APS9: The Politics of Doing #IA for #HighEd

Aaron Baker 
Web Services Coordinator, University of Arkansas at Little Rock


The audio for this podcast can be downloaded at http://2011.highedweb.org/presentations/APS9.mp3



Announcer:  This is one of the series of podcasts from the Hyatt Web Conference in Austin 2011.

Aaron Baker:  Everybody, my name is Aaron Baker. I work with the University of Arkansas at Little Rock. There's a bunch of us here from UALR. There's also several other Arkansans here so a big shoutout to them. This presentation is going to be about information architecture, IA and specifically the pitfalls of doing this kind of work for a .edu website in the realm of higher education.

Myself, I, uhm, just to give you a little background, at UALR we don't have a centralized web office. I guess I'm the closest thing to that. I worked in academic affairs and to the Provost for eight years or more or less and then was reorganized to IT, that was a fun process.
01:05 And we have several people in IT that do web development in systems or in a development shop and then we have several people in communications under advancement and those of course worlds apart. And so we've been really lucky because we all get along pretty well. We have a very good working relationship, sort of centered around the website and we have agreed to do things a certain way which is always really helpful. You know having a working relationship with your colleagues across the board is always a good thing.

My job as web services coordinator, a lot of the work that goes around the website comes through me as a central point of contact which is really helpful. The design and content responsibilities are more or less in the communications office and then in the other parts of IT are the let's keep the lights on kind of policies, you know make sure the thing is running.
02:08
And then anywhere with those two meet is where I'm involved. So I do things like usability studies. I, of course do information architecture. I manage the information space, the name space, so when people come to me, and requests something like you know, gohogs.ualr.edu, I have to tell them no, we're not the Razorbacks it's a different campus. And some people, you know, want justifiable name spaces but unless you have that institutional picture in your mind, everyone else you talk to is just going to know about their own little world, right? I'm going to try something during this presentation but I can already tell you that I don't know how well it's going to work because it's just, I only have two hands but I've got suggested tweets just in case.
03:07
Just in case you are sitting there wondering what to take away from this presentation, I'm going to try to keep it fun. Information architecture can be kind of boring but I am not boring so I don't want to ever, I don't want that ever be the case so here's just a sample tweet saying, "Hey, come to this session. There's plenty of seats left." So what is information architecture or IA and how to explain it.
 
One of the benefits that you'll get from a good understanding of information architecture, the best thing that you could do is to communicate what information architecture is. You probably do all of these things already right? You have a homepage. Decisions are made on how things get on that homepage and specifically I'm talking about navigation. How do these links get there? Did you just make it up? Maybe you did.
04:01
But you need to sort of express the process that you went through and by doing so and communicating that process to the campus, to your boss, to the upper administration, you're better served because that is the number one way you can mitigate conflict over these decisions, right? If you simply put everything on the homepage that people ask you to number one you're going to have a mess. Number two, you're not going to have any way to defend yourself. You already have a process by which you make decisions. So get it on paper, write it out and get it approved. Get people talking about how these decisions are made.

This is really about web governance and if you don't have web governance yet, that's OK, you still make decisions. You still have to defend the decisions that you make. Information architecture is ah, it exists in the boundaries between design and development, right? It's the wire frame.
05:03
It's the interaction points. It's the, you know, if you had to strip away all of the graphics and most of the content from your homepage, you'd be left with boxes and labels and sort of areas where things happen where you think things to happen, the words that you think people should click on or want to click on.

This is the, ah probably the seminal book on information architecture it's called the Polar Bear book and it was written by Peter Morville and Louis Rosenfeld. It's information architecture for the worldwide web. It's a big undertaking to read but you'll be better served for it. It really covers the gamut. And it's fun reading these things because they're written for Amazon, right? They're written for large websites but they're not specific to higher education. Higher education websites are mentioned quite a bit in here but we as higher education web professionals have to take this plethora of good advice from the internet regarding the web and what to do right and standards. 
06:10
And we have to figure out how we're going to apply it to ourselves because we don't often work in a model that is monetized. And you'll find almost all of the advice in the web exist for people who are trying to monetize their websites. And so, because we're not explicitly trying to put a dollar value to everything that happens on the website, sometimes we get a little lazy and the decisions that we make because we don't have a bottom line. We do have a bottom line but it's really hard to measure from the website's perspective. 

