And it came to me then that every plan, Is a tiny prayer to father time

Thoughts on organizing information

So at the day job we are dealing with how to effectively organize and structure our data so that the largest majority of our visitors can find the content they care about... quickly.

The mantra today is 'everything within two clicks.' But how realistic is that?

Being the webmaster of a college/university website means that you have to build your structure well enough to accomadate 2-3,000 pages of unique content. Anyone else sick to their stomach right about now, or is it just me?

The Usual Approach

So the time-honored approach (if I can even say that about a web related issue) is to create a monolithic site structure, trying to divine the most logical, obvious place to place information. In this scenario you try your best to group like pieces of content while keeping an eye on those generic pieces of content that will show up in multiple locations.

I have a problem with this scenario. It assumes that every visitor that hits your site thinks like you do... not only would this be bad for the existence of the human race (we don't want lots of people thinking like you... yes you, I am looking right at you who else would it be! Sheesh) it is impossible. I am an almost 29 year old web geek, I do not think like a 16 year old Chilean girl; so how could I ever determine where she would think to look for information on our Equestrian Program?

I can't. So what is one to do? Should we not create a stable, well thought out site structure? Yes... yes you should, but I don't think that is where we should stop.

An unconventional approach

So what is an enterprising webmaster to do? Sure there is always the search engine, but if the page doesn't have the term you are looking for associated with it, then searches are useless.

And that is the core of this really isn't it? We can't anticipate what term Joe Junior High School will use when searching for information on Dorm Life, so we don't. We allow Joe to restructure the site as he sees fit. To associate content on our site with terms and words that make sense to him; and probably a whole lot of other people in his age bracket.

This is the approach that I am going to explore in the CMS we are designing for Asbury. The generations we are targeting in our next recruitment cycles are the social software generations. Tagging and Digging is not only something they do, it is something they will expect. So why not give them the technology that they already use, and allow them --the user-- to help us organize our content?

The plan is to have an 'add a tag' area on each page, with some explanatory text. Then once we have been running for a month or two collecting data, expose the user inputed 'tags' via heatmaps and the like.

This should also increase the reliability of our searches since we will be searching against the tag info as well, not just our indexed content.

Now a question for you

So the real reason for this missive is to see what you, my lovely readers, have to say on this topic. Do you think my plan is a good one, or not? If you are in the not camp I would like to know why as well as any alternatives you might have.

The floor is now officially open.

Tagged: technology and web-design

February 25

Skip to Comments

Patrick
Stroll on over and visit Patrick
March 1, 2007

I think that two-pronged approach is brilliant. It's Dynamic. You create a structure to allow conventional organization of data, but allow users to structure their own in a more "fuzzy" cluster.
The Social Software Generation is who we see in our freshman and possibly sophomore univeristy classes, but more importantly its the high school senior and junior students who are our target audience. I'm going to have to forward this article to my coworkers.
Shane
Stroll on over and visit Shane
March 1, 2007

This is off subject but I don't know how else to contact you so please excuse me if this annoys you.
Thank you very much for the "CJD Spam Nuke" plugin. I've been getting hit with a ton of spam lately. I was highly suspicious that marking spam as "spam" gave no indication that it was actually being deleted. I didn't want a bunch of crap filling up my database so I googled for a plugin and yours was in the top of the results.
I'm baffled about the authors bitching at you about giving users a way to delete this garbage based on some mysterious future benefit. I'm no expert but I see no reason to have my database cluttered with this junk. And I don't appreciate them assuming we are too stupid to make a decision about this on our own.
Thanks again for the plugin.
Seb
Stroll on over and visit Seb
March 1, 2007

Chris
This is exactly what we are trying at the Powerhouse Musuem with a new project which will give access to (in time) well over 100,000 individual collection objects across a range of dfferent international museums.
We have similar problems - how can we decide where a particular type of object sits in a strict hierachy? More so once we start connecting collections that aren't classified in the same ways our are - an art museum may classify and describe a chair in a very different way to a design musuem or a science musuem for example.
This is less of a problem for a fixed specialist audience who shares a similar mindset for classification (and uses a controlled thesauraus) but to expose the long tail of our collections to wider audiences we have developed a way of using three types of tags - one automatically generated, one classifier/specialist generated, and the other user generated; combined with use tracking to generate a dynamic site structure more rhizomatic than tree-like.
I'd be very interested to hear how you go.
(why am i reading your blog? i just installed the k2 beta . . . and followed some links)
Luke Wertz
Stroll on over and visit Luke Wertz
March 1, 2007

