Location: Home » UXblog » Thoughts on CLF 3.0 From Outside the Firewall...

21.03.2010: Thoughts on CLF 3.0 From Outside the Firewall...

Category: standards     Posted by: cornelius     Discuss: view comments     Views: 33031

Over the last few years, the Government of Canada has done its best to attempt to control the look and feel of its internal and external web properties by instituting the first two versions of the Common Look and Feel (CLF) Guidelines and mandating that every department and agency adheres to them by certain specific dates. And while the current version (CLF 2.0) is a definite improvement over the original version of CLF in terms of accessibility, coding standards and visual interface features, ask any user experience designer out there (and by out there I mean outside of GoC since I'm outside the firewall and therefore have no idea about the CLF pulse on the inside) and they will tell you that it is still nowhere near a modern web standard.

I mean no disrespect to the guys at TBS, but after a couple of iterations, it's time to get this thing on the right path. I think the first two versions have proven one thing: that creating the visual blueprint of our government's web presence should NOT be left (entirely) up to internal organizations. I can't even begin to guess the total amount of taxpayer money spent on creating the CLF specifications, and the even more obscene amount spent to port government web content from CLF 1.0 to CLF 2.0. And while there are a handful of departments (e.g. Service Canada) that have done a nice job of designing within and around the constraints of CLF, most of the Government of Canada sites (including the TBS site) are still a visual eyesore. And it's not just TBS' fault, it's the individual departments' complete ignorance of color theory and their inability to select design shops (because let's face it, most of them hire external consulting firms to do the web deed for them) that actually understand how to design a professionally pleasant website in the 21st century.

I could go on an on about why CLF 2.0 is bad, but that's not the point of this post. What I'm going to do is attempt to give you my thoughts on what would make CLF 3.0 a much more successful endeavour, from both the GoC perspective, as well as the taxpayers'. Feel free to download the template (MS Word document, 100Kb) that I created for the purpose of this discussion (you can also do so by clicking on the image above). The template is distributed under a Creative Commons (CC) Attribution Share-Alike license.

1. Crowdsource the visual design template (look and feel)

The first thing I want to talk about is CLF authoring. So far, CLF authoring was entirely up to TBS. These days, the online lanscape is overtaken by the social web, and it's time for the GoC to use social tools to get a design that we all enjoy. Yes, I do believe the CLF 3.0 visual design should be crowdsourced. And while TBS will still maintain decision power of the final look, I think GoC should create a competition (you know, with cash prizes and everything, not that every designer wouldn't do it for free simply for the exposure that comes with CLF authoring) and a judges panel that will declare finalists. The public should also be allowed to vote in their preferences, with prizes going to the panel winner, the winner of the popular vote, behind and outside the firewall winners, best student submissions etc. And in the process of (hopefully) making the right decition, TBS could even combine visual and content features from multiple submissions. This way, just like in any other user-centered design methodology, users will have an input, and a say in what they will use. More details about this top can now be found in my next blog post, CLF 3.0 Crowdsourcing: A Public Traction Pill for OpenGov Initiatives

2. Provide departments with color pallettes

I am personally not a big fan of rainbow-looking webpages when it comes to visual and web design. I'm also fairly confident that most people's understanding of a professional look also includes a notion of using a restricted color-matched palette. As a result, I believe the mandate of CLF should also include providing each department (after consulting with the department of course) with a primary and secondary color palette that should be used on their web properties. This will eliminate much of the current rainbow feel of most government websites and should improve the overall impact of the CLF design. Another option here could be to create a (relatively large) number of color palettes and the departments could choose on their own which palette they will use.

3. Use a common visual header for all GoC web properties

This statement may seem redundant as the first two CLF standards have used common headers since their inception. What I mean is not simply using the same structure and standardized dimensions, but having the same visual header, with the only difference being the data and the hyperlinked destinations. This will not only eliminate the possibility of butchering header graphics (and this is a huge problem of CLF 1.0 and 2.0), but will keep the header simple and professional. An interesting starting point towards that is provided by Tom Bradley, in his blog post, A Proposal for CLF 3.0. I personally think there is still a lot of room for improvement there, but the idea is on the right track. Simpler, minimalistic, ample use of white space, less screen real estate, practically a lot of the principles of usable design. My own thoughts on the header are that it should be some kind of white or black-based gradient background, with the elements flowing left to right, with an integrated search bar. Again, if you haven't done it yet, feel free to download the accompanying high fidelity wireframe (in MS Word 2003 format). The visual concept is still in its early infancy but should still do the trick in illustrating some of these concepts.