Information architecture, these are four definitions from the Polar Bear book. It's the combination of organization and labeling, navigation schemes within an organization. The structural design of an information space to facilitate task completion and intuitive access to content. That's really important.
07:01
One of the ways that information architecture is formed is through user and task analysis. Who are your users? Well we've got perspective students and current students, alumni, family, friends, faculty but think about perspective students and current students and then think about the tasks that they are trying to complete using your website and that shows you immediately why information architecture is necessary because that task list for those two different constituents using the same website are completely different.

And so, you know, it's really important to notate what people are trying to do to get an understanding for yourself because that becomes the basis on how you'll measure success. So, and that whole process can be considered the information architecture. Information architecture is also the art and science of structuring and classifying websites. Well you do information architecture in internets to help people find and manage information and finally an emerging discipline and community of practice focused on bringing principles, design and architecture to the digital landscape. 
08:136
So here's a sample tweet just in case you're falling asleep, "The Polar Bear book is pretty cool. You should read it aps9." By the way if you're having trouble in wireless in here I find that the BB1 network is a lot better than the Hyatt network. That's just one to grow on.

To me information architecture is convergence. It is where users interacting with content in a specific context. So it's almost, it's not enough to just think about what, who users are and what they're trying to do, there's often a situation that exists beyond our control in which they're trying to do it. You know, you're not simply trying to log in to blackboard to view your online courses, you're trying to do that from your phone or you're trying to do that while at somebody else's house, there's a whole context that really can complicate things.
09:14
And going over these each one by one users, we have audiences like the perspective and current students. They have tasks. They have needs. Those are different. They have different information seeking behavior. Some of them like to browse. Some of them like to search. Those are two very different information seeking behaviors. Some people like to skim. Some people like to ignore.

And they have different experiences when dealing with the web, right? They may have never used the web before. I mean it's hard to understand that these people still exist but they do and we still have to account for them. Or they might think that they know everything about the web and those are equally as dangerous. In the content realm you have different documents and data types.
10:02
You have content objects. You have an existing structure. It's not like we are just going to build a university out of the ground and have this perfect, you know, information architecture or landscape. No. Even if you go back home tomorrow, you are, you have to deal with that structure that already exists. You never really get to destroy an information architecture and come up with a new one. And if you did, your users would not be better served.

So you always have to deal with something that already exists, the history. That's why information architecture is messy. It's never very perfect. You know, the science behind it comes from taxonomy studies and librarians and all of those kinds of things and you want to believe that they exist in a really well put together world and that's just not true. We know this, right? We know that our homepages and the structures and the world we have to deal with is not perfect. And that's OK, we just have to accept that and move on.
11:05
And then the context. And this is where things get really tricky because we know what the users want and we know what they need or we can, we have ways of soliciting that information or they'll tell us. But then on the other side the institution has its own goals, right? You want graduate students in six years. You have funding dilemmas or needs. There's politics, of course. There's a culture at your university. There's certain technologies that you have to use or you wish you could use, resources and other constraints. 

And so you know, classic Venn diagram. This actually comes from the book or from Symantec Studios which is Peter Morville's sort of consultancy group. And information architecture sort of exist right here in the center of all of this 
12:07
Information architecture principles. So these are kinds of, the kinds of things that information architects do and are concerned about. And they are also deliverable organizations. On face value it seems like that's an easy win. The organization is not necessarily the most easy thing to do. I can organize my closet in a way that's easy for me but is my organization scheme the best for anyone else, right? 

Imagine, I'll talk more about this in a second but imagine a kitchen. A kitchen that you have in your house that you sort of organized in a specific way. You know, you like the silverware drawer on the right side of the stove so you can get at things if you need them. Or maybe the glasses go right above the dishwasher because as you're emptying it, you just want to out them right up there, right?
13:05
You've sort of thought about these things but then someone else comes into your house and they're like, "Pfff, I can't find the spoons." How do you figure this out? If you think about it, a higher education website is a kitchen that everybody has to sort of agree on. And we want it to be this like perfectly designed space that anybody can sort of come in and use. But then we also have this history of how the kitchen was laid out. And there are certain walls and things we can't tear down and that awful island that we keep tripping over.

