I’ve been a little quiet lately. One reason for this is that I have started a full time job where I actually have an office to go to five days a week. Aside from the rude shock of working full time after not doing so since my son was born 5 years ago, I am enjoying it immensely. I have moved into a new environment, that of local government. I am learning a lot about the structural design of Council websites. I’ve done a bit of poking around, looking at what people are doing. Generally the structure of the sites is ok but there is plenty of scope for improvement. In general they lack a strong user focus. They have not taken their IA far enough. So, even though I am not actually employed as an IA (specialist IAs are a bit of a luxury in the country), I am excited about the possibilities.

A large amount of my IA work has been done in the higher education sector. I have worked on sites with wildly different audiences and purposes but in terms of the overall goals and culture of the institution, the affect they have on the site has been roughly similar. Over the years I have developed a keen knowledge of the politics of universities, the history of their websites and how both of these things affect the sites. Generally when you are designing or even just maintaining a site structure you have to wrestle with these things daily. There is pressure to prioritise the material of influential groups. There are people who think the web is there to showcase their work, rather than as an extension of their work. There are people who insist that org chart = site structure. Of course, these problems also exist in the local government environment.

Problems like this seem to stem from people’s conception of information, their relationship to it and how they make sense of it. Most successful site structures work because they categorise and organise information according to how the audience makes sense of that information. The paths to the information, the context in which pieces of information operate combined with things like usable, predictable interfaces make a site easier (or harder depending on how well it’s done) to use. The problem seems to be with people being unable to step outside their own conception of the information and how it all fits together and into someone else’s shoes. Doing IA involves a lot of this: put all the information in a big bucket, pretend you are someone else, arrange it all in a way that makes sense. It’s the being-someone-else bit with which most people have difficulty.

Generally most Council sites I have looked at fail to get totally into the shoes of the audience. A lot of them are part way there but they still tend to organise their information according to the structure of the Council to some extent.

Councils also seem to struggle with the diversity of audiences and depth of information they require. A broad audience / deep information site is very difficult to get right. Often you have to break it up into smaller pieces and concentrate on those, even creating a set of sites rather than one big monolith. Having said that, the punter usually looks at the web presence of an organisation as one big monolith so there has to be some kind of big picture view from within the organisation. Those who are deep in the content, the content specialists, the policy makers etc, are never well-placed to see this big picture. They are very good at knowing their material very well, they’re not that good at seeing where it sits in terms of the rest of the information output of the organisation, nor how the audience interprets it. They sometimes forget that people come to websites for basic information; everyday, normal information - not deep, strategic or heavily-documented information. They usually want an answer to a question. Quickly.

So, my first challenge, once I get my head around how a Council works from the inside, is to step back outside, look at all our information and see if it makes sense for the average punter.


Six Australian universities and one in New Zealand are poised to start distributing lectures through an Australian version of iTunes U:

The participants that will offer their teaching and research free for download on iPods are Griffith University, Swinburne University of Technology, the Australian National University, the University of Melbourne, the University of NSW, the University of Western Australia and Otago University.

iTunes U interface


Although it’s not up on their website as yet, winners of the CASE Circle of Excellence awards have been named. See the results of the web section and read comments from the judges at mStonerblog.


Webmonkey has been home to one of the most referenced information architecture tutorials ever so it was with some happiness that I found out it is back online with a visual spruce up and more collaborative.

Highly recommended.

Via Jeffrey Veen.


When I did my first IA of a big university website, one of the largest hurdles was trying to convince people that the information they were responsible for providing should not always go on their department website. This was most pertinent when applied to information for prospective students. Future students are not familiar with the university, often they find the terminology intimidating or just plain foreign. They have lots of questions but they do not know where to direct them. They want answers, fast. It was not always easy to convince people that their information was better-placed on a central site. It didn’t make sense to force them fo figure out who was responsible for what, because chances are they wouldn’t bother.

That was in 2002.