4. Mandate the use of RSS and social media

We are at the point where every department should be able to (and some already do) syndicate news, press releases etc. Since every government department and agency has a communications department that prepares press releases on a regular basis, I think it's time Canadians have a choice to follow the news coming out of the departments that they are interested in without having to always browse to the department site and find the press releases page. This will reduce traffic on the department websites (which is a good thing) for those who are only looking to see what that department is up to. Also (and i know this is a bit of a contentious topic), every department should have at least an unofficial Twitter account (and some already have official ones) that will broadcast whenever a new press release, tool or item is published to the general public. That way, not only can we subscribe to the RSS feeds, but we can be alerted via Twitter in real time when new material or web tools are published. Some of the departments are already doing this, so good for them to embrace social media with the rest of the world. And one last point on this, the link to the page containing the RSS feeds should be included in the common header so we don't have to look for it or have to rely on the underperforming GoC search engine. At that point, a directory of GoC feeds and social media accounts would also make a lot of sense, but that's just wishful thinking now :O)

5. Move to a primary horizontal navigation system at the top

Let's face it, the CLF 1.0 and 2.0 experiment of using the leftmost vertical column for primary, secondary, tertiary and n-tiary navigation has been a disaster. Trying to find information on most GoC websites makes it a tall order even for those of us that practically live on the web. The current navigational model is not working, and it's time to move to a navigational paradigm that is more flexible. As shown in my proposal, I would suggest including a horizontal bar containing primary and secondary navigation, right under the common header. The advantage of this, is that the height of the combined header and navigation bar should still be less than the height of the current CLF 2.0 header (in my example, the total height is 120px), and therefore would allow more white space and real estate for the main content area of the page templates.

6. Make the left navigation column optional

For those sections of departmental websites that require additional navigation, the left column of the main content area should still be used display additional deep links. But before jumping on doing this with everything, please do me a favour: when working on the IA of a government department website, hire a COMPETENT IA (information architect), because this is the most important aspect of your site that you will have control over.

7. Use a real common footer

Unlike the header, which uses different data for each government department, the current CLF footers are identical. However, they are largely useless as they provide very little information that modern web citizens expect to find. My submission contains a CLF footer that is based on the same visual design theme as the CLF header, and provides a lot more information. It includes four data columns, with the leftmost column reproducing a greyscale version of the Government of Canada logo, the copyright information and the "Return to Top" link. The next column contains links to all the important notices, from privacy and terms of use, to disclaimers and accessibility notices. The third column contains the contact information of the current department, and the last column will display all of the social media links that this department is in charge of.

8. Accessibility

CLF 3.0 should move at least to WCAG 2.0. Simple as that. Flash and Silverlight should also be allowed as long as transcripts or description of the Flash animations are provided, just like it is the case today. But unlike CLF 2.0, I believe TBS should provide templated accessible components to be used in the development of GoC sites to ensure that department don't interpret these requirements loosely. Such components would include sliding panels, advanced HTML controls, Flash/XML wrappers, etc.

9. Main content area

As this article is getting rather large, I will restrain for getting further into the main content area for the time being (maybe a future blog post?). The information architecture of the main content area should be left to the devices of each department, and the only thing that should be mandated in the overall layout is its width, which by the time we get to CLF 3.0, should cater to screen resolutions of at least 1024x768 pixels (my template is 980 pixels wide to account for various browser borders). If departments choose to use 1, 2, 3 or more columns of content, it should be up to them. There will also be a lot of freedom in how each departmental website will be composed, and the relatively large size of the main content area should allow for some interesting combinations of photographic backgrounds and (accessible) rich internet and web 2.0 features.

What do you guys think of something like this? Is it too restrictive? or too loose? Is it better than CLF 2.0? Or are these simply dreams from outside the firewall?

9 votes
1 2 3 4 5
Tags: , , , ,
» Share article on:

steven (22.03.2010 10:51:40) wrote:
I like the idea of crowdsourcing CLF, the government should definitely look into using the concept, even for other activities. I don't think anyone is madly in love with the current CLF design, and it's a shame that there are no avenues to kickstart a change.
Mark (22.03.2010 11:28:19) wrote:
Love the wireframe and the breakdown.
My 2 cents:

- set the px widths a minimums and use % for a fluid layout
- change RSS to "feed" or "subscribe"
- your initial design is simple, clean, and effective = very well done

On colour palettes...
Why give the departments a colour palette at all? Isn't the point of the common look and feel to make all GoC properties feel like 1 big property? Traditionally departments have not executed well (as noted by your comment of headers). There are too many departments to assign a colour to each.
Unify the palette.

One point you didn't address is the sad state of GoC URLs. CLF dictated that URLs needed to be bilingual. This lead to a bunch of URLs that are acronym heavy. v3 should mandate understandable unilingual URLs. Yes more complicated to manage on the backend but we already have 2 sites for every site (EN and FR) why not 2 URL sets.
cornelius (22.03.2010 13:01:26) wrote:
Steven, Mark, thank you for the comments. Mark, the state of the URLs completely escaped me... I absolutely agree with you, understandable unilingual URLs are the way to go, most departments are now using digits as identifiers since they don't quite understand the standard as it is...

As far as color palettes, I actually like the idea of the main content area being based on different palettes (and i would use vector graphics or photographic backgrounds that are wider than 980px with this to maximize impact). But letting the departments make the decision on their own has been a complete disaster in the past so that's what i would like someone to make 'informed' decisions for them. If there are two many departments, at least create 10-15 palettes for them to use, it's not really a difficult undertaking to come up with them... There are only so many primary colors out there :O)

Fluid layouts with min pixels with are a possibility, the reason I would still stick with pixels is because i'd like to achieve the desired effect with the right background combination (see Canada Health Infoway website as an example).