Kitchens are incredibly hard to design when you're trying to design them for everybody. And the accessibility model fits in there too, you know. It's like you might want high countertops but that's not easy for everybody. So you have to think about all these things. Labeling is very important too. We get into a lot of labeling. My colleagues and I get in a lot of labeling discussions.
14:01
You know, is this thing going to be called an index or is it going to be called a directory? Those are two very loaded terms and essentially we wouldn't use them both but, and they're the same so we had to pick one, right? Is it going to be an index of sites and people? Or is it a directory for students? Those are all hard choices. And then rising above those two things is navigation. Navigation subsumes all of those previous problems but then another one space. You know, you can't have an infinitely long navigation area. So now you're sort of juggling the psychology of choice when you're talking about navigation because organization is just a collection. Labeling is just what you call something. Navigation is how you communicate what people choose. People choose in navigation. They select something or they don't. The reverse is also true.
15:02
And then when navigation isn't enough or when navigation is ignored, we have search. Search systems are something that I think we are dangerously ignoring in higher ed because of resource constraints. How many of you manage your own search engines? And by that I mean you don't use Google custom search. OK. Not too bad but you see that, and we use Google custom search and I'm not ashamed of that but were we to have the time, we would be better served by, it should be somebody's job to sort of look at every keyword and make sure that we either have something to respond to that content with directly and have the most customizable and programmable interface. It's also a great way to find labeling, right? These things sort of are interwoven.
16:02
When you look at what people are searching for, you all of a sudden know more about what you should label things. And all of this is sort of researched and tested through usability studies. I've never been so frustrated in my life than doing a usability study with a perspective student from a feeder community college where they decided to use the search tool and they just immediately skipped past that first result which was the best one and then started to get into really questionable territory and then didn't complete the task because, you know.

And what can you do? When you're doing a usability test, you can't sort of guide somebody in what to click on. The whole point is to figure out what you should be labeling things and even when you think you've done the best job, you know, there are reasons, each user brings their own reasons as to why they do or do not click on something.
17:02
Uhm, and for each of these, I'm going to just get back. Each of these has research potential to figure out more about each component and more about how they interact. So for your users you can do personas, right? Information architects deal a lot with personas that helps us sort of figure out what certain kinds of students will do in certain situation. One persona I have is, you know, her name is Betty and Betty is a non-traditional 40-year-old female who has children and they have grown up and are in college themselves or maybe out of college or maybe they're in the last years of high school. Betty is middle aged. She has a job. She's a professional so she goes to school part-time. 
18:04
She embodies all of these certain characteristics of that kind of student and so when we do things we often have to say, OK, what's Betty going to think about this. It's important that you name your personas and I wish I could have white sides cutouts of them so I could point. We get in heated arguments and it's like, well I don't think this is good for Betty.

[Laughter]

And most of the time, you know, Betty is put up against Billy. And Billy is that rock and roll 18-year-old top of his class graduate from the science magnet school. He's majoring in engineering. He lives on campus to go to school full time. He plays World of Warcraft. He hacks into the network when he can. He's got a blog about it and he tweets to me all the time. Something's going wrong on campus and you ask the web guy.
19:06
So  Billy's got different needs than Betty and Billy has different experiences. He uses the web a lot more but he's, and completely younger and those are the kinds of things that's the kind of user-research we do. We develop personas. We ask questions about what each would do. We do user testing. We find people like Billy, actual people. We find people like Betty, actual people.

With content, you do different kinds of research or you struggle with content in different ways. You do indexing. Indexing doesn't sound very sexy but it can be very effective. You look at your content and you find out what those keywords are, what are the tags. That's indexing, cataloging. You  look at your metadata, you start to develop not only the keywords or the words that are in that content but what are the related words?
20:03
So if you're, if you're talking about a policy on academic suspension, a related policy is academic probation. Students don't really know which one they're looking for if they're looking for it at all but the information is related and so you should develop connections between it. All of this, the content piece, I wish it weren't just a circle. I wish it were more of an iceberg that existed well beyond to discover this building. It is a mammoth thing to undertake. And I'm glad I'm not talking about it today.

The context is also special in terms of how you go about researching it. You've got to get handle on what the institutional goals are. What is your strategic plan? What is the institution trying to do, trying to accomplish?
21:00
How does your university measure success? What are the success metrics? Is it just head count? Is it SSCH? How did they calculate return on investment if they do? What's the business analysis? And how do you manage expectations of the administration. So the context piece is more of an administrative piece because when you're thinking about the context, you're thinking about the situation surrounding your environment, your university environment.

