See usability-design [not available anymore] on forge.
ISO 9241-11 definition of "usability": effectiveness, efficiency and satisfaction with which a specified set of users can achieve a specified set of tasks in a particular environment.
The HCI team is responsible for two things:
- The usability of 4.5
- Future usability, preferably embedded into the development culture.
This page represents the working area for the HCI team. The initially published information here is a summary of the first 3 weeks of mailing list activity, collected by the interim team leader, Kasper Skårhøj.
See all pages that belong to HCI [deprecated wiki link].
I find it necessary to define what the focus of our group is since some suggestions easily fall into other categories. The highlighted bullet is our focus:
- Architecture => go to "5.0 Core" team
- Improve ease-of-use, newbee appeal, human-factor => 4.5 (thats us! Our criteria!)
- General features and nice-to-haves => Any version as an extension or core implementation through bug-tracker)
From the Roapmap of 4.5 (May 2005) you find this nice graph:
From the R&D results in May 2005 you also find these perspectives on what "ease-of-use" means:
- consistency
- transparency
- unique, descriptive names
- user/task centered options
- proconfiguring template users as editors, administrators, developers and as beginners, advanced, expert
- Consolidate displaced functions in new central modules. (User Manager)
- Substantially improve usability with the help of scientific testing and university experts.
- Create pre-configured users narrowing options down to beginner, advanced and expert level for editors, administrators and developers, making all modules configurable.
(At this point I feel we should only look to this list as inspiration since the most accurate definition of "ease-of-use" is the one we will jointly discover on this team! -kasper)
An objection to working on 4.5 and 5.0 simultaneously has been that we waste time on 4.5. That is not the case if we do this right. Mainly because the value of our work lies in knowledge about working usable solutions for TYPO3, not the implementation itself (generally speaking). This leads to the following system:
- Ideas, concepts, experience, visual elements: All goe directly to 5.0 development
- Technical implementation
- Some will be the 4.x branch implementation only.
- Some will feed into 5.x (thats what we will want to do as the
major development works), for example:
- Forms/Wizard framework?
- New backend interface (requires some MVC?)
- XUL possible
- Other client interaction?
In the process of working with usability for version 4.5, we have the following work areas:
- Usability No-brainers [deprecated wiki link]: There are obvious improvements we can already name from our own experience (these are abundantly exposed on the HCI list already, on mailing lists and we could make a survey for the community). This is where you can contribute your ideas offhand. Also, contributed by creating a Usability Survey [deprecated wiki link] for the community.
- Usability Discoveries [deprecated wiki link]: There are improvements we have not realized yet (these should be discovered through usability tests on the current system). This is where we work systematically with analysis of the current system.
- Usability Innovation [deprecated wiki link]: There are possible innovative ideas to be discovered (seeking inspiration outside our software?). This is where we innovate new approaches
Eventually all improvements needs to be tested (we change only to improve, not for the sake of change).
"TYPO3 has got it's own style and IMHO we should try to keep it that way, since this has been a part of the success story during the last few years. Improving things should never lead to a loss of identity." JoH
I think the sentence above from JoH is very important to keep in mind!
Below is a compilation of statements we can refer to in the discussion of what usability actually is and means for TYPO3:
- Well taught, TYPO3 never is a problem (for any target group?)
- Short term, long term thinking - do we sacrifice long term efficiency with improvements that appeal to novices?
- Is a usable design one that keeps a user in a "Dependent" state - or makes them independent?
- "Give a man a fish and feed him for a day, teach him to fish and feed him for a lifetime": Is it better to teach users the concept instead of the specific action?
Does these paradigms hold true depending on user profile?
- once-a-month user? Should be guided
- everyday editor? Should be trained in the concept so they help themselves.
- Long-term developer? Should be trained
- First-time-installed? Should be guided into the experience.
