In his State of Drupal presentation at DrupalCon Barcelona Dries mentioned usability as one of the key factors in improving Drupal.
One of his practical advises was to do user testing.
So let's act upon that. Pick a friend, relative, colleague. Show them Drupal, let them try it out, and watch and listen them doing it. See what works, what they are struggling with, what their frustrations are. And write it down. Here in this group.
It doesn't have to be scientific research. Any feedback will help.
To structure this a bit: I believe 4 user roles can be defined.
- visitor - someone visiting the site
- editor - someone doing the day-to-day content management
- administrator - someone installing and setting up the site, e.g. creating content types and views, positioning blocks
- developer - theming, coding, doing PHP stuff
(Please note that in official Drupal terminology a user is anyone logged in, and a visitor is anyone not logged in. This is not what I am talking about here. Also note that one real life person can act in several roles.)
I guess what we're aiming at here are the editor and administrator roles. So please keep that in mind, and specify when you're reporting youre results.
One more thing: when we as Drupal developers talk about the "admin" interface, we're talking about the interface for both editor and administrator roles. In Drupal there's no difference. The Drupal interface doesn't make a distinction between editors, administrators or even visitors. For usability this distinction is important, though. Different roles mean different background knowledge and different expectations.
Comments
Wonderful
Fantastic Idea. Actually, I will start to develop a survey similar to Dries's, just exclusively centered on Usabilty. I asked him about help, and he offered the support of the Association. The idea is, to use the same technical framework as he did (which was not drupal), and I think this survey should be published also on other websites than drupal.org. e.G. general open software sites or also well-known blogs that cover roughly the subject would be appropriate. In the end it's just links that point toward a site where the thing is hosted, but that does not matter. If we want to know about completely new users or maybe ones that tried to install and ran off for some reason, we got to go outside our parallel universe... ;)
In Germany I think I'd find some sypathetic blog gurus to help us.
Life is a process
Life is a journey, not a destination
Workflow and roles are everything
In fact, I think that the entire admin interface should be reorganized around workflow, so that admis tasks are clustered around stages of a sites development by different groups
here are the developer phases:
1) Installation. These correspond to much of the "site information" content2) Configuration (managing modules)
this is for the interactive architect:
3) Architecture (setting up menus, blocks, views, etc.)here's where the front end design comes in:
4) Templating (setting up fields, blocks, views, etc. as well as look and feel)these are provisioning stages for readying a site to be turned over to a client
5) Security and Notification (setting the site up to work with various client roles)6) Performance
this is where the client or webmaster will spend most of her time:
7) Content managementIt would also be terrific to have a series of default roles built in. We're finding ourselves building these into every site:
administrator
developer
moderator
contributor
webmaster
editor (or client)
@modulist
Those roles are fine for
Those roles are fine for your sites, but would be inappropriate for a small site or blog.
Setting up default roles would be useful in install profiles, which can be targeted at a certain types of sites.
I like this
The taxonomy of the current Administration menu is somewhat logical, but far from ideal. The groups haven't got any relation to the workflows of installing, setting up and managing a Drupal installation - which is what you use the Administration for.
I'm not sure I agree with your grouping thou, and drumm has a good point on the roles, but the overall suggestion is very good! What groups should be used and what should go where is something a user survey should discover.
That being said, here are my suggestions:
Installation (Admin pages you visit to set up the site and then rarely use.)
Configuration (Admin pages you may use more frequently, and settings pages for most contrib modules.)
Appearance (Admin pages used to change the appearance of the site. Views goes here.)
Content management (For setting up and managing your content. Same as current, except I moved Menus here.)
User management (For managing your users. Same as current.)
Reports (Instead of "Logs", as Status report isn't a log while all the logs are reports.. right?)
Edit: Added some comments in paranthese.
On second thought
I tried this out on a D6 installation and it doesn't work that well.. I'm actually thinking that the current Administration page isn't that bad at all. The "Site building" group contains the pages relevant for "building" your site (the first part of the site life cycle), while the "Site configuration" group contains most of the configuration options aimed at further tweaking of your site. There could be a group for "Look and feel" or "Appearance", containing Theme related stuff, but it would be a very small group.
Something else that could be improved is the placement of some pages. I feel Menu belongs in "Content management", while Triggers and URL aliases should go down into "Site configuration". And if Translate interface belongs in "Site building", why doesn't Langages?
Anyway, this is just me trying stuff, it doesn't hold any ground at all, I'll shut up now :P I'm no usability expert, a thorough user survey is the best way foreward.
