Thursday, June 07, 2012

User Instructions For USB

User Instructions:  If you support a large number of users, one of the hardest parts of any project is creating documentation.  If it's more than one page, no one will read it ... even if it's required for their job and something they should know.  I'm sure there are many techniques, but I have found instructions work best when they have screenshots and arrows and post it notes over the top.    Here is my first draft on the notes for how to interact with USB sticks and our software.  The left column is the new UI which eliminates file management almost completely.  A few clicks and it's into email and already sized.

I got about 5 people that indicated they would test all the new USB code and should be getting feedback over the next few days.

I also took the time to create a one page cheat sheet related to using the desktop.  There are a lot of keystrokes and shortcuts and if they are not known this lowers productivity; or they say they want another operating system because of memorized technique -- be it Windows or Mac.  The items mentioned are to assist with issues that cause user frustration.  Computers do really cool things, users just often don't find the steps intuitive..  User top issues:

- I find it hard to navigate to folders and remember where I saved my documents  (not one person has ever figured out the GNOME shortcut concept in the file manager without training)
- I need help finding those documents, and also documents that other employees have created.  Where did they save them?
- I saved a document last week, but forgot where it was saved and the file name.
- How do I change the speaker volume?
- We have to plan for the weather, how do I easily see the forecast and radar?
- How do I find other software applications that are not shortcuts?
- How do I check for new email messages quickly?
- I want to just type in the file folder instead of having to click in GUI mode. (not one person has never figured out that you can type in / on the wallpaper to open an entry dialog).
- How do I lock my screen easily?
- How do I take a screenshot on a timer, so that I can capture drop down menus?
- How do I take a screenshot of just one window, to avoid having to print the whole screen?

This one page document went out today and already people are trying new things.


Anonymous said...

I'm sorry, but this documentation is pretty unclear. Also, it seems that a piece of software such as this one wouldn't need documentation if it had a UI that was a bit less messy.

Dave Richards said...

The culture of some people is that they'll use the lack of documentation as a way to not do work. One could design the perfect UI, and some would not use it because they don't want to learn. This isn't a home user that is doing what they enjoy; there are people working that literally do not want to be here. There is a huge bell curve of skills spread out over almost a thousand people; so no documentation or UI will work for everyone. These types of handouts have been effective and yielded results here.

Anonymous said...

I don't mean to sound disrespectful, but you need to get a UI and UX design person to help you with that.

You are trying to help people who struggle with the most basic tasks, yet you are offering them solutions for advanced users (keyboard shortcuts? changing volume with mouse weel?).

Why is it so important to tell users that the search tool is called Beagle? What is the point?

If the problem is the fact that "users just often don't find the steps intuitive.", then the solution is to make those steps intuitive. Creating documentation will not make them intuitive steps; it will just make them non-intuitive documented steps.

Anonymous said...

I have to agree with the other Anonymous here.

Dave, I've been following your work from Planet Gnome for months, and I've never understood what you think is the benefit of this massive amount of work that you must be putting in.

This is a slightly rude rant and for that I apologise. Perhaps I missed the blog post where you said what it was all _for_.

My philosophy with any software installation is that defaults are usually best. They may not be perfect, but the benefit of not having to tweak things when you re-install, not having to remember how things work when you switch from one system to another, not having to train people, outweigh them.

Why can't you present your users with a standard Linux desktop such as Ubuntu? Train them to use that and they will be happily using it forever more, wherever they encounter it. They may even install it at home.

Ubuntu already has user interfaces for handling USB devices, for resizing photos and so on. If something's missing, perhaps write some small app to do it and maybe contribute it back to Ubuntu or Fedora or whatever.

Some of the screenshots that you show on your blog boggle my mind. They look like flight traffic control centre software. So many options and buttons and labels, and - I know it's subjective - but ugly as sin. None of your UI follows any kind of theory of UX at all.

Aren't we moving towards commodity systems, bring your own device? You're building a custom terminal system here.

Most of all I pity the guy who comes after you and finds this huge indicate custom system. If you got replaced tomorrow, I can imagine the next guy just scratching his head, and then throwing it all away and doing a simple install of Windows or Ubuntu.

I'm sorry, I'm sorry, but I just don't get why on earth you're doing this. Please show me where I'm wrong as you seem to be putting so much work and thought in.

