Recently in Usability Category
If Jakob Nielsen is writing a usability article, we know Twitter is now 100% mainstream. Interestingly, this article though only covers release of corporate news information with tips for both writing punchy copy and for timing (9:01 is better than 9:00).
A little chillingly though, he indicates that Twitter "decay" (when people stop clicking) is fast. Twitter updates so quickly that messages are very quickly lost in the shuffle. As Nielsen comments "Once [the messages] scroll off the first screen, they're essentially 6 feet under." As a result, he still recommends e-mail for messages with a relatively long life span so that users can sort through their "pile" at a later date. No argument from me.
I like to troll for new academic sites in my various lists, especially the Internet Scout Project, but I find that a large percentage of online archival repositories are so unusable I've stopped adding them to my own links library. They are just more trouble to use than they're worth ... which is sad because I'm bypassing great information on civil rights, diversity, Pennsylvania history and more. And I bet most people who aren't professional archivists/scholars are too.
There are a lot of symptoms of an unusable archive (I'll get to those), but I think the main problem is that the content experts are so focused on "metadata" that they've forgotten the context. The result is that you pretty much already have to know what exactly to look for before you can find anything. And by exact I mean keywords like "Jackson, Andrew -- 1767-1845 -- Innauguration, 1828" - got that?
The Quest for Ladybird Johnson
So to clarify my angst, let me pick on a site which has some great resources, but isn't quite so user friendly - the Library of Congress Portraits of Presidents and First Ladies, 1780 - Present. First, let me say that my inner instructional designer salivates because I am fairly confident that somewhere in here is a public domain photograph of President Bush 43 in here somewhere. How many bloggers need a legal image of President Bush? Or President Clinton? Or maybe a student would like a picture of Abraham Lincoln? I do believe this is our tax dollars at work.
Since I'm not a professional historian, I'm really hoping there's a list of Presidents in chronological order (or maybe a second alphabetical order, just in case). Then I can find my target and download my image. So let's try the Browse link.
As you can see I do not get a list of presidents, but a list of all keywords (president, location, event, etc) in alphabetical order. A lazy searcher (the vast majority of us) will stop here. A more persistent searcher will notice that keywords include things like "Adams, Abigail" - meaning that if I target the approximate location of the presidential family name, I may get a hit.
So I will actually look for Ladybird Johnson, which looks to be in the "Grant..." to "Photographic..." range. Fortunately there is one entry for "Johnson, Ladybird."
And finally, I get my photo (very striking if I may say so), along with lots of supporting metadata, like the date (interesting), call number (will I be going to DC anytime soon?), the medium (a photographic print - duh) and digital ID (I may actually need this if I lose the URL), and, best of all, alternate subject headers (if I found it, do I need these now?).
What I don't find is any information on what Lady Bird did, what her husband did, when she lived or died, when she lived in the White House or even a link to any sites a historical society might maintain. I can infer that her husband was Lyndon Baines Johnson, because she is "Mrs. Lyndon Baines Johnson", but not all readers are good "inferers."
I'm sure this information is useful to the LOC staff, but I'm not sure about the rest of us. They do say that the Reproduction Number can be used to order high quality prints, but this is really not clear on the page as is.
Spelling Out the Steps to Unusablity
I'm picking on the Library of Congress, but it really is a common problem. I cannot tell you how many sites:
- Classify holdings by media type (do you want an manuscript, illustration, photograph, video or audio?)
- Give a catalog number, the original archive source or alternate keywords, but only minimally describe what the "artifact" is or how it fits into the larger historical narrative.
- A seriously hardcore archive site will scan the original, but hide/skip a text transcript. You don't have to be be relying on a screen reader to realize that reading a PDF version of a mid-19th century print (with red marks) is not the way to process content on the Web.
- And the most hard core of all - the one with just the search box. Because I already know which piece of 19th century tobacco advertisement I am targeting.
Honestly, if I need a free photo, a lot these sites might be very useful. But I feel like I'm missing so much more in all that metadata.
I should add that if you really want to make an educational archive especially unusable, you'll have to add these steps
- Call the list of external links "Resources" (it's more academic that way)
- Classify artifacts with abstract scholarly categories like "Discover" vs. "Learn" (not of all this can be bad translation issues).
- Mention all sources of funding on the front page (that's why they give out the big bucks), but skip the table of contents, integrated tutorials or anything that a non-professional could use.
I think I'll be visiting Wikipedia now. They might have a free photo I can use as well.
Sites that Do Work
You can build archives/educational sites for non-scholars. They key is to use comprehensible search terms and put information into more context. Some of my favorite archives.