Today I find the same battles are still raging. I (and countless others) have written before about the need to separate site structure from organisation structure. The structure of an organisation should not be the primary key when trying to organise your information into groups that make logical sense. Your audience needs to use your website. Your audience doesn’t understand or even care about the structure of your organisation or how it ‘works’. They want information to enable them to complete certain processes or tasks. They want access to applications that will enable them to complete these processes and tasks. Sometimes different parts of a university look after information related to the same process or task. Separating this information does not do your audience any favours and will no doubt annoy them immensely. Arranging information according to who owns it or looks after it then seems absolutely ludicrous.

Well, it seems ludicrous to me but not to some. Of course, if YOU make sense of the information output of your organisation by who looks after which bit, arranging your website accordingly makes complete sense. Unfortunately that’s not how most of your audience makes sense of your information.

I’ve often wondered why some people can’t seem to see the bleeding obvious in this case, because if you spend two minutes thinking about it, it IS bleedingly obvious.

So why do they do it? Why do they fight tooth and nail to keep information in some place where it will never be found?

One big reason: ego. Websites have always had that wonderful ability to appear to be an extension of an individual or an organisation in an almost anthropomorphic manner. I’ve said it a million times and one more time won’t hurt: to the audience a website is not just an extension, it IS the organistion or the individual.

If you work on websites this is a pretty scary thought. Your work is incredibly public and is directly related back to you. If you’ve ever received some ‘feedback’ after you’ve sent a new site live or you’ve re-modelled or made-over an existing site you’ll know all about this. This feedback can get very personal and quite irrational. You learn to disassociate yourself from the site because you have to. You can care deeply about how it works and how it is progressing but you have to learn to turn this off sometimes. You need to be able to separate the constructive and deserved criticism from the personal and sometimes almost visceral reactions. You need to care about your site to make it better but not care so much that it becomes something that you would defend with your life, beyond what others would consider rational.

Those who find it difficult to see the needs of the audience and needs of the site as a whole are struggling to separate the site from themselves. The site is an extension of themselves and an indication of their hard work. If you try to convince them to move information or functionality to a part of the site that better suits the audience they take it as an attack on their department and themselves. It must be pretty painful to live like this.

Such an attitude is really a hindrance to creating a site that is user-friendly. If you can’t separate the site from yourself, and it is, as Auden says, an “extension of (your) powers to charm”, you don’t have much chance of seeing the site from the eyes of your users.

There are ways of managing this and managing, or manipulating, is very important in a devolved web environment like that presented to you in a university. But I’ll save that for later. This post is already over 700 words long, after all. Far too long for the web…


Yesterday I was watching Twitter and:

see your organisation as a broadcaster. diverse content. getting staff to blog is like feeding kids

@georgh ha ha i so have my work but out for me. am contemplating bribes.

You’ll need more than bribes I thought. For several years I blogged at the University of Sydney. I was also involved in providing a blogging platform for staff. Later I moderated the student blog. In all cases I was disappointed by the lack of actual blogging.

So what’s the problem?

The idea of a blog always sparks up great enthusiasm in people, especially managers. A couple of years ago it was seen as an easy way to make your organisation seem abreast of the times.

“We need to do something about our website, make it a little more sexy, or something”
“We could do a blog”
“Hey yeah, that’s a great idea, we’ll get the web people to set one up tomorrow!”

(Unfortunately they discovered that technically it’s easy to set up a blog, making people write is hard).

Then there was the phase that saw it as a solution for staff communication (soon overtaken by wikis).

“We need to make sure everyone is getting the right information. I think staff need to share information a bit more. Make sure everyone has everything documented”
‘Yeah, we could do a blog”
“Great idea. Let’s get the techies to set one up tomorrow”.

(Unfortunately they discovered that people don’t really like documenting processes and giving them a blog to do it on isn’t going to make them do it).

Then there was the blog-as-humaniser. CIOs were given blogs to make them more ‘human’. Or something.

