This is sort of an off-topic post, obliquely related to things geospatial by virtue of the fact that our geospatial products come with help documentation. As LizardTech’s technical communicator, I attended WritersUA 2009 at the Westin Hotel here in Seattle a few weeks ago. The WritersUA Conference is the annual shindig for user assistance professionals (a.k.a. tech writers and our ilk).

Overall I was impressed, and in a lot of ways. There were over 300 attendees from all over the country, nearly 25 exhibitors and sponsors, and dozens of presenters over the course of three days. There was free wifi throughout the entire conference space, a huge Twitter monitor, and best of all, the opening speaker was Scott McCloud, who has written/drawn award-winning books (“Understanding Comics” et al.) and did the famed Google Chrome comic book documentation.

As is probably normal, the sessions were a mix of things that sounded information-packed and edifying and things that sounded irrelevant, although there were surprises. For instance, the closing session on illusion and mental models by a comedian/hack magician turned out to be one of the most inspiring sessions of the whole conference, as was Scott McCloud’s opener, even though you might have thought cartooning doesn’t have much to do with user help. And one session I was looking forward to turned out to be just bullet points that I might have come up with myself. The most useful session in terms of how to do things better was a session on improving knowledge bases for better troubleshooting.

I think I heard the word “cognitive” at least twice a day, which I thought was significant. It tells you how important the mind of the user is to UA professionals.

In my opinion there was a superfluity of sessions dealing directly with the Darwinian Information Typing Architecture (DITA), an XML language for documentation that the government is mandating wherever it can these days, and there was an inescapable prevalence of tool-specific sessions, lamentable only because most were not about the tools I’m using.

Notes on some of the most inspiring (in one case depressing) sessions:

Opening Session
On Monday morning Scott McCloud submitted to a Q&A session in which he described how the Google Chrome comic came about, what it was like doing it, and how the end result was received (wildly successful, borrowed from and parodied all over the Internet).  

The big thing for me in Scott’s talk was the idea of isolation through sequence, which is what distinguishes comics from other drawn art. Few of us would want to turn our help into comics, but the idea of directing the user’s attention by using the same tools the cartoonist uses – removing everything from the user’s focus except what they need to know at that moment – is something that all of our user assistance projects would benefit from. He also talked about how the mind remembers static images more readily than moving ones, and about the notion of “two cognitive altitudes” (context and detail) and the importance of the balance between them.

As for the man as a speaker, I was amazed at the prehensile-ness of his mind. You could tell from the way he listened to questioners and then reframed their questions that his mind was completely wide open like those big dishes in the desert listening to noise in outer space, and when he listens you can see his brain reaching for whatever metaphor or symbol or phrase that will best capture the essence of the conversation. He’s really a technical communicator at heart. If you ever get a chance to hear him, take it.

User-Centered Design of Context-Sensitive Help
Matthew Ellison’s talk was a great intro to context sensitive help (CSH), in which tips or other help-ish material are built into the app, and are either visible to the user as they work or easily accessed by an icon or underlined phrase. CSH is becoming important as UA writers begin to acknowledge that no one reads help unless they are forced to at gunpoint (more on this, unfortunately, below). But more than an intro, it really gave solid best practices. One point that stuck with me was, “focus on answering likely questions rather than documenting the application.” Many times, we tech writers are tempted to write from our model rather than the users’, but again, more anon…

Best Practices for Embedded UA
Scott DeLoach talked about the most useful ways to embed assistance in the UI in exactly the amount and timing (and even phrasing) needed by the user. It seems to be a scientific fact that as long as you don’t call something “help” and it’s right in front of them, users will consult it. Lots of good info here, such as the fact that a novice will eagerly click “Quick tip” while an expert will want to see something like “Tell me more…”. One takeaway: sometimes thoroughness is less important than usefulness. Scott asked us to ask, “What are the things the user HAS to know in order to successfully complete the task?”

Lessons Learned from Research on “Help”
The percentage of users who consult online help and printed help, respectively, are “virtually no one” and “actually no one.” Those are the facts. Professor David Novick of University of Texas at El Paso seemed from the start to want to make us cry, and from the beginning he drilled home the dour and research-supported truth that we’re wasting our efforts. There was no good news to balance this glum panorama, either. It was a dizzying ride through a milky-way of dismal data points, at the end of which he said, “and when they DO use the help, research shows that it doesn’t make users more productive.” We all went out of there and shot ourselves.

Better Knowledge-Base Articles for Complex Troubleshooting
To my lights, the best session of the lot. I’m glad I have the slides, because I could barely keep up with this guy. David Farkas is a professor of technical communication at the University of Washington here in Seattle. He moved through his slides very quickly, rightly using them only as touchstones for the actual content of his presentation, and if you took notes you missed the next thing. David offered lots of good info I will be reviewing, like how to title KBs most effectively, how to structure them, and how to “reduce the cognitive load” on the user’s mind as they troubleshoot problems. He regards the KB’s purpose as a troubleshooting mechanism for “bugs, known issues, things that fall between the cracks” rather than a body of help topics per se.

