various times of contribution, various sources etc… for X websites (multiply by 22).
Secondly: We are the guardians of the visual identity of the site. For this to work we choose a fixed, sure, template. It is a challenge. Especially in the header / menu / footer. But that way we are sure that we are proposing menu entries and not a catalogue. Moreover we know how the page will (should for the moment, we agree) look like. Etc…
It is how we succeed in offering a visual coherence for the EP on the web.
Like Android or iPhone development tools: it is fixed: you have to follow the guidelines.
And we love rules and guidelines. We are tigers.
Some fine tuning is on its way.
As Steve pointed out: it is just the first phase.
The site will continue to evolve, step by step, taking our constraints and user inputs into account.
Best to all.
PS: this comment was written by Fred, not by me. I’m using my ID here ’cause Fred can’t find his password and we kinda feel lazy ;-)
Yep, multilingual sites would definitely put some constraints on progressive design (but then they put constraints on design generally). I'll discuss it with my circle of designers.
Hi Mathew,
Responsive / progressive design is a very interesting approach for web design. Nevertheless, we (well, our Web Studio) decided to go for a grid approach with a "to the pixel" sense of building the pages. There are many reasons why they chose this approach, one of them being the impossibility to garantee a correct result in 22 languages. Finnish, for example, might behave weirdly in the menu and/or in content's blocks when automatically resized.
I'd like to thank you all for your comments. The Web Studio checks this discussion daily and they might answer some of your remarks… when they'll be done debugging and polishing what still doesn't look right to their eyes right now.
Please bear in mind it's the first (and most) visible step of the website's revamp. Next updates (including the correcting releases) will be more discreet. Things will evolve constantly.
Best to all,
T.
Yep, I was surprised about the fixed width thing too. Why not go for progressive design? Content adapts to the user's screenwidth – e.g., images resizing dynamically, and content blocks rearrange themselves for the user, even reformatting themselves,. All done by one stylesheet, handling everything from a widescreen to a smartphone. No new technology required.
Overall, I feel that the redesigned EP website is heads and shoulders above the previous version, or any other *.europa.eu website for that matter. Criticism about the site navigation and hierarchy choices is normal and is likely to cease after a while, as users "find their way" again after their "cheese" has been moved. That is why I would like to put forward a couple of suggestions on the design side (which can be fixed very easily):
- The new header is bold, clear and appealing, but definitely too big in height, at least in the "portal" page. Two easy fixes for that:
1) Remove the blue bar below the navigation icons (div#headershortcutlinks) and make it appear only when one of the icons is clicked. Voilà, 78 pixels saved.
2) Remove the white bar with the social media icons and the search function. Both can easily go in the rightmost column, and in the case of social media Flickr and Twitter are actually there already.
This will ensure that users on a small scren actually get to see some content without the need to scroll down. A quick mock up:
http://imageshack.us/photo/my-images/710/europeanparliament20111.jpg/
- Why fixed width? A flexible, "liquid" layout for the main content would have been much better. Statistics show that in January 2011 85% of web users have a screen resolution higher than 1024px wide. You are wasting a lot of valuable screen real estate by limiting the width of the main content to 927px.
Other minor points:
- Integrate RSS feeds in the "head" section of the html code. This will activate the RSS button in modern browsers such as Firefox and Safari.
- Providing a PDF version of news articles is great. Providing a print CSS would be even better.
Thanks for all these comments everyone. I'm afraid I'm on the road right now and rushing between meetings, so I can't go into too many details in responding.
However a quick general response. Obviously, the launch did not go as smoothly as we would have hoped. Though everything worked beautifully in pre-production, the publication of some parts of the site ran into server infrastructure problems which caused our IT people quite a few headaches. Still, by the end of day one they had everything up and running. I gather that there have been new issues with the committee and delegation sections today, but I have just seen a message that these are now OK again.
What can I say? I'm not a technical guy, so I can't comment very easily on any of this, but my hunch is that some local difficulties are not unusual in an operation on this scale and that, once sorted, these problems will be quickly forgotten.
In the longer run, it is the observations about the site content that matter more. The first point to make here is that, as I tried to point out in the post, this redesign was essentially done on an "as is" basis in terms of the content, with the exceptions I mentioned in the post. Several of the observations actually reflect pre-existing situations that it was not possible to address as part of this exercise. Ron, by the way, the order of the committees is a protocol order the precise origins of which merit further research. Old timers like me remember when the number of the committee in the list was widely used, and when, for example, the Budgetary Control Committee was meteorically "promoted" from 14th (if memory serves) to 4th overnight. I'm pretty sure this order is related to the distribution of committee offices negotiated between the political groups at the beginning of a mandate.
