Posted by gaele on August 15, 2008 at 11:34pm
Forget about the user. There is no average user. "The User" does not exist. There are different users, with different goals, different backgrounds and different levels of experience.
And forget about usability. From Matthew Paul Thomas' blog:
What do you mean by “usability”?
Every so often, in a software project’s mailing list, bug tracker, or wiki, I see someone suggest a change they claim will “improve usability”. This is an unhelpful choice of words.
It’s unhelpful because usability has multiple components:
- Learnability — how easily a beginner can use the system, and how easily they can become an expert.
- Efficiency — how quickly people can achieve what they want.
- Memorability — how easily people can remember how to use the system or feature, after not using it for a while.
- Safety — how rarely people experience errors, and how easy it is to fix any errors.
- Satisfaction — how pleased people are with the overall experience.
Sometimes a design change may be entirely beneficial. Other times, the change improves one or more of these usability components at the expense of others.
Comments
Target
Put another way, "usability" often targets specific user groups. An improvement for website designers may not be one for an end-user. One of the main reasons why usability is so difficult in Drupal is that everybody uses the same admin interface. Of course some parts are only used by power users, but a lot is user by all the users. How can you design something for "anybody"?
Design is a balance between the components described above. End users probably need learnability and safety the most (GUI and alerts). Power users need efficiency, even if it comes at the expense of learnability and safety (command line: no undo, no warning).
Do we need a different approach, like offering multiple interfaces for the same task depending on the role/profile?
How can you design something
1. An adaptive interface
Think about help texts and descriptions turned on or off. Or default/advanced settings invisible unless you want to see them. Not bases on role/profile, but based on user preferences.
2. Abstraction
Take for example the Drupal Welcome page ("Welcome to your new Drupal website! Please follow these steps to set up and start using your website:"). This page is great for learnability for beginners. On the other hand it is bad for efficiency for experienced users. For them it only gets in the way.
Now think about an experienced user installing a new module. A page guiding the user to set up this module might be great for learnability for this experienced user.
This suggests a tutorial module, not just a welcome page.
Edit: Please note that Matthew Paul Thomas has the first one wrong. Jakob Nielsen says: "Learnability: How easy is it for users to accomplish basic tasks the first time they encounter the design?" He doesn't mention "beginners".
Adaptable
Ooops. Make that adaptable, not adaptive,
And by adaptable I mean easily adaptable. Not going to the modules page and enable/disable a module.
I'm not a fan of lots of
I'm not a fan of lots of guidelines and tutorials - I think they just cover for a bad interface. And from what I've seen, people don't like to read instructions, they just want what they're using to work.
I don't think it's necessary to dumb Drupal down for a beginner either (I tested web developers in the UB study with 5-10 years experience who couldn't figure out the Drupal admin interface). If an admin interface is designed around the tasks and needs held in common by all Drupal users, all should find it easier to use.
I think user research to find important tasks in common, and an interface designed to support those real user tasks (rather than just a list of everything available) is a good approach. I'm working on putting together an independent study for the user research part at the moment, input is most welcome.
"I tested web developers in
"I tested web developers in the UB study with 5-10 years experience who couldn't figure out the Drupal admin interface."
At a large Dutch media company, about 7 fulltime editors, use Drupal. They don't find it hard to do so. They do find it hard to wrap their hands about oddities, wich are all there because of weird specifics for that specific site. Things like "How do I get last months friday-night-video online in tomorrows video-feed". We did not spend time+effort on those specifics. But we did spend time on interfaces, cases etc for our cases.
This point out two mayor issues with your concerns:
1. There is no "the Drupal admin interface". Once you get your case rolled out, finetuned and so forth, you should have interfaces specific for that case. I.e. we have several dashboards, batch-moderation, quick-views and so on developed, because that is what our cases need.
2. There are no "real user tasks". Or, actually, any task is one. Define "the user" in drupal: is that an admin? an editor? The development-team? the garage-startup-proof-of-concept-maker? The lonely blogger? If you look at Drupal as a development framework (which many people do) you need tools&stuff (modules, widgets, manuals, guidelines) to improve stuff for your cases. For your use-cases. For your users.
http://www.webschuur.com | http://bler.webschuur.com
Lets be realistic
Beccascollan : We are not dumbing down Drupal, we are making it more clear for beginners with a background in web development not neccairly complete beginners its unlikely that they need Drupal for there solution(Usefulness). Can you send me some of your setup work already - I did user research for the redesign of drupal.be. I am a bit worried about who you are researching upon.
gaele: Let me put it this way, it ain't going to happen. I think everyone who is working in the line of usability or is interested knows what you said (or atleast they should). But its hard to grasp the differences between specific user groups because it depends on so many factors that even the most tuned out usability research can only get you to the point of doing things how a user expects to do it (which is very far) but as we are working on cutting edge technology we will always be faced with crafting new experiences.