OK, skipping ahead principles. Controlled vocabulary and that's something I'm going to talk about in a second. A controlled vocabulary is that part where you sort of developed your own in-house dictionary for the words that you use on campus. It's a subset of natural language. It's words like academic clemency or the advising, you know. The special university, your university's specific words.
22:07
And when you, the controlled part is really important. It's like the French language. You can't just invent a word in French like you can in English. In English, we have a very pragmatic language. If we want to say, "Hey, I'm going to Google it. yeah Google it." All of a sudden that means to Google. It's like we just built that word out of nowhere and we didn't get anybody's approval for that, and wow thank God for the English language for that.

Such freedom doesn't exist in the French language. Maybe they can just do it but the point is that with the controlled vocabulary, you sort of define the words that you're going to use for these particular situations. Back to the index, you know, we decided to call it the A to Z index. We've made the decision in 2005 and it's done. That was a point in history where we now have something that we've named it and that's what it's going to be.
23:07
We can't just call it the directory over here, the vocabulary's control, we named it the index that's what it's going to be. Also, in the controlled vocabulary is where you developed the related words that's for sore eye. Oh yeah, tweet moment. Managing information architecture is like planning or maintaining the perfect kitchen. I should have said that a while ago.

OK. So here's show and tell time. Again, my role in the university website has little to do with design or any of those other things, so you know critic away. And I would probably use something else except I have really good access to this and the data behind it. So the areas I have, can I just point with this? Yes I can. We have a tools navigation area. So this is like a quick link to blackboard or email, the A to Z index, class schedule map, things that we've determined that people like to use, quick link to.
24:09
We have this thing called boss. You all have that special thing that you've named it. It's sort of the student portal. How do they get into self service or whatever else. This is a good example of the history. Would I have named the thing boss? No. But do I have to live with it? Yeah. Some people have things like tiger mail or, you know sounds like or it could be worse. I should just start calling it Bruce Springsteen. Hey I lost my password. What do I do? Just log into Bruce Springsteen.

Uhm, an Apply link. This is what we call the cafeteria navigation or audience-based navigation. We call it the cafeteria nav because these people sit together in the cafeteria. So all the things for perspectives is here.
25:00
In case I don't get to it later, this navigation is equally terrible for all of us. The audience-based navigation, perspective students hardly ever identify themselves but the win with this navigation, this is a drop down menu too. I mean this is a screen shot but the win with this kind of navigation is it's that quick links. How many of you have quick links or the thing you hate to maintain and you're right for hating it but guess what? That's that one place where you can put everything to sit. Just get it in the quick links. It's actually very valuable and I'll talk about how I measure these things here in a second.

Banner area. This is the topic navigation. So we call this information about area we have about the university. And you all have these same links. You know, academic programs, admissions, financial aid, libraries, outreach. The giving link, you got a giving link? Yeah. We got a giving link. Athletics, more news, blogs and then the footer navigation and social links.
26:09
This tools navigation on the top right, something I have done recently, which I think is really helpful is I added event tracking to that area so I can measure it explicitly. So if, how many of you are running analytics of some kind? Google analytics specifically. So if you add on-click event handlers to links, you can measure that link specifically. Otherwise, you really can only measure the distance between one page and the other.

And we've got like five links to the index on this homepage. I need to know which one they clicked, right? I need to know did they click the one on the top right corner? Or did they click the one in the footer? So I did on-click even handling to these things and all of sudden I'm able to measure, you know in the past month I pulled this one today.
27:03
In the past month, those links in particular, the ones in the top right corner where you use, you know, almost 300,000 times. What are people clicking on on the homepage? That's the question I get. Or that's actually now a rhetorical question I ask everyone else when we start talking about the homepage. What do you think people click on? Well they click on blackboard and most people, on our campus, I don't know if this is similar on campuses but there is the situation where people don't know how to get anywhere unless the homepage gets them there. They don't know how to log in to that specific system. The homepage is down and that means the world is over.