Also like the idea of changing RSS to Feeds, RSS is too technical and Feeds would be much more descriptive. Like I said, if TBS would ask for help, I'm sure a lot of designers in the know would be able to steer them on the right path with this endeavour...
Everett Zufelt (22.03.2010 16:09:19) wrote:
8. Accessibility
CLF 3.0 should move at least to WCAG 2.0. Simple as that. Flash and Silverlight should also be allowed as long as transcripts or description of the Flash
animations are provided, just like it is the case today. But unlike CLF 2.0, I believe TBS should provide templated accessible components to be used in
the development of GoC sites to ensure that department don't interpret these requirements loosely. Such components would include sliding panels, advanced
HTML controls, Flash/XML wrappers, etc.

* I agree that CLF 3.0 should adopt WCAG 2.0, using priority AA as the standard. I think that the adoption of ATAG (Authoring Tool Accessibility Guidelines) would also be a good idea, even if only a subset of the ATAG success criteria is adopted.

I really like the idea of the provision of accessible complex UI controls (widgets). I would warn against using ARIA 1.0 as the technology to ensure that widgets are accessible, but it would be good to implement ARIA for progressive enhancement.

As for the use of Silverlight, Flash, and related technologies, I agree that where these are used soly for multimedia presentation that an accessible alternative must be required. However, it is also necessary to make sure that any UI controls embedded in the objects be available natively through xhtml / JS. Silverlight and Flash objects are not accessible to some assistive technology users, particularly those on non-Windows OSs.
Steve Buell (22.03.2010 16:19:17) wrote:
CLF 2 is being updated. http://www.tbs-sct.gc.ca/cl...
The CLF 2.0 standard will be updated to:

* take into account the most recent version of internationally accepted Web content accessibility guidelines;
* increase flexibility of Website layout and design;
* enable institutions to incorporate the use of innovative and emerging technologies to their online information and services.
cornelius (22.03.2010 16:49:54) wrote:
Everett, thank you for the accessibility insight. Regarding the use of accessible alternatives for UI controls, well, my suggestion would be to ensure that the UI controls do not reveal content that is not accessible via any other mechanism. This is why I proposed the use of a sitemap link in the main nav bar, so if a visually impaired visitor decides that a sitemap is the best navigation alternative, they can get to the content fairly easily.

As far as WAI-ARIA, I know it's in draft, but I haven't had a chance to look closely at it, so I will defer to you on this as you seem to have much more experience with it. I would assume that by the time CLF 3.0 or the revision to CLF 2 hits GoC there will be an established standard for RIAs that can be followed...

Steve, I know CLF 2.0 is being revised, this is largely the reason for this post (the fact that we're a little afraid of what's gonna come out of it). Since GoC has not stated that it's moving to CLF 3, I'm assuming the changes will be simply WCAG 2.0 adherence, guidelines around RIA integration, and some more flexible visual guidelines for the main content area, but I doubt they will seriously address the issues in the header, footer and overall layout flexibility issues. All I can do is assume the worst since the TBS statement is overly vague. Again, crowdsourcing anyone? :0)
Everett Zufelt (22.03.2010 17:03:45) wrote:
@cornelius It is true that ARIA 1.0 is currently a W3C draft recommendation. Some browsers and assistive technologies have begun to implement parts of the recommendation already. The pitfall of relying on ARIA to convey accessibility information, even after it is recommended, is that many users will continue to use non-compliant browsers and assistive technology.

There are clearly some complex UI widgets that cannot be made accessible without ARIA (if this wasn't the case I expect ARIA would not exist). However, ARIA should be relied upon as a last resort to make UIs accessible. ARIA should be used to progressively enhance otherwise accessible UIs. And, it hopefully goes without saying that the information and interaction provided through any UI that requires ARIA for accessibility must be made available through an alternative interface.
Thomas J Bradley (23.03.2010 13:04:42) wrote:
It seems we had the same sort of idea for our CLF3 proposals.

You had a really interesting point in that many departments' colour palettes could use a lot of work--I really like the idea of providing sample colour palettes.

I hadn't really thought about a footer in my proposal, but it does make sense. As long as it is used for more information as in your proposal.
cornelius (23.03.2010 15:52:23) wrote:
@thomas They are both good starting points of discussion, and just based on the comments I've received here and on Twitter I'm sure they can be improved a great deal. There are a lot of passionate individuals who could potentially drive the conversation and the creation of templates (static HTML, Drupal, etc.), but at the end of the day, GoC makes all the decisions...
Mike Gifford (23.03.2010 16:08:26) wrote:
Great post @cornelius. Love that you've made the wireframe available under a creative commons license for others to remix.

I've posted a related article to my blog - http://openconcept.ca/blog/...
Steve Buell (23.03.2010 16:59:32) wrote:
Open consultation falls under POR (public opinion research) and all that entails: approvals, briefings, notices, etc.
I work in the public service and see how change happens. I can understand the "assume the worst" perception. That way you're not surprised. I know there are some really smart people working on this and they are aware of what is going on in the community. Keep the discussions going.
cornelius (23.03.2010 17:13:21) wrote:
@steve But you see, this is the problem... The 'being aware' but not engaging. 'Approvals, briefings, notices' are not two way conversation. If TBS is cooking something so good, why not engage the community and get true feedback out of this... Why choose dry process over critical thought?
Everett Zufelt (23.03.2010 18:03:18) wrote:
Another thought about WCAG 2.0.