+1
+1 for arranging the admin menu around workflow. Currently I find myself repeatedly thinking "Where did they put that menu item?".
I'm sure we can work out a grouping. Personally I'd say 7 groups are a bit too much.
Main problem for me was getting started
I am a highly-experienced PHP/MySQL programmer, relatively new to Drupal. My main problem, in the administrator and developer roles (as defined in the original post on this page) has been in finding information on drupal.org.
For instance, for my first Drupal site, I wanted to create a collaborative Drupal-based site for my high school reunion. I wanted to set it up so that I had sole control over a few pages, and class members could easily find information and memory book pages, edit their profiles, and add to the memory book. Figuring out which modules I needed, how to install them properly, and how to get the site set up involved way more trial-and-error than I would have liked.
Then I needed to create a new module for the site. Figuring out the hooks API and the basics of module development was also a matter of trying to read existing module code, rather than reading a good guide on drupal.org. Once you know the basics, the api.drupal.org site is not too bad, but you can't read an API until you have at least overview knowledge.
It's possible that information is out there on drupal.org for what I needed, but if so, it was not obvious to me where to find it, or the documentation I found was out of date, incorrect, or misleading. It also doesn't help that the drupal.org site is so slow on its page loads. So my overall impression of Drupal is that it is a very powerful site/content management system, but very hard to use, with terrible documentation.
That's my 2 cents...
--Jennifer
Drupal programmer - http://poplarware.com
Drupal author - http://shop.oreilly.com/product/0636920034612.do
Drupal contributor - https://www.drupal.org/u/jhodgdon
I agree
The information is there, it's just very hard to find. It took me a while to understand that drupal.org requires patience and determination if you are to find what you're looking for. I've learned that searching through Google (e.g. "site:drupal.org search phrase") is the easiest way to find something on drupal.org.
But fear not, the drupal.org administrators are aware of this, and there are efforts to redesign and improve drupal.org in the not so distant future :)
user-interface
Also see this: http://drupal.org/node/172681. sepeck has made a great start creating documentation for beginners.
That said, this thread is not about documentation or drupal.org. It is about improving the usability of the Drupal user-interface.
4 types
Last year I also coined a classification with 4 types of Drupal users. I can't find it anymore but it was used in a presentation by Steven in Brussels, if I remember correctly, so the slides may still be around...
Anyway, the 4 types were approximately the same, and mean each of them use the admin interfaces in a different way: Drupal developers (core coders) and website builders can probably use any admin page, editors use far less, and visitors only a few (profile, registration...).
I think most of the "bad" interfaces are website developer interfaces, so they are the ones who would benefit the most from improvements in this area. They also probably use the admin pages way more than any other type of users.
I don't mean visitors or editors wouldn't benefit from improvements in the screens they use! They have BIG usability problems. But the biggest ones are not related to Drupal core yet (image handling, wysiwyg...)
Edit: ah! Found my classification, here on slide 21 and it was a presentation by Neil Drumm in fact:
I guess now with CCK, Views, Panels, etc. we could add the "website administrator" (a kind of webmaster?), as they can do so much without a line of code.
I guess that's more or less
I guess that's more or less the same classification as mine, only the other way around. So let's use these names then. I still believe we should aim at roles 2 and 3.
Yeah
The developers (those who set up the installation) and the editors (those who use it) are the ones who use the administration pages the most, so we can leave out developers (who I suppose are more occupied with the technical aspects) and the visitors (who rarely see much of the admin pages) :)
4 types works for me
Like Neil Drumm points out, the seven categories I posted above are probably too much for a simple blog or quickie site.
His <a href="#comment-18047>four types are a great starting point, and the admin menus should be split up accordingly. But....
Drupal needs to grow up and address the needs of those large-scale projects with page counts in the thousands and teams of professionals working in specialized roles. We've built sites for major universities and corporate clients and the current interface has nothing to do with the right workflow or process for this kind of site. If anything, it scares off the good architects and designers.
So if we don't split the interface up according to job discipline or workflow, then the handbook or documentation should be at some level. Otherwise, there's a book to be written for the corporate publisher or the marketing communications professional. Anyone willing to co-author one? ;-)
modulist@drupal.org
@modulist
Wonderful
Fantastic that things are getting on the move here, now.
First thing for me is still: We should do a survey similar to the one Dries did before Drupalcon. This survey should focus only on usability and find out the major flaws in Drupal from this angle of view.