I can't get into my email, well just go to mail.ualr.edu. No I need to click on it and so there they are. All of those click blackboard email box. This information's really helpful for me because people want to add something like giving to that top right corner and I want to say no, uh uh.
28:04
You've got your giving link on the left and I can tell them exactly how many clicks it gets, 0.1% of the total clicked. I've measured it and you can do the same. Now what's the political fallout if you try to get rid of that link? Immense. So in the perfect space, you want to develop your navigation through users and tasks and you want to prioritize that into a nice neat and tidy list but the reality of the situation is not every link is going to be there because of user and task analysis. And that's got to be OK. Well those are the situations that we have to live with when developing these name spaces.

More about controlled vocabulary. I talked about it's a defined set of natural language, the metadata. Uhm, grouping equivalent related or sibling terms. This is something that's ongoing and this is actually a really cool way to engage faculty in the website.
29:03
I don't always try to get faculty and put on how to do my job because that's a real dangerous situation. We are like everyone else, we're struggling with governance and web strategy and you know. Just what are we doing with the web? Those really philosophical questions that administrators ask you in the hallway or outside the Starbucks. And you just wish that you had like the elevator speech in your pocket. You're like "Whoa, the web is a place where users meet interactive goals for the institution."

But this project that I did once was really helpful because the faculty really got into this idea because on some level faculty are just as concerned about student's success as everyone else. And they know that the university is a complicated space and they are the ones that invented these terms. So give them ownership in some of these things, you know. Help them help you develop what the parameters are. What are we calling this and why are we calling it? Why? What is the difference between these two policies and how can we communicate that to our students?
30:17
I want to talk a little bit about information architecture process. Research, I talked a lot about the kinds of research I do. User research, who are these people? What are they like? Task analysis. What are they trying to accomplish? And what does that look like when I watch them try to accomplish that? Those are all very telling the strategy. The strategy part is the perfect information architecture that you want to sort of arrive at but you know you never will.

This is in the perfect world, this is how the information architecture space would be designed. We'd have these links and I'd forget about that giving link or I'd find some other way to do that. I'd put it somewhere. It's not like it can just go away but develop that, get that done and write it in your journal and close the book. No just kidding. Have that strategy down and talk about why you think that that is the perfect information architecture space.
31:19
Again going back to the communicative part about this, each one of these, you sort of think about what would best be articulate your ideas, communicate them and I'll show you a little bit what I mean by that but basically you've got to start talking about this process and talking about what you'd like to see and then the decisions that you're making instead because of the political realities.

When you start doing this and engaging with people, number one they'll trust you more. Number two, they'll respect you more and you will begin to be able to communicate this, you know, you'll get some empathy in the hard decisions that you have to make.
32:01
I mean I joke a lot about the decision to choose index over directory but if you think about it, if you know and understand the gravity of these decisions, then you can appreciate that just making arbitrarily isn't going to cut it. You're going to decide a term that you'll have to live with. Somebody did not understand that gravity when they called the thing boss. It's actually an acronym for banner online self service. Welcome to my world. I got to live with it unless, you know.

Sometimes I hope, now here's a strategy piece, I hope that this portal project that we're working on, you know, will have the ability to rebrand that thing and just sort of push that term under the rug. But let me tell you once people know what that is they love it. They are just, "Oh go to boss. Where's boss? Oh I got my boss in here." And we have a test in the portal. I'm telling you it's the truth.
33:04
Each one of these things is iterative I believe wholeheartedly in iterative design. Iterative design is a concept from user-syner design where you sort of do something and then you test it with real users and then you just repeat that process. It's like rinse repeat, you wash your hair and then you wash it again. I never understood that when I was reading about shampoo, you know, it says rinse repeat. Why would I repeat this process? I just washed my hair. Do I need to wash it again? But yes you do, you need to do that for each step.

Each step takes the sort of loopty loop process. You go through a design phase where you're translating your perfect world architecture into an actual physical space. All of a sudden, horizontal nav becomes a pretty big problem because it's not infinite, right? And you're constrained by that page boundary and unless you're going to go the way of Amazon, where they don't care how many tabs we have and how many levels, you got to understand how this thing is going to look.
34:09
All of a sudden it's not really about organization or labeling but there's some sort of visual constraint on the navigation itself. That's just an ethema to architecture in the information space. Then you go through an implementation cycle when all of a sudden things, you have to implement it and the technology that you have and maybe that works, maybe it doesn't and then the administration piece.