As some may know WCAG 1.0 (required under CLF 2.0) is technology centric, namely html. WCAG 2.0 is technology agnostic, it can be applied to any accessibility supporting technology. One problem with this within a large organization like GoC is that developers will need to be trusted to make the determination of which technologies support accessibility. It is important not only to consider the accessibility features inherent in a particular technology, but how readily the accessibility information can be accessed by ATs on different operating systems.

As an example, Adobe Flash supports an accessibility API, which can allow some properly designed Flash objects to conform to WCAG 2.0. However, this API is only exposed to AT through the Flash player on the Windows operating system, which in my opinion means that Flash technology should not be considered to support accessibility for the purposes of WCAG 2.0. PDF is another media type that supports accessibility. However, PDF forms, as far as I'm aware, can only be accessed by screen-readers on Windows, again I would say that PDFs (at least PDFs with form controls) should not be considered to support accessibility for the purposes of WCAG 2.0.

I use Flash and PDF here not to pick on Adobe, but because these are very popular media types, and their accessibility issues are often poorly understood.
Jason Kiss (23.03.2010 18:28:57) wrote:
I want to second Everett's comments on WCAG 2.0 and what should be considered accessibility supported technology in the next version of CLF. To his examples of Flash and PDF, I would also add JavaScript. While JavaScript works well with many assistive technologies, and can even improve accessibility, government info and services need to be as accessible to as many people (and so devices) as possible, and you can never know what combination of software, hardware, and preferences someone will be using when they visit. This doesn't mean that CLF would prohibit Flash, PDF, or JavaScript, but only that these technologies couldn't be relied on to be fully accessible for all users.

New Zealand may serve as an example. It was, I believe, the first government to go with WCAG 2.0, and doesn't consider Flash, PDF, and JavaScript to be accessibility supported in the context of government sites, with some minor provisions: http://www.webstandards.gov...
Steve Buell (23.03.2010 19:37:18) wrote:
And now we arrive at the crux of the matter. I, as a public servant, have limitations on my public engagement. Anything I participate in has to bear public scrutiny, 'Approvals, briefings, notices' are required to formally engage in any public consultation. Informal participation in public discussions still requires disclosure of my status as a member of the public service and has to follow certain guidance.
Do I agree in this day of interconnected communications? Not completely. Do I respect these limitations? Most certainly!
At some point (crossing my fingers and hoping) the limitations on members of the Public Service engaging in on-line forums as individuals will be freed from attribution to their departments or the GC as a whole and not subjected to things like litigation.
This is why dry process may still reign over critical thought.
Could this comment draw some internal critique? Probably.
cornelius (23.03.2010 23:12:14) wrote:
@steve It would be a shame if your individual opinion on twitter/blogs is taken as some kind of indiscretion by the official powers. The presence and voice of various GoC employees on either side of the fence is the exact reason why we all have hope that at some point the virtual web presence of our Canadian Gov't will be recognized as a leader in all areas of open government. That being said, when I noted that GoC should engage in a conversation, I didn't mean informally, individually or unofficially. I find it so 'public service' when GoC departments have mediatized policy hearings, dedicated feedback sites for various initiatives, etc. but when it comes to CLF, they create a one pager that tells taxpayers 'we are working on something, and trust us it's good... but we won't tell you exactly what it is, and it's gonna take us two fiscal years to do it'. Call me cynical, but after two butchered CLF versions, I won't take their word that this time they're creating CLF Nirvana.

If TBS is not interested in crowdsourcing (in which case i'm assuming the reason is the number of people employed for the next two fiscal years whose mandate is to come up with CLF 2.x-3), why not at least create a public hearing when both internal and external users give feedback on the design? And mediatize it so we all know about it? I'm not the first, and probably not the last person who's openly criticizing the standard, yet TBS/GoC always seem to find a way to choose dry internal process over common sense...
Wendy (24.03.2010 09:53:46) wrote:
I'm interested in seeing more discussion on the PDF accessibility issue as our GoC website posts publications regularly in both PDF and HTML formats.

For consideration, here is the Accessibility Support Documentation for PDF on the W3C site: http://www.w3.org/WAI/GL/WC...