Magic and Mental Models: Using Illusion to Simplify Designs
Jared Spool’s “magic” presentation on Wednesday was a fantastic end to the conference. It seemed at first to consist of amusing but unsuccessful attempts to read people’s minds, but that was all just a way of demonstrating Kano’s model and the role of delight, which, nutshelled, says that if you get the basic functionality right (things people need and expect), you can add another aspect Kano calls delight. This is a little extra something the user is not expecting. One example he used is that when you attach your iPod Nano to sync it up the image changes to an image of not only the model you have but also the same color. Jared said most users won’t even notice this, but those who do will experience delight.

Jared also explained the difference between the designer’s model and the user’s model this way: none of the Disney imagineers wants a vistor to the haunted house ride to come out at the end and say “Oh my gosh, the speakers around turn #3 were amazing…I have to have that in my house!” The user is supposed to have a model in his mind that says THIS IS A SCARY PLACE, THERE ARE GHOSTS AND HALLOWS IN THERE. (Jared didn’t say “hallows”, that’s my paraphrase.) The designer’s model is different; it’s about wires, mirrors, speakers and other functional components. The point is, we who make products don’t need to explain our model, we just have to make sure the user’s model works.  

In the end Jared pulled off a bona fide magic trick and we were all delighted.  

What I came away from the conference with was a motivation to reduce the “cognitive load” on our users, give them the assistance they need (and only what they need) when they need it, and finally to delight them.

Our customers deserve all that. After all, they have the world on their shoulders.
 

We added floating licenses to GeoExpress a couple of versions ago to give customers more flexibility, but many customers aren’t aware of what floating licenses are or how they work. This Q & A addresses the most common questions we get about floating licenses.

What is a floating license?

A floating license is a license that moves around among the members of your organization as needed. It’s a means of enabling a user on a networked computer to lease or “check out” a GeoExpress license for the period of time they are working. When they quit GeoExpress, the license is returned to the license server and becomes available for another user.

What’s the benefit of using floating licenses? I need to tell my boss what the benefit would be.

Try this on the boss: “Using floating licenses makes for very efficient use of our organization’s software license budget.” Rather than purchase licenses for every single person in your outfit that might possibly need to use GeoExpress at some point, you can buy licenses for just the number of people who are likely to need to use the software concurrently. If that weren’t enough, floating licenses also take a load off your IT administrator, who no longer has to go to each computer to run GeoExpress and write down a locking code. With floating licenses the licenses for all your organization’s copies of GeoExpress can be managed centrally.

Which editions of GeoExpress can we use a floating license with?

Floating licenses can be used with GeoExpress Tools, GeoExpress Standard and GeoExpress Unlimited editions.

GeoExpress Remote Edition is one of those, right?

No. GeoExpress Remote Edition is its own edition and floating licenses don’t work with it. If you’ve got GeoExpress Remote Edition you‘ll need a remote license. Contact your LizardTech representative to get one.

So, how does a floating license work? How do I check one out?

When you run GeoExpress, the software “looks for” a license on the network and if one is available, you’re off to the races. You don’t have to DO anything. It all happens automatically and invisibly. In fact, users at an unfamiliar workstation running GeoExpress with a floating license checked out on the network would not necessarily know they weren’t using a local license. Users generally won’t even have to think about how their software is licensed.

You said “if one is available”. Why would a license NOT be available, and what happens then?

The license server only distributes as many licenses as your organization has purchased. After all your licenses have been assigned, the license server waits for a license to be checked back in before it allows any more to be checked out. If this happens, the next user that runs GeoExpress gets a message alerting them that all the licenses are in use, and they’ll have to wait for the next available license.

What’s a license server? It sounds expensive.

Generally speaking, any computer on your network can be a license server. You just have to install the license server software on it. The license server software costs you nothing. It’s provided for you on the GeoExpress installation disk.

As a user just trying to run GeoExpress, do I have to know which computer on the network is a license server?

No. GeoExpress automatically searches out the nearest license server on your network. However, if you wish you can tell the software where on the network you would prefer it to look by specifying a license server.

Can I also specify that I’d prefer a NITF-enabled license if one is available?

Did we pay you to ask that? Why, yes. Yes, you can.

Where do I do all this specifying about licenses and license servers?

Choose Floating Licenses from the Options menu on GeoExpress’ menu bar. A dialog box appears where you can specify a license server and whether you’d prefer a NITF license.  

With floating licenses, does GeoExpress still get installed on a local computer?

Yes. GeoExpress is still installed on individual workstations, but the workstations do not have individual licenses. The license server stores all the licenses and keeps track of them.

About 250 LizardTech users took our customer satisfaction survey in April. I wanted to thank you for taking the time to tell us about how you use our products and what we can do to improve. Ninety-eight percent of you said you’d recommend LizardTech’s products to a colleague, and as a product manager that makes me about as happy as can be.

Some of you said you’d be willing to talk to me in a bit more depth about what you’d like to see in future versions and things LizardTech can do to make your life easier. I’ll be starting to make calls in early June, so thanks in advance for your help.