- The Quilt Index - Usable Browse categories, and photograph metadata tastefully submerged under full record link.
- NASA Images - recategorized for non-astronmers with excellent descriptions
- Paleography Tutorial - It's a tutorial, which is what most of us need.
- Met Museum Timeline of Art History - It shows only selected artifacts in browse mode because it understands that most non-scholars can only absorb a few at a time.
- Library of Congress American Digital Collections - the entry pages to the various Library of Congress collections is improving, but the collections themselves are still of mixed usability.
I live in hope that I may learn some of these lessons myself.
From the Penn State Web Standards list
http://deeplinking.net/paper-web/.
For my own reference, this site has some good examples of different interfaces sketched on legal paper (and maybe some cocktail napkins). Also has a video showing a usability test with a paper set of paper prototypes.
While I'm at it, I should add this link from A List Apart
http://www.alistapart.com/articles/sketchingincode
Interestingly the author argues for a "medium quality" prototype because wireframes may be so bare that users may not be able to infer functionality. FYI - The problem with "professional" quality is that users get distracted with suggestions on fonts and colors (been there, done that).
I've had some success with the "mid level", but I've seen projects where the bare paper does work, so the key seems to be keeping the concept minimal enough for users to focus on function and not fonts.
At the Web 2008 conference this year I was able to listen to the keynote from Steve Krug, the author of the classic usability book Don't Make Me Think.
The one take-away I got from that session was that modern users expect the Home Page to kind of be an orientation page to the site (or part of the site). That can have several interpretations, but to me this would include linking to a table of contents, a short statement of the purpose of the site, an explanation of some tools , one or two unique features and who to contact for more information. Interestingly, the home page is likely to be the second or third page visited (after the content page Google found for the user).
It reminds me of another metaphor heard from a usability expert of comparing the home page to an entrance of a store or office complex rather than the cover of a book. That is, when you enter a new store, you may be looking for aisle labels and maybe some hot sales. You may or may not be interested in the perfume lady.
P.S. My other impression was how accurate his observations of the higher education Web culture. In fact, I was wondering if he had sent an astral projection of his spirit to some of the Web design meetings I've been. Spooky
I just ran into the site German GROW, a site with cultural links for German teachers at http://www.aatg.org/content/view/255/ It's run by the AATG (American Association of German Teachers).
Like the icons? It looks very Web 1.0. But click each link and it actually takes you to a del.icio.us page that the AATG has set up. For instance the Musik link goes to http://del.icio.us/GermanGROW/Musik As you can see, the AATG has established their own folders within del.icio.us.
This is an interesting design which I like because the interface is intuitive to a Web 1.0 educator audience, but actually leverages Web 2.0 technology. If you do understand del.icio.us, you can go directly to http://del.icio.us/GermanGROW/ and even suck in the RSS feed. You can even add your own resources if you know how to tag.
But if you haven't had a chance to get acquainted with tagging yet (and many instructors are still in this position), the Web 1.0 interface will still let you get to that content without learning a new set of terms. As my colleague Tim Perry noted recently, "RSS remains an elusive concept."
An interesting case is Studio 204. I must say it is a well-designed site and nowhere does it mention that it's tied to ETS, TLT or even ITS. It just says it's a video studio any student or instructor can access.
However...in order to access the site, I MUST know it's a part of TLT because of the URL http://studio204.tlt.psu.edu/ (see that little "tlt" between "studio204" and "psu.edu"?). My first guess was "studio204.psu.edu", but alas I got a 404 error.
I'm really not trying not to play a blame game here (in fact I can hear some of you grinding your teeth an muttering "I know I know"). I also know that you do have to go through "channels" to get a URL like "studio204.psu.edu" - although I also know that it's not especially burdensome.
But if you're ever wondering why people can't find our valuable resources - maybe it's because we've made some of our URL's more exotic than people can digest.
I regularly read the Jakob Nielsen usability newsletter, and this week he presented results from some eyetracking studies from the U.S. Census Bureau site. In the scanning images, red shows frequent hits,
http://www.useit.com/alertbox/fancy-formatting.html
The point that struck me the most was that users were attracted most to navigational elements - either the search box, the left hand navigation buttons, the category links or the links list. Interestingly, the users skipped over the red text with the population countdown and the bottom Annual report link even though they were meant to be highlighted.
According to Nielsen, anything that looks like it could be an ad is typically ignored, so too much highlighting can be just as bad as too little. In this case the red text combined with the large font made the census count look a bit like an ad. Even worse, it's "not navigation", so users are really not paying attention.
I suspect that red text is most problematic on a home page where lots of other navigational choices are presented. It would be interesting to see how users react red text on a page with more information.