Test file results are listed at http://www.w3.org/WAI/GL/WC...
Chuck Henry (24.03.2010 10:40:23) wrote:
I am a former public servant, and owned CLF for about a year. I believe TBS (CIOB) is doing a good job of broad internal, i.e. across all departments consultation. It started before I owned CLF and I did my best to encourage and expand it. The issue of external consultation is thornier. TBS is trying, but the reality of external consultation is, as Steve explained, isn't controlled by TBS (S standing for Secretariat, i.e the bureaucracy that supports TB Ministers). External consultation is a ministerial decision, so write to your MP and the PM about what you'd like to see. In the mean time, TBS CIOB is following the process to do external consultation, time will tell if they are successful in getting permission. In the meantime they are doing their best to follow external conversations, as Steve has (bravely) demonstrated.
PS - Jason nice to hear from you!
cornelius (24.03.2010 16:02:11) wrote:
Chuck, Thank you for the comment.
While working at Deloitte, I was on a TBS project that required cross-departmental consultation. And while I know that TBS was well intended to listen and incorporate cross-departmental feedback, the actual process was a bureaucratic formality. Decision makers in other departments simply don't care about look and feel, usability or standardization (in this case we were working on a cross departmental financial application)... We had meetings scheduled at PW for example, and they would cancel, skip or send their junior resources who right off the bat would say 'I have no authority, I don't really understand the context and I can't give any feedback'. Aside from the insight of 2-3 people from other departments, we ended up creating whatever TBS felt was right. Yet, the TBS press release and website debrief praised the process and the 'extensive consultation' that took place. So again, call me skeptical...

It seems like whoever is or was involved with CLF is content to say 'we have a great group, they are very capable people, etc.', and they've been saying that for years, but aside from real progress on the accessibility side, the look and feel was and still is exceptionally stale. You're also noting that external consultation is a 'ministerial decision'. Seems like something I would read about the Vatican. Let me paraphrase: those same people who have made the bad 'ministerial decisions' with respect to CLF in the past, are about to make the same bad decisions the third time around. Now I'm relieved.

The last couple of weeks I was out for beers with a couple of current and former EX level personnel. They all told me that they are aware of the personnel issues, but there's absolutely nothing they can do... Like you said Chuck, you 'owned' CLF, and the only feed back you can give to this post is to write to my MP and the PM. Indeed, two very qualified design professionals with a keen eye for esthetic detail... Luckily we are now (hopefully) dealing with a new generation of public servants who seem to be more open and proactive about improving things. And by new generation I definitely don't mean my MP or Stevie Harper.
Everett Zufelt (24.03.2010 16:39:28) wrote:

I would love to see examples of the "real progress on the accessibility side". Some GoC services (epass) are still incredibly poorly accessible and other sites are just poorly accessible. I wonder if any degree of accessibility understanding is required by developers hired for GoC departments.

I mediate the above comment by saying that some sites, or pages of sites, are quite reasonably accessible. But, it isn't the millions of reasonably accessible pages that matter, it is the one inaccessible page that a citizen wishes to access that matters. I really don't understand how epass ever was allowed to be acquired with such poor accessibility. Granted, I don't know for how long epass has been around, and perhaps there were no CLF accessibility requirments at that time. But, surely epass doesn't predate WCAG 1.0, over a decade old.
cornelius (24.03.2010 17:07:48) wrote:
@everett Unlike the CLF 1 and 2 visual standards (look and feel) which in my opinion were sad since inception, I believe CLF 2 made an effort to mandate accessibility across GoC. But my comment is only related to the standard, and definitely does not apply to the implementation. Do I think the standard still needs work, absolutely, it has fallen behind the times as you guys passionately pointed out, but when it was originally released, it was a clear improvement over CLF1, unlike the visual standard which was marginal.

I've worked in a UX capacity with development teams in GoC and I've been quickly shut down both internally and externally when I've even implied that we should actually test websites with screen readers or hire someone who really knows what's going on on the accessibility front because we are dealing with high-profile, public facing websites. The standard today still seems to be 'accessibility by checklist'. If I had my way (which would never happen :O), I'd get well respected accessibility peeps (someone like @feather, no offense to you guys but he's my a11y hero) to lead the effort on GoC CLF accessibility standardization and testing...
Laura (24.03.2010 21:02:41) wrote:
@cornelius First of all, let me say that I fracking love the template. Ok, that's my opinion, so let me back it up with some facts from usability testing that I have seen performed on gov sites. Oh yah, I also love the crowdsourcing idea!

1. In one test I was involved with in the past year, every person tested looked to the bottom of the page for the contact information and not one person clicked on the top header bar where it says "Contact Us". In fact, anything I've ever seen tested on the top nav bar has been ignored. Good idea to scrap it, at least, in it's current iteration.

2. In another test I was involved in, people couldn't figure out why the "important notices" was not at all important. I like the way you've listed out the various disclaimers etc and linked to them from every page, especially that obnoxious PDF warning.

3. Haven't seen a test yet where people didn't look at the top right'ish for the search bar.

As for opinions, these are mine after nearly 10 years as a public servant working in gov web shops:
1. I don't think Twitter should be mandatory as bad use is just as bad (or worse) than no use. There are a number of people working on Twitter use policies and best practices so I believe that in good time most of us will be on it.
2. Totally agree re: RSS. I would love to see RSS replace eNewsletters as the communication feed of choice.
3. Totally agree that W3C guidelines should be the de facto accessibility standard but defer to @Everett & @Mike as both of them have proven themselves to be experts in this field.
4. Love the idea of the colour palette, would like to see more in the way of examples of good style guides that talk about more than just colours, but tone and style too. Not sure that it's a good idea to make any of that part of CLF.
5. Would love to see TB provide a centralized place for all dept sites to link to the common notices if they choose.
6. We have made progress! point final!

Couple of things to consider:
It would be neat to see an annotated version of your mock-up linking each design element to best practices/usability tests. I'd add stuff on it.