Turning a college website into a social-net style site? Great for the kids. Very Web2.0. But bear in mind, only a tiny percentage of the most Internet-saavy people understand the ideas behind Web 2.0 and why it works. I'll be 100% honest with you -- I've been using the Asbury website quite a bit in the past few weeks (re-applying is work, man!), and when I'm using the site, I'm not sitting there thinking "Huh, how can I help Cris & Co. out today." My thought patters are something more along the lines of: "Okay, Aldersgate... Where are the pictures, I gotta finish up this application and get it in before the offices close in 12 minutes and I'd like to visualize what I'm applying for!" (by the way, your "tours" of Aldersgate are broken... no go on the visualizations!)
No, I highly doubt tagging or user-sorted data is going to work on a more corporate (professional as apposed to blogger) site. In the blogging community you've got a group of people that basically understand the basics of blog-style-design... you don't have "about" -- you have the "colophon." The "Archives" are where the past posts are, and the pictures of your cat are under "Flickr" (whatever the heck that is). Talk about a like-minded group of people. I don't think that transporting a design structure that is understood by that group of people onto a corporate framework is going to work. My sister would sit and a tag cloud and be 100% confused. I don't think it would be practical, I guess...
Just my $.02
Jeremy
Stroll on over and visit Jeremy
March 1, 2007

I hear Luke's point of view, and I believe that it is a valid one and should be considered. However, there is always going to be a deal of learning involved when approaching technology. It is not far away from being a social responsibility to learn the new tools. It is a college....
I think the interesting part of the problem is how you go about teaching people, to use the system by design, and encouraging them to use it, without having to spell it out for them. Which, leads you back to your original problem. How do you get into the mind of 18 yr old freshman and know what they are looking at and clicking on.
To encourage the future users to take a moment to learn, I think we, as designers, need to start thinking about how to 'teach adaptability' through our design of these systems. I think the 'product' associated with the design, should actually be the interaction with the information and the experience of being connected to it, rather than designing the product as a static item, like a 'form', or 'file' that users like Luke might be searching for. I think it is about getting users like Luke to see the reward in his interaction, and that the 'file' that he is looking for, is a form of interaction and adds to a experience.
It is also a tough thing to do, when you content is rather dry, and when you are not myspace, or delicious.
(are 16yr old chilean girls bookmarking on delicous?)
Either way it sounds like you are setting up a learning relationship with the students. Keep it up! And post the results!
-jeremy
Patrick
Stroll on over and visit Patrick
March 1, 2007

I hadd lunch with the guy in charge of our servers and is one of the principles in our web site management. I explained this dual-path structure idea to him and he sees lots of potential. It's the dual path approach that makes this so exciting. You have your existing same-ol structure in place. You add hooks to it that allow for tagging as well. I would think we'd allow tagging only after a login to the university portal. That unlocks the tag capability. Perhaps you have certain pages outside the login that show the top tagged pages, most popular, most recent, whatever. Perhaps you start with a list of tags, but allow users to create their own, too. Talk about some interesting usage reports that can be generated...
Anyway, good ideas and worth mulling over. When you work at a place with a continual source of youth, you have to embrace change or be made irrelevant.
Luke Wertz
Stroll on over and visit Luke Wertz
March 1, 2007

>> you have to embrace change or be made irrelevant.
The point is though that a lot of youth -- and I do mean A LOT -- simply use the Internet to play Homestarrunner and check their email. I'm guessing that the average student coming into Asbury won't understand what tagging is or how it works.
Jeremy is dead-on right though -- if you can find a way to spur people to tagging, then yes, it is a good idea. I'm not opposed to the idea of tagging, I just don't think that it's very practical in this style of deployment. Maybe it wouldn't be a bad idea to have Resident Life slip in a question or two in their next mass-survey (they do those about twice a semester) to get a very rough idea of where your current student body is technologically. It might save you lots of work (in either direction!) If I'm completely off base and like, 98% of the students are using tagging on other sites, then you know that the work you need to invest in the whole "teaching" side of it is greatly reduced since they'll already be up on the advantages, etc. :)
I love these "idea" conversations. It's very helpful to evaluate and re-evaluate ideas of user-interface design and information architecture theory!
Jeremy
Stroll on over and visit Jeremy
March 1, 2007

I too enjoy these types of discussions. I dont know that I see them very often, So I wanted to thank Chris from bringing it to the table.
I say go with it Chris... Take it to the highest level. You may alienate a few, but they will come around in time. Especially if they have to use the system from time to time. Teach the masses, to teach themselves!
skippy
Stroll on over and visit skippy
March 1, 2007

As the recent inheritor of a legacy departmental website at a gigantic midwestern University, I'm watching this discussion with much interest.
I've never been keen on tagclouds -- they seem like more clutter than useful navigational aid. But the idea of prioritizing links based on historical traffic is not a bad one.
Chris, do please share your experiences as you move forward with this. Share also any code you develop! ;)
Sebastian
Stroll on over and visit Sebastian
March 1, 2007

Luke Wertz said:
I’m guessing that the average student coming into Asbury won’t understand what tagging is or how it works.
I totally agree with you. But there are two things to consider. First, it's a solvable problem to design the tagging to be as easy as possible. Second, and that's the nice thing, the average student doesn't have to fully use the unconventional approach.
In my opinon it's enough when the 5% geeks tag everywhere they want to and the rest can benefit through a search engine.
Another advantage of this approach is that we, the web geeks, have the feeling of being able to change something, make it better. And when the demanding expert group is satisfied, chances are that the rest is too.