“Hey, the CIO wants a blog. He reckons it’s important that everyone sees him as a human and approachable.”
“Great. I’ll set one up tomorrow”

(Unfortunately they discovered the CIO had no time to blog, nor could he write).

Some success

There are two settings where I have seen blog success in an organisation:

1. The university-run student blog
There is one big reason why such a blog works: money. Students are paid to write. It doesn’t guarantee a spectacular blog but it does ensure that posts are going to be relatively constant. The blog will thrive with skilful moderation and management. I saw one particularly good case of this at Sydney (and no, it wasn’t me!) Like all employees, student bloggers need management, they need coaxing, they need positive feedback and sometimes, hand-holding.

2. The enthusiast blog
This blog is essentially the child of one or two people who take it upon themselves to write it. Typically they like sharing information, they are excited and passionate about their topic and they can write. They are committed enough to write often and they don’t mind going the extra mile to moderate comments, seek out similar blogs and promote their own. They comment on other blogs, they respond to comments on their own blog. They keep abreast of their field and they feel it is their job to keep others up to speed, whether it is written into their job description or not. This is a critical point: there will be very few job descriptions (in a university setting) that include blogging. And with no requirement to do it, why would someone devote so much of what is essentially their own time to writing a blog for work?

And finally

The number of people who blog in general (and I mean serious, effective blogging of a high standard) is very low. A good blog requires enthusiasm, skill, commitment and knowledge. Despite what some will have you believe, not everyone can do it. Transpose this then to the work environment and trying to impose a blogging regime is doomed to fail. Unless there is already a culture of blogging that respects the place of blogs and takes them seriously it is very hard to convince people of their worth. This doesn’t mean there is no place for blogs in the workplace, there most definitely is. Just don’t be surprised when only a small percentage of people read the blog(s) and an even smaller percentage want to write, do it well and do it often. Value the ones that do it well, know your limits and don’t be defeated by the tens of abandoned blogs set up because someone thought it was a good idea.


Having been through a number of university website ‘redesigns’ Rosenfeld’s presentation hit a nerve with me. The complexity of both the university website and the organisational structure and politics of the university itself make the grand redesign a hopeless task that is doomed to fail. The ‘project’ mentality that goes along with such redesigns makes matter worse. The website is not a project, it’s an ongoing part of your organisation that requires constant fine-tuning rather than a cumbersome and political overhaul every two years. I’ll quit the rant and go straight to Rosenfeld. Enjoy.

(I’m also a little jealous that Illinois got Rosenfeld to talk to their web people).

Via College Web Guy.


Now I am out the other side of a university website redesign I have time to return to an article I read while in the midst of it: Menuing in Content Management: Implicit v Explicit

I found this article at a time when the project I was working on was experiencing menuing difficulties. Specifically, how to get the CMS to produce the menu system the design called for. The IA of the site had been designed with a relatively shallow menu system as an antidote to the previously very deep menu system that had grown somewhat organically over time. Once the site was put into the CMS we noticed the menu system was not that which was originally designed. Sure, all the content was there in the same structure dictated by the IA. The paths to that structure however were horribly deep.

The CMS was set to dynamically build the menus according to where the content sat in the backend. So every time a page was created a new menu item was created and placed according to where the page sat in the directory structure. Ugh. I could see how the menus on the previous site had become so out of control. In an attempt to make website maintenance easy by creating menu items for you, the CMS was becoming somewhat inflexible. (As a Mac person I see this as similar to a Windows problem: the system tries too hard to make things too easy, in the process basic tasks become convoluted and the system itself becomes inflexible and annoying. For example, who finds Clippy the paperclip ‘helpful’?)

To get around the problem we had to create a set of ‘minisites’ that forced the system to create menus as they were supposed to work. The backend organisation then became rather wider and less deep than the original setup. Several people expressed concern that the backend was becoming messy but others (including myself) expressed concern that the dynamic menu facility and the desire for a tidy backend could not dictate navigation paths on the site.