Why do we need official consultations? Sounds time consuming and costly. The great thing about social media is that anyone and everyone can have their say and others can choose to listen/read respond or not. I volunteer to add a link to your blog post on the gcpedia page to redirect traffic here should those involved be so inclined.

@Steve I volunteer to help on the usability testing of CLF3 but only if we do it before it becomes a government-wide directive. ;) Let me check with my boss to see how much time I can dedicate but I'm sure I can make the pitch if you're interested.

And finally, I just want to say that as much as I appreciate this forum and the venue for discussions, as public servants we do have a responsibility to our employer and must be respectful of our conditions of employment. That is true for anyone who works anywhere. So, whether we agree with everything or not, we've chosen where we work and we've got to abide by the directives laid out for us, while at the same time using creativity and common sense to guide us in the grey areas. I say that because I think you are misinterpreting some of the comments above to mean something that they don't and have responded with ageist & biased responses. Happy to fill you in.
cornelius part 1 (25.03.2010 00:07:05) wrote:
@Laura So happy that you've joined the 'conversation'. I'll start by addressing your last point. Believe me, I understand the politics. But for this post to work, I had to go big or go home (isn't that @nickcharney's new #W2P slogan? :O)... Or a little 'ageist' / biased if you'd like... I had to get enough people who care (and I am secretly aware that for the first time there are currently more people who care about this inside the firewall than outside, yay!) to start an open conversation about a very touchy subject. Being nice and polite wouldn't have worked. People would've read this and moved on. You of all people should understand, your own blog post is titled "Everybody Hates CLF"... I know, sorry, I had to bring it up :O).

That being said, apart from the #a11y guys who are doing a great job of sharing information and ideas across various channels, myself, @thomasjbradley, yourself (Laura) and the other tweeps who have commented on this post are amongst the few that are openly sharing anything (i'm talking about ideas, usability data or visual designs) related to the CLF user experience (or lack thereof). No disrespect to the guys at DigitalOttawa, but they built commonlookandfeel.ca and that went nowhere, no new material in months and the material there is not persuasive (or edgy) enough to actually start meaningful conversations... On twitter they are all saying they have some great CLF 3.0 ideas, but other than Thomas Bradley's post, no one has posted anything... What are they waiting for? CLF 3.0 is on deck, they want the public service to be open about it, but they won't divulge their groundbreaking CLF sketches? (DigitalOttawa, pls don't hate me, I actually like your work :O).
So I decided to put my money where my mouth is (and I am aware my mouth happens to be quite large this time of year since I don't currently have a full-time job and I spend most of my time doodling and learning a couple of new things about the business of UX) and actually spent half a day or so to do some UX research and put together a CLF 3.0 template that could potentially work... And it seems to have had at least a little bit of impact.

And I too have other ideas, a lot of them stemming from discussions I'm having online and offline with people that are reading this thread. Here's one that I think it's also pretty rad. I believe the header and footer should not only have the same style, but GoC should have one single CSS file for the header, footer and layout container that is common across all government sites, hosted in one single location on the root of the gc.ca server. That way, when CLF 4, 5, etc. rolls around, you tweak the common CSS and all the government sites are there immediately. No need to pay exorbitant fees to consulting companies to write the same lame CSS files over and over for you. The main content areas can have their own departmental CSS files and that way, you can't blame programmers for butchering the most identifiable part of the Government's common look and feel.

You asked about color palettes, those require some real work. To create them you need an understanding of what colors are used on gov't websites and with what frequencies. What are the typical primary/secondary colors? Want to standardize across based on a greyscale structure? And based on this data sift through and create professional style guides (not just dumb #Hex lists but actual usage scenarios that can be audited that include tones, blending, negatives, etc.) And I have other ideas about GoC social media widgets standardization, CSS background selection, common (accessible) component design etc... Maybe another blog post. (cont'd in next comment)
cornelius part 2 (25.03.2010 00:07:44) wrote:
(cont'd from prev comment): And to conclude this rather long post, again, I really hope you, Steve, Chuck etc. understand that a lot of my online tone is a factor of the medium, and yes, I planted my feet firmly within 'outside the firewall' territory and purposely misinterpreted Steve's MP and PM comment. Like I said from the beginning, Steve's accessibility work is awright by my book (pretty sure Everett might disagree on this tho :o) but a11y is not my main specialty or interest for that matter, and I leave it to the experts in that field. I'm a user experience professional (in the end-user research, interaction, information architecture, design and UT sense of the word), and part of my professional mandate is to make usable products. There are some comments on some of my old gov usability evaluations (haven't done GoC ones in a while as I worked large private and public sector projects in NY and LA for the past few years) that make my comments here seem like bedtime stories. I had to persuade people to think about design not just code. I had to make it clear that regardless of the backend, all people on the street can see are front ends that are pointlessly convoluted and are based on incomprehensible IA paradigms. Even my CLF 1.0 designs from 2005 were more modern and progressive than the CLF 2.0 stuff that was subsequently released. So now that I am back in town, I've decided to restart my half-decade-long beef with CLF in the name of… I don’t know, social conscience? Again, if I offended anyone with my comments, I truly apologize, but I still stand firmly behind what I think. I may be young(ish), but I led a Big 5 UX practice in the past, and was successful on some pretty crazy public sector UX projects so I've learned my lesson to speak up before things go well beyond the boiling point :O)

If you guys don't think there's any value in what I have to say (forget how I say it, this is not an official forum, and I don't work for GoC to be bound by the same ethical dilemmas that some of you face, with the caveat I obviously exhibit a different tone when acting on behalf on an employer/client), you can stop reading, commenting, and paying attention to this thread. Or, here's an idea, let's find a way to work together on this this CLF 2.x-3 UX/design/crowdsourcing geschäft, I promise I'm way cheaper that Deloitte :O) Oh, and I also promise I'll play nice. Most of the time.