But I digress. The other thing to say, as I said in the post, is that an operation of this sort is always going to throw up a range of bugs and errors across twenty-two language versions. There is a planned process for correcting these, and for correcting the things that simply don't work in practice quite the way you envisaged them beforehand. Here, it is important to spot as many of them as quickly as possible, and the reactions to this post have been fed into this, so thanks. On the legislative observatory, that was actually a separate project on an independent application the timing of which happened to coincide with this one. There will be a corection loop process for that too, no doubt.
On user testing, Mathew, yes there was a lot of that, using numerous different profiles (off the top of my head, 30 or so) There has also been a quarterly satisfaction survey for a year now which gave us a lot of information, plus numerous sessions with experts ("masterclasses with web gurus", if you prefer). So no, we didn't just stick a finger in the air. That said, pace the purists, we live in a real institutional world where there have to be some trade-offs between the needs of umpteen user groups and stakeholders we serve and some pragmatism about the timetable for technological innovation. On the latter point, this is why we decided to go for a process of redesign followed by progressive technological and editorial progression, rather than a massive relaunch (in 2025?) which would resolve every issue with the old site at one fell swoop and realise our full vision for the site overnight. If only…
Let me conclude by saying that of course I personally would have loved this whole operation to be 100% glitch-free and to have received nothing but unadulterated praise for every aspect of the upgrade. But this is the real world… That said, I for one am convinced that, inevitable teething problems aside, this is already a huge improvement over what we had, and gives us a basis to do many great things from now on. My overwhelming impression from the feedback so far is that outsiders seem to love it and insiders actually also, but just that they need to get used to it and see their issues addressed. Give us and it a bit of time, guys.
I don't know what's a 502 proxy error but you guys surely love it. I'm not reluctant to change, usually welcome it and don't have any issue with adaptation, but launching such a "refonte" should be done when you're sure that everything works. Or at least most things. Or at least a few things past the front page?
Getting back access to committees would be nice, hope it will happen before the end of the year…
good luck, for now all I can think of is "FAIL" in big red letters.
Wow, that's a big 'ole header, huh? ;-)
I was never a major user of the old site so don't have major headaches with the new one, but I am curious about the process behind the redesign.
Did you carry out a lot of statistical analyses on the old site before embarking on the redesign? If so, were you able to look at user paths, or just the basic metrics like views, users and bounce rates?
How about user tests? If so, did you use personas to segment your users, with different scenarios for different personas? Did you bring users into the taxonomical development, through e.g., card sorting?
What did all of those exercises tell you, and how did that lead to the new structure and design?
Finally, any plans to do post-relaunch testing, for example, or do your systems not allow it?
Ask a lot of questions, huh. Sorry about that – just curious. Feel free to ignore.
I wrote a blog post yesterday and tried to find relevant information and documents, but I found the way that some things are organised quite – and I excuse for my harshness – crappy.
A) There are some more useful things on MEPs' pages now indeed, but as I saw here it is totally unlogical to have the oldest parliamentary activities on top and newest at the bottom. And nobody outside Brussels knows when the "5th parliamentary term" was. It's also unlogical (and that didn't change from the previous version) that there is no link to "draft reports" and "draft opinions" or (which is impossible to find generally) "amendments" tabled by an MEP.
B) The design of the legislative observatory is awful because the static header of the procedure page covers almost 50% of the page on my laptop monitor. This means you can't really get an overview over a legislative procedure. but you have to scroll through it with just a miniscule extract of the full procedure frame. I also find the document search quite confusing, but that hasn't changed from the previous design.
C) Going to the website and finding documents is generally confusing. Why not have a clear link to "the document register" instead of hiding this behind "Legislative Observatory" under the "others" menu on top of the page? (Today, the Observatory seems to be down; the link is not even there.)
D) As on the old website, the order of the list of committees doesn't make any sense.
E) I know this has been replaced by the document search on each committee page, but I found the link list on the left we had on the former committee pages (e.g. to amendments, draft reports etc.) quite handy. Needs an additional click now. And it's also not very handy that on my laptop screen the search results are presented so that once I've clicked "Search" you don't see the results but still only the search form, for the results you need to scoll down (but it first feels like nothing happened).
I'm aware that these are just personal views (and I'm sure I'll find more such things over the course of the coming weeks), and I'm also aware that each end every use has different demands to the page and that it's difficult to satisfy als the personal demands. But I think at least 2-3 of the points mentioned above are simply illogical or complicating things, favouring the self-presentation of the Parliament and its news and don't help to make the actual work – draft reports, amendments, voting records, legislative procedures – more accessible and transparent.