I agree that Adaptive Interfaces is the key to crafting an experience that adapts to the context of the user, but its unlikely that we will get far enough to see this go live, since it takes quite some mindset and research to get it "right". What is more realistic is that we will see this in instalation profiles, that they are becomming so good for there target groups that they move towards changing Drupal core from a visual perspectice.
Everybody likes the idea of everyone, making it work for grandma and for chx but its unlikely to be done with one interface. You can focus on the task, the activity or the user as a whole, it's all good depending on your insight. The problem is that not everyone shares the same vision, which is likely to be more important then any research will give us. Dries his audience is likely to be completely diffrent from a big group of developers, we have to sync this and be a bit more realistic in our users (Like Becca said, people dont like to read instructions - we shouldnt be solving interface usability issues with documentation and instruction text).
Not sure if you saw my talk on Drupalcamp Colorado but I coverd most of these topics with some hands-on tips, I wrote more about adaptive interfaces here : http://www.bojhan.nl/adaptive-interfaces/ and http://www.bojhan.nl/adaptive-interfaces-part-2/
Hi Bohjan - my course of
Hi Bohjan - my course of study is under development at the moment, and it will be user research on how people use/perceive content management systems - not necessarily just Drupal, but I'd like to include Drupal and will be happy to share my results. I was just planning on recruiting people with high computer literacy (great-grandpa does not necessarily equal a new Drupal user, someone who needs to make websites is). It's a selfish endeavor - I'd like to do an intensive study in user research and get more practical experience in it. I'd love to hear more about your research and what worked/didn't work, and will take any advice that might make it worthwhile to Drupal.
And not accusing anyone of trying to dumb down Drupal - it's an attitude I've encountered both against usability and an interpretation of it, but people aren't dumb, interfaces are :)
For Drupal it might be
For Drupal it might be interesting to explore the different type of users, going more towards people who want to achieve more complex tasks with websites. Which is more likely to match Drupal's audience then Joomla for example.
When it comes to the word dumbing down, I just think it makes little kitties cry. If we use it, developers might use it even more - so therefore I always want to point out we're trying to make it more clear (which has a completely different meaning).
On a note of contribs, I do think they can be crafted way better when it comes to the usability (if they only considered of importance - more then just adding UI functionality and calling it a usability improvement). After we got beyond the step of adding stuff to increase usability, the problem becomes finding and testing those who use the module (UTS might solve this issue). We probably get further by providing developersg uidelines/pattrens that they can grasp and hold on to, you can always say its all about the context but it helps put the focus on really finding out about your users.
We have the luck with the UTS that it's target audience simple. It is webchick, which to me makes a lot of sense as she is interested in usability but doest know exactly how to do it and requires some help from the interface to set a study up, its probably harder to find such a person then it is for contrib developers to define them.
While creating the interface
While creating the interface of a contributed module, it would perhaps be best to focus on the lowest common denominator and make the interface as intuitive and easy to pick up (learn-able) as possible for new users. I prefer the term 'simplified' over 'dumbed down' as dumbed down means an interface/function has been simplified more than necessary to increase usability.
An adaptive interface could take longer to implement (time not only researching multiple user requirements, prototyping such interfaces and implementing them) than implementing the functionality of the actual module(s.) As of now I see adaptive interfaces/expert oriented interfaces existing solely in custom install profiles and/or custom modules whose target user would most likely have a certain level of proficiency/training using a CMS and as such the interface can be built while focusing on the interface needs of such users.
While creating the interface
Do you mean users who are new to Drupal? Or users who are new to the module, but who may be proficient in Drupal?
In the case of contributed
In the case of contributed modules, I would lean towards new to the module but not necessarily new to Drupal.
Core modules/interfaces should however, imho, cater to users who are new to setting up websites. Its a lesser evil to have experienced site builders click through interfaces designed for new users than for new Drupal site builders to be overwhelmed and confused by an interface that can only be understood by experienced site builders.
Dumbing down
From TFA:
I was looking at creating a
I was looking at creating a selenium-ish, zbrush-ish interactive tutorials run by JavaScript of course. I never even got to the prototype stage but I think it would be a great tool if created properly.
Tj Holowaychuk
Vision Media - Victoria BC Web Design
Victoria British Columbia Web Design School
Tj Holowaychuk
Vision Media - Victoria BC Web Design
Victoria British Columbia Web Design School