PS @Laura: the overlap between my design and your UT data is unreal. And i don't even know you (yet).
Laura (25.03.2010 10:44:35) wrote:
No offence taken here Cornelius.

Thought of 2 more things you might want to incorporate into your mock-up.
What's your proposed approach for multi-lingual sites? I have worked on gov sites with up to 5 languages besides English and French.
Also, knowing that client data is difficult to obtain for both policy and practical reasons, have you thought of including a "Was this page useful?" feature somewhere on there?

Personally I love the idea of having a common CSS etc. Good stuff...you've got me thinking.

P.S. I totally regret ever saying "hates" lol. I wish I had called it "CLF: everyone's a critic". #hindsight *sigh*
Everett Zufelt (25.03.2010 10:48:30) wrote:
Just to clarify, I have knowledge or no opinion one way or the other of @Steve's knowledge and expertise in the accessibility realm. :)
cornelius (25.03.2010 16:32:05) wrote:
Well, between all the peeps/tweeps that i'm meeting this week over coffee/emails/twitter DM's regarding this post, there's one theme that seems to be universal: people inside GoC are walking around with a newfound swagger and are confident that open government is well on its way...

And if the fact that I made crowdsourcing the number one thing on my beef-with-CLF list is getting a bit lost in the follow up commentary, well, I have news for you: Monday's (March 29) subsequent blog post will give you all the background info on why the Canadian Government should crowdsource CLF 3.0 and what it would mean for GoC's opengov initiative. I'll update the post and this comment to reflect that, as soon as the new article is up.
Martha (29.03.2010 20:21:38) wrote:

Congrats on kick starting what has to be one of the more stimulating and informed discussions I've seen on Government web design and accessibility.

I've managed Government websites for over 10 years. I've seen the pre and post-CLF 1.0 and lived through the migration to CLF 2.0.

The spirit of CLF is noble, respected and necessary. Our sites need to be easily identifiable as GoC entities and must certainly be 100% accessible to their users.

Beyond that, I say we open it up. Provide a suite of templates designed to meet the needs of various audiences and communication objectives. Some will require full social media integration, others may not.

The sites I manage (over 100 in 30+ languages) have multiple target audiences and visitor profiles. One design can't possible meet the needs of those audiences while at the serving those of other departments and agencies.

What's lacking to help inform CLF 3.0 and beyond? A collective understanding and agreement around our ambition for being online and the vision needed to instill a sense of effective communication through design and content that meets our users needs. Does that rest with TBS? No. But where?

Canada launched a Government Online initiative that garnered much international attention, deservedly so. Money was pumped into this initiative and resulted in a 'swagger' you now see around open gov.

And so, here we are. Continually grappling with the look and feel of our websites while overlooking their internal governance and external purpose and audience.

As Chuck and others note, TBS has done an admirable job over the years of steering the Gov't of Canada online and in keeping us in line. The primary challenge as I see it? We must find a way to bring together the web worlds (design, usability, performance measurement, content, development etc.) to take the helm.

Could we crowdsource design? Yes. Do we need to? No.
That being said, I look forward to catching up on your crowdsourcing post!! ;)

Martha (29.03.2010 20:22:27) wrote:
p.s. I'd love it if you could create a bigger comment window? It didn't lend itself easily to me proof-reading ;)
cornelius (30.03.2010 03:52:37) wrote:
@mjmclean First of all, thank you for your commentary Martha, you've hit the nail on the head with pretty much every one of your discussion points.

As far as the spirit of CLF, i'm not questioning that in any way. We need instantly recognizable, professionally designed modern government web properties that have to be accessible to 100% of end users. But GoC needs to look at this twofold: first is related to CLF standards, the second is related to educating business, web/UX and communications teams on how to work with each other on conceptualizing and constantly evolving GoC websites.

I'm glad you recognize that one design (in this case CLF) cannot meet the requirements of all government website permutations. What CLF has been unable to provide, is a framework (this would mean a main template and a manual on allowable modifications) that supports easy customization without being obtrusive. This is why I created the template in this post, it is simply a way to provide a more visually 'open' paradigm. The design allows for instant recognition as a GoC site, as well as increased flexibility, freedom and space to brand and design the site itself in the main content area. But it's just an idea, based on half a day of research and design, I make no claim that it's perfect, but I would argue that it's at least a significant improvement.

I also still think that it's not a bad idea to add social media to my design. However, because social media accounts are in the bottom right visual space, those could be replaced with other information if so desired (this all depends on TBS research on what features should be mandated across the entire GoC spectrum, based on audience).