What about politics? So you could guess that none of these decisions are ever political. I have a few stories to tell. That left hand navigation, you know, where those academics and giving and libraries? There's a few situations where it got political. The first one the faculty wanted a link that said research.
35:01
Research is becoming a very, much more important focus of our university than it has in the past. We've been designated a Carnegie Research Intensive University, whatever that means and researches becoming a kind of a big deal. Why isn't it reflected on the homepage? Well in all actuality, when I did the strategy for this navigation in 2005, we had a research link but we had nowhere for it to go. And so I wasn't going to put a link that didn't go anywhere or to a page that said yeah we do research.

And so we came with this, we started these discussions about what do you want to see behind that research link? Yes we need a research link. What does it need to be? And they just got all excited about these kinds of content and you know, research stories and things. And I was like, yes! You know getting that faculty involved and we stood behind that idea because they came up to me like I was the enemy, that I was preventing this from happening. I was like no. I've always wanted this to happen. I'm with you on this one. Let's have a research link. Let's have a huge research section. Who's going to write it?
36:16
And that's where that conversation moved. When you talk about the implementation, the administration parts of that I was like, I'm spending most of my time changing usernames and passwords for people. I can't necessarily write about your research about ants which is apparently very important. Or you know it was like, it doesn't matter whatever.

And that is ongoing right now. If you look at our homepage today you'd still don't see research. We have created, we have started to create a research section and academic affairs and Provost officers are sort of taking that on but it's nowhere where it needs to be. Not for me to put a link to it on the homepage. The second story is about the library's link. If you noticed it said libraries and collections, which is not uncommon. A lot of you have more than one library right? Especially if you're at a big school.
37:07
We're sort of halfway at that point. We have a central big library and we have a lot of other collections and for political reasons we had to call that libraries and collections and it took you to a jump page where you were like did you want the main library or one of these other libraries that nobody else uses. And that was a situation where I was like this is ridiculous because I looked at analytic, you know the quick traffic and 90% of the people or more that went to libraries and collections just went on to the actual, the main library page.

This is another situation where they were adamant, the new dean of the library was adamant about that link needs to go here. And I was like again I agree with you, let's make that happen. So we had to work together on a resolution. I was like you can understand that if I just change that link to go to your homepage that I've got all of those librarians from the other collections on my case.
38:02
And it was true I did but the process that we went through, we, ahead of the change we talked about it with all of them. The main library agreed to put a libraries and collections page on their site so that that quick traffic could still be achieved just in a reverse order. Click on libraries, you get the main library click on other collections and that's your portal to those other collections that are a part of the main library.

They agreed to do that, that was a good win. And then I had them go through their channels. So I had to have the, I said that I have to have approval from the Vice Chancellor in Provost to make this happen. So that if any of those other academics who wanted to get on my case I could simply say, your VC approved this change. And then we communicated it all and it was happy. Well it wasn't really happy but I was happy. There are few things that I want to talk about really quickly. Prevention. Try not to get into these situations, right?
39:07
Understand the existing political climate. Those librarians don't agree with each other. They did not like each other. It's the reason we have different libraries. It's the truth. Have a policy. If you don't have a web policy work on this to continue. We don't have one explicitly yet but we're working on it. Have a policy that explicitly states who or how IA is determined. IA is the reality. Start pushing that. Explain this to your administration that these decisions get made somehow, we might as well have an explicit process.

Do your homework. Have the data. When people say I want this link or I don't want that link, you need to be able to say, well I looked at the numbers and it turns out not a lot of people are clicking on that or that's just not a heavily used thing so maybe we shouldn't put it on the homepage.
40:01
They'll immediately say, "Well if you put on the homepage, people will look at it." And you might actually have to do that but the point is you will still then be able to look at the data and say, no actually now all of your traffic are bounces and exits. So have all these tools ready and then make yourself known as the person who implements these changes and be transparent about the actions you take and why.

An example of that, here's an article I wrote in 2009 about navigation theory and practice at ualr.edu. If you Google this phrase I think you'll find it. I have a short code linked to it in the next slide but I just sort of go into the whole process and say, you know what these decisions have to be made and this is how we did it. This is most likely in response to that whole research debacle where, yeah we did identify that as a major focus of the university and something we should have on our homepage.
41:01
I didn't state in here that the content wasn't there so it didn't link to it but this is my article on how these decisions get made. And here's what you can tweet. It's bit.ly/UALRnav and don't forget hash tag aps9. I've got five minutes, I'm going to race through.