What we need is the opinion of new Drupal users who have a fresh look. I've got the following suggestion: the Association would help us even financially, of that I'm sure. How if we would swarm out and find new users to fill out the survey when it's finished. To encourage everyone we could give away Drupal T-Shirts, Books and DVDs, or maybe even a free one year membership in the assiciation.
I'm thinking of something like this: One who gathers 5 new-user surveys gets a T-Shirt, for 20 a book or DVD, and for 2000 gets elected President and a private Jet with 5 Supermodels to his service... For women a personal Dinner with Brad Pitt and Johnny Depp, of which to choose one afterwards.
Everyone needs some encouragement.
I will start on a skech of the survey soon, here my ideas so far:
Add more Ideas if you have some.
Would't it be a fantastic goal to make Drupal 7 the "Usability Release"?
But there is an enormous amount of Work to do, four in Usability Respect, Drupal is quite bad. So many things to structure. But obviously we got the manpower now.
Life is a process
Life is a journey, not a destination
Great suggestions
Great suggestions (both above and below this comment)! I'd like a dinner with Angelina Jolie though :)
As for a survey, did you know about this one (http://drupal.org/node/127285)? It started on March this year, but it doesn't say anything about what came out of it, or if it even started. Neil (drumm) looks to be the one behind it, maybe he wants to pick it up again? What do you say, Neil?
There have been other surveys before as well: http://drupal.org/node/126140
Yes, I started that and
Yes, I started that and worked with a couple usability-types, I do not remember their exact job titles, for two days to write that up. I have a few more notes to flesh out if someone is interested. I would love to see the effort continue.
Exploring the fundamental usability questions outlined on that page is important, and I think would be a great long-term asset to Drupal. Asking about specific things being easy or hard gets a decent snapshot of what we might want to work on, but does not help us actually understand users.
I lost interest because:
- I was concentrating on Drupal 5.x. I did a lot of small usabillity reviews as part of that.
- It really is a lot of work to take this from the start to a well-written report which is an excellent resource for usability practitioners and developers.
Let's continue from there
Can't blame you for not getting things done, the usability in Drupal improved a great deal in 5.x! Also, Dries' call for focus on usability really gave this group the support we needed. We should really take care of this oportunity now that the issue has the attention it needs.
We should obviously continue with the interview script and eventually coordinate a survey. Hopefully, someone who knows how to do this will step up and lead this. It takes a lot of work like you said, but I think with the focus on usability, there will be a lot more people helping out than earlier. Advertising the survery (once it's ready) on drupal.org would help a lot!
Two levels
Usability can be improved on two levels :
a low level approach: user testing, gathering data about what people use, which menus, etc... This could help define the new workflow some people would like to see.
a higher level approach: changing tiny but important things in the current interface.
The latter comes from a discussion with Neil in Barcelona. I pointed out some admin pages that are very hard to grok for newcomers (ie: menu settings, primary and secondary...) and wondered how we could change things as most of us are not "real" coders and don't have CVS installed. So he said we could simply file issues, and explain in details what is bad, how it could be done better, or even post a mockup, that way Drupal developers may be more willing to fix it.
The menu settings is a bit like my pet peeve, as it seems every developer friend who tries Drupal asks me how he can set two levels of menus in the header... So I'll definitely have a closer look at it!
Changing tiny but important things in the current interface
You mean issues like this one:
Easier date-time format setting
or this one:
user registration settings: more logical order?
These issues didn't grab any attention at all. What should I (we) do differently?
I'm all about the usability
I'm all about the usability focus... but I have to ask... even though I'm a dude, could I choose to have dinner with Brad Pitt instead?
I think the real benefit from the survey will be in nailing down what common website administrative tasks were just too confusing or difficult. For example... "Were you able to create content?" "On a scale of 1 to 5 with 1 being easy and 5 being difficult, how hard was it for you to post content?" Similar questions for creating user roles and assigning them to new and existing accounts. How about creation of menus? That's gotta be confusing to folks not used to Drupal's interfaces. : P
Once we have key areas like that nailed down with items for improvement on each one, we can assign folks to particular tasks instead of just "making blogging more like WordPress" for example.
(EDIT: My bad, this should've been a reply to the above post.)
Administrators Report of March
Looking at the outcomes
http://drupal.org/node/126151
We can state, that even in the given group, that had
more than half of experienced users or Administaters
Drupal comes out the most mediocre.
Doing a new survey we should take care that we are
able to filter by skill level. It would be interesting to differenciate
the skill groups - most important are the not-so-experienced users.
Life is a process
Life is a journey, not a destination