Dave Richards said...

I don't ever take comments in any other way except opinions, which I value. There are things unmentioned but should be clarified:

- This blog is for customizations and things that I have done that might be of interest to the GNOME community. These UI tweaks do not consume the bulk of my time. Would anyone want to read a blog about me spending two hours with a debugger on Evolution trying to figure out a crash? Or tuning a post office, or testing a new web site or keeping up with all of my bug reports in the various products? I also handle a good amount of escalated calls from users having problems and go in and fix issues. I also do all of the other admin functions such as tuning and watching for problems and setting up new servers. The point is that these dialog changes are just what I post, but I'm not spending 8 hours a day working on them.

-- These mods are posted too to give the GNOME developers thoughts and ideas for future changes that have been tested for enterprise use. Customizations are made only when we have downstream problems because of design in regular packages. If some of these ideas are integrated, that's great and then customizations would be minimized.

- And that leads to the most important issue, using 100% stock solutions on Linux: All one has to do is find out how many Government agencies have done this. Maybe a dozen at the most in the US? Maybe not that many! I can virtually guarantee that if we deployed stock desktops, that we'd be running on Microsoft Windows within a year. Let me explain, many users fail epically on Linux, Mac and Windows. But when you fail on Windows people assume that it's a training issue. But if they fail on Linux it's the operating system to blame. Failure would lead to complaints and complaints would bubble until this technology is removed. So it's a mind share battle and not a technical one that must be addressed.

-- Generally most of GNOME is pushed as-is and works great, there are just some rough edges related to files. YES, it works the same as WIndows and is just as easy/complex...but that's not how people view it. People literally believe if they shoot 10 megapixel pictures and insert it into a PDF as-is that it would be 10K in size on Windows because somehow it would just magically work that way.

- I met with the UX people a few years ago in Boston and will be meeting with them again in Orlando this fall. If Linux is equal to WIndows it fails, it has to be better and a small amount of time spent on new ideas might yield better designs from those that know better.

- The professionally designed software products sometimes have huge usability problems with users whether on Windows or Linux. The UI probably could be better, the look and feel of these few add-on applications are working because I can see the increased productivity and lowered support calls.

- There is an opt-out feature on the Desktop that allows you to turn off these MIME and file helper dialogs and only 5 or fewer people out of hundreds have done so. When users have to move around information by hand (as they would on Mac and Windows), they fail.

- Related to it looking like we are customizing everything, there is a lot here that I haven't really focused on in terms of the blog. All of the software on the destkop, 100s of applications runs as is (good or bad) and unmodified.

- BYOD is coming and very probably will be in the form of tablets. We are working hard at supporting them in a way that complies with records retention laws for Government. We will support email and other tablet functions, but there are still hundreds and hundreds of users that need a very simple desktop to log in and pick up their email, and type a few documents and do some printing.

Passionate comments are good, and we are all working for common goals. If we reach a point that stock GNOME can be deployed, that's a great day for me.

liberforce said...

I mostly agree with the other comments...

Users don't read popups, error messages, and printing a whole sheet of instructions to teach them why they're wrong is IMHO pointless.

Less is more: put less information, but only the required one, not more. You "USB detected popup" has way too much text. One sentence telling that an USB stick device is available, and both of your buttons would be more than enough. If you want to be more verbose, just add an help button, with more details, but you should hide what users will need in 1% of the cases.

Then your pictures navigation UI is terrible for a novice and for an experienced user. You're basically recreating a wheel, but a squared one. F-spot or Shotwell got it right by using a timeline widget. Everyone has seen that at school, and that's an intuitive way of moving in time. You could use the remaining space to add shortcuts if you really want them (last week, beginning of the year, etc.), and again, an help button which will launch a documentation browser. Even if you don't like a timeline, or find it too complicated, you should at the very least put your previous/next day buttons near the date, not near the pictures.

As for messages like "sustainability for app a or b"... You should think of patching these apps to warn the user when the pictures are too heavy (and provide a way to resize them on the fly). This would be a good idea and could even be added upstream I think, benefiting to every user of these apps.

Customizing an environment to that extent is a lot of work, but from the exterior, it feels like you're giving the wrong solution to the wrong problem.

Anonymous said...


I would create videos on these questions for the benefits for future users.

Also, do you make use of an internal wiki?