Your international.gc.ca portfolio is impressive, and from what I've seen the sites you manage are easily some of the better conceptualized PS properties. I agree that GOL is working, and that it is responsible at least for some of the 'swagger' of the opengov folks. TBS has been (let me say partially) helpful as you've pointed out, but where I still think they failed is providing departments with the right approach on the inception and evolution of web properties. This doesn't just mean visual guidance, but also internal governance and basic user experience principles (end-user research, persona development, mandating usability testing, reacting to metrics, content strategies/monitoring content, etc). All these elements are not really optional, they are required to create successful web properties. The size of their team is one reason why this is happening. They have a (ridiculously) small team so they may have even overachieved given the bases they needed to cover so far.

Do you need to crowdsource? Again I still tend to lean heavily towards Yes (the small team size and lack of cross-disciplinary personnel is a pretty strong argument for this). Unless of course, you're contracting CLF to a private ad agency (for design) and a UX shop (for the UX/accessibility elements). Expecting all of the work to be properly done internally may be overly optimistic, and by the time it would be done in the next couple of fiscal years, it will already be well behind the times just like CLF 2.0.

Thoroughly enjoyed reading your comment and I am really happy to see that there is awareness in the higher levels of the GoC web layers concerning the things that need to be done in order to achieve successful online properties.

PS. I've increased the comment area as well, so hopefully this will make it easier for you to comment in the future :O)
kris (05.04.2010 22:25:30) wrote:
I'm behind on this thread, so my comments are coming in late. That's kinda related to my first point...

Point 1:
We need official consultations because we are the Government of Canada and represent Canadians. Conversations, postings, and information that happen on web sites such as yours must be available to anyone who asks - the other meaning of accessibility - and the information therein must be authentic - from a public servant, said on the date captured in the posting - and finally, the information must be reliable - captured in trust and produced when required, i.e. ATIP.

Unless someone knew of your site and this conversation, they would not know that this CLF3 conversation was happening. It was brought to my attention (thanks Laura!) and I am definitely interested for several reasons that I will get to, I wanted to make the point about public consultations. I do apologize for "raining on the parade" but as public servants and those representing Canadians, we have to make sure to give stakeholders a fair voice.

I know that's boring and very parade raining. That's not to say crowdsourcing isn't possible, but it's important to ensure it's done reliably, authentically and the information & discussions are available.
(I might be obvious that I used to be an archivist. :)

Point 2:
Now, as an information architect, I have worked on several departmental web overhaul projects - everything from content strategy and rewriting, to search to tagging to taxonomies to common labelling and more. On every project, there are commonalities in content, search, tags and taxonomies, even across a wide range of departments, kinds of content, and audiences.

For this reason, GOC web templates definitely need simplifying from a content strategy and content organization point of view, and these templates should to be reused during content renovations, which happen regularly (eg. CLF1 reno, CLF2 reno, re-organizations, new ministers, policy updates, etc. etc.). There are tools that exist to facilitate this, and many IAs in the GOC have been through this process several times.

There is a tremendous amount of confusion on the part of user (non public servant user) when they are looking for stuff on a GOC web site, and simple reorganization, simplification of the content menus, user-centric menus and content organization - types, formats, function, subjects for example - would go a long way in fixing this.

From an IA perspective, and from my experience, user tasks and content delivery on a GOC web site can be standardized (I probably shouldn't use that word), at least to some extent, with which the wide range of services and content and languages the GOC is engaged.

I am definitely a supporter of reuse and minimizing effort, so I pretty much agree with simplifying design. Now, if we could make all our teams all agile, that would *really* make me happy…

cornelius (06.04.2010 11:44:55) wrote:
@kris Well, when you say information has to be from a reliable source, you're talking about legislative information right? This conversation is centered around my post because there are no other avenues open for the public to voice our concerns about the CLF fiasco. I would also argue that a lot of times, GoC gives the public the politically correct washed out version of the truth, rather than the raw truth (for example, I don't see the GoC being openly critical of their own work... ever). This is a very bad thing, so that is the reason why I call out certain things that I believe need some real improvement.

Yes, everyone quotes ATIP, official communications, accessibility, but to me that is just another way of saying 'We're not creative enough to do the morally right thing and address a real problem by going around employer hoops'. And to say that information is only authentic if it 'comes from a public servant'... Well, that one I flat out disagree with. I think CLF 1 and 2 are outdated standards and have been from Day 1. The GoC can say how great it is all they want, when it comes to design and UX competence, they (at least TBS/CIOB) don't seem to have it. So whatever information comes from their camp, public servants or not, is not reliable, it's patchwork. The filter of public opinion is what would make that information reliable and GoC is not ready or able to properly open it up for discussion or for true crowdsourcing.

From an IA perspective, a process needs to be followed, no more comms teams making all the visual content organization decisions. Tagging is great, but the job of IAs should be to work on both human and machine findability. Most users don't care about metadata tagging because they don't use the search, nor do they access archives. They simply want to find information within a visual hierarchy. Which is not very intuitive these days on GoC websites, with exceptions of course.

Note: Your email address will be protected and will not be shared with anyone.