You can see our dilemma. As an IA I would fight tooth and nail to make the CMS deliver the structure and navigation paths called for in the design. People more focussed on the CMS itself would fight for a clean and tidy directory system. What was difficult was trying to get the concept of navigation as opposed to structure understood. The problem was not that the relationships between the information in the IA was not working, it was that the menus as represented on the page were not creating the correct paths that would make the information accessible. Also, the idea that sometimes the way the information is arranged does not always reflect the navigation was a curly one for some. Sometimes the navigation is different in order to accommodate users needs, expectations and web conventions. The trick is getting the navigation to give the user a mental picture of the site that is backed up by logical organisation of information. It’s a very fine line between the two and sometimes getting the balance right is difficult.

Another difficulty comes about due to the fact that IA work for university ‘corporate’ websites is not always straightforward. Due to the fractured nature of the organisation where every department/school/faculty has its own identity and the wide range of potential audiences for the site, it is almost impossible to design a ’single’ university website using one approach. Rather, a university web presence needs to be approached as a set of individual sites attracting different audiences, with varying needs and expectations. The reality is though that single audience members need to traverse a number of these sites in any one visit so paths to information are critical. These bridges between sites can make or break a successful and seamless visit.

Such a situation does not lend itself to the tree-like approach to navigation that is required for a dynamic menu creation situation as described above. There will always be some tweaking and hacking, for lack of a better term.

The post from Gadgetopia then gave me some respite in that it reminded me that this problem was not uncommon and that there was no real answer in terms of which is better: “explicit menuing” (which requires more maintainence and provides more control) or “implicit menuing” (which require less maintenance but give less control).

1. Implicit, meaning the menus are driven off the content structure of the site.
2. Explicit, meaning you have a specific “menu” structure, and it links to specified content in the site, regardless of where that content is.

What we were creating was somewhere in between where the structure of the site does reflect the menuing but not as implicitly as the CMS wanted to have it. Considering this, I will be interested to see how the site copes with growth in a devolved authorship environment. Control of structure is important in this case and in such an environment it can prove difficult, but discussion of THAT little problem is for another post.


I was working in a CMS yesterday, using a browser, and I got this error message:

I am not sure what I think of it. At first I thought it was pretty silly, it seemed to be treating me like an idiot, talking down to me like some techies and Helpdeskers have in the past. It seemed to epitomise that attitude that accuses users when things go wrong with machines: “what did YOU DO to IT?” Once you’ve worked around computers for long enough you will know that sometimes there seems to be no obvious answer for errors, otherwise there would be no need for the millions of troubleshooting forums on the web. Taking out your frustration on a user is not really going to solve the problem and is one reason why many novices are cautious and scared around computers, often feeling they are going to break them. (Not without an axe is always my answer).

Another reason it annoyed was the fact that I didn’t know what I had done to cause the error. The screen was indicating that my cursor had selected only one line of a bullet-list. As far as I could tell, there was no error.

Perhaps what irked me most was the anthropomorphic nature of the message, a style that is outdated and pretty geeky. Things like Flickr don’t have the ‘page’ talk to me, they just give me a gentle prod in the right direction. This message seemed not only overly geeky and condescending while trying to be human and cool, it didn’t give me any more knowledge about what had happened and how I could stop it in the future, especially since the selected text on my screen backed up what I thought I did.

And then again, maybe I spend way too long thinking about error messages. Yes. Life, get one. Ahem.

PS Flickr now does video. Well, I’m excited.


Last week I gave a brief presentation on usability. It was meant to be a quick introduction to the concept of usability and to provide tips on how websites maintained in a devolved environment, where content authors are not always web professionals, can be improved through small usability improvements. I have been asked for some accompanying resources so here are the slides and following, a brief list of introductory usability resources.

Usability

Accessibility

Higher Ed

Book