Stop, look, listen. What are you doing? What? What? What are you doing?

[Laughter]

Recognize when you've got potential conflicts headed your way. That's the moment where you need to have that meeting with your boss and your boss's boss and be like look people are getting unhappy. Here's what they want to do. Here's what I want to do or don't want to do. Let's start talking about this, get everybody involved. The worst thing is when your boss tells you about these problems. Sometimes that happens. When the storm is brewing take yourself into a neutral position and get everybody engaged.
42:03
The worst thing that could happen is this being about you. Architecture's not about you. Architecture is about the website and about navigation, about users, and content and context. So don't make it about yourself. Get yourself out of that situation and let others grapple with these hard decisions. It's possible and likely that they don't understand the gravity of these decisions so that's the role you need to take.

Don't be afraid to try it both ways. That sounds kind of dirty. Measure the difference. Maybe one works better than the other. That's where, you know, it is the web. This is the part of the web that can change. Try, there are ways that are appropriate in which you try to do controlled choice to pick the best option, AB testing. Do more research.

Finally if it's too late to mitigate conflict, you need to find a resolution quickly. This does not need to be something that takes a lot of your time, right? And the more. it's like faculty, when one of them gets angry at you, they have this ability to get more of them on your side and all of a sudden you've got an entire group of faculty who are sort of against the website and then they'll soon be against everything you do. And that's not what you want.
43:24
You need to be effective they need to trust you. So find that resolution, don't make it about you. If necessary refrain the issue. It's like this isn't really about that link in particular, it's about me saying yes to that link and everyone else consequently asking me for that link, right? ROTC wanted a link on the homepage like a giant ROTC graphic under the Trojan athletics head. I was like how many people are in ROTC? As a component of our 13,000 students. I have to say, listen I can't give you that link because I can't give all student orgs that link.
44:02
And then whatever you do measure it. Use analytics. Use the event tracking. Use it as a way to communicate why it's a success or why it isn't. Conflicts arise, get yourself out of the situations, read about it. All right. We have like less than a minute to talk about stuff. Any  questions? You first.

Audience 1:  How do you link on the homepage?

Aaron Baker:  Diversity link where does it go?

Audience 1:  The board does it.

Aaron Baker:  I mean what is that content? Was it like yay we're diverse? This is...

Audience 1:  Yeah.

Aaron Baker:  It's in the footer next to the we are accredited link. That one too. Yeah, that one too in the footer. Yes. In the orange?

Audience 2:  I have a situation where the dean decides what goes on the front page.

Aaron Baker:  Oh.

[Laughter] [Cross-talk]
45:02
When the dean's decide what goes on the front page, you get this everything needs to be accessible in three clicks. That's what I read in that business magazine.

[Laughter]

I was like, then every page is one million clicks. Congratulations!

Audience 2:  You know like, oh well this nationally ranked stuff, someone put it on their school front page.

Aaron Baker:  Sounds like a news or a feature item rather than a navigational concern but I smell what you're stepping in.

[Laughter]

Audience 2
:  It's a problem at times or...

Aaron Baker:  No comment.

Audience 2:  OK.

Aaron Baker:  Sorry. Yeah?

Audience 3:  Is this like the same part of it's not like the others do when you have specialized language. Do you think that the previous step was to prevent from happening, right?
46:00
Aaron Baker:  Yes but you're always going to have things that are specific to your institution and sometimes you can't not plan these things boss. Boss is in our controlled vocabulary and it's got to be there because somebody put it there but we're not going to all of a sudden have other links to it called the student information portal. That's what I mean by that. And again it's messy, it's messy. But I agree with you that if you did a lot of natural research you could use things like... Here's an example, you know if you try to log in to our email system and you fail, it says instead of saying wrong username or password, it says, "The credentials you provided cannot be determined to be authentic" or something. You probably had the same message. What's that? It's not natural language.

I agree with you that doing the right kind of research should prevent a lot of these weird terms in your controlled vocabulary. But really I mean things like advising, that's a natural word but depending on who you ask at my university, they'll tell you that advising is a different thing and has a different process. Each of one of them will say that. And that's kind of a problem.