Usability is the function of. (-Alex/Erik)
- A good working UI
- Splendid teaching
- Teachers who understand the users need
There seems to be two opposite approaches to using software. Some people like a tool where they know all the options and can combine them quickly into solutions. The user is in control and uses the tool as an artist to do his job. He is an expert with the tool. Other people like a guided process which doesn't require deep understanding. They use the tool so rarely or for so diverse tasks that the time consuming wizard process doesn't matter. They can use the software uneducated. Typically they are novices. These two opposites could be expressed with these keywords:
- Soundmixer - state - tool - user controls - educated - expert
- Wizard or bot - process - robot - software guides - uneducated - novice
Does this make sense? (-kasper)
Big topic, gets its own page: Usability Target Groups [deprecated wiki link]
- Old users (so keep icon, location of things, menu to the left as option) (alternative: create the ability to turn off left menu as an option for those who wish to customize the BE in this way? -Alex) (alternative#2: may be too limiting if we have to keep all the same. Allow a new skin and extension that would give the new BE interface look, and so support the new look until major version change -- David)
- Novice users: Intuitive for first-timers
- Expert users: Allow "keyboard short cuts".
- Occasional users: Intuitive at first but can learn power of system.
To start the team, these questions has been asked, trying to address each of the two team responsibilities:
How can we collaborate in the HCI team on improving usability? How can we innovate on TYPO3s usability in a network community? How do we identify true solutions from a pond of opinions?
Input from HCI list, divided into "No-brainer", "Discovery" and "Innovation" categories from above:
- Typical complaints (mailing lists)? (-Gaumond)
- Improvements entered in usability category in bug-tracker so we can vote for them (-Martin Ficzel)
- Usage statistics collected inside system itself (-irene)
- Research: Competitors golden eggs? (-Gaumond)
- Consistency: Once learned functions / concepts should be used throughout system.
How can we (the HCI team) sustain an improved usability and consistency in TYPO3 over the next generations of the software?
Not many reflections on this yet, 2nd priority for now. Might result from experience with 4.5 development:
- Paper-prototyping (-Gaumond)
- Task oriented / Roles (-Gaumond)
- Define "Personas" instead of "the elastic user" (-fabrizio branca)
- View things though the eyes of the target user (end user / developer etc)
- Don't be technical in concept, but technical in implementation.
- Test the usability, improve the interface, test again and so forth.
Personas are fictitious persons, an archetype of some user type, described in detail so a developer can better evaluate the usability of his work in the eyes of the users. Personas should be carefully described. We might be able to do so later from the discussion of target groups above (which is a more categorical concept). Personas might be a part of the solution to long term usability aware development?
When discussing concrete solutions we will need to know what requirements to compatibility we operate within: (R&D committee has final decision on this): All features must run in these browsers regardless of platform (Kaspers suggestion):
- Konqueror (latest stable)
- Safari (latest stable)
- Mozilla/Firefox (1.xx+?)
- MSIE (6.xx+?)
Two-track principle for all basic functionality:
- Modern, Ajax whistle and bells (client validated) [inaccessible, unreliable - but fun and fluent]
- Conservative GET/POST (server validated) [accessible - but boring and slow]
Problems:
- Typical ActiveX/JavaScript/security/firewall settings that could prevent features will require a fall back solutions?
Backend Accessibility (blind people etc)? http://www.w3.org/WAI/intro/atag.php
What methods and actions are we going to go about for improving usability:
- Online-Survey (view draft: Usability_Survey [deprecated wiki link])
- Remote-Usability-Testing (Skype/VNC)
- single observations (see Usability_Discoveries [deprecated wiki link])
Post links to external usability related resources, including a little description of the content:
10/90: Ten percent of features used by ninety percent users? "Apple products typically have lots of features hidden below the surface for the power user to find, but leave the user interface unmarred by frequently unused functions" (http://www.ofb.biz/modules.php?name=News&file=article&sid=351)
Summary of User Interface Design Principles "Excellent article on major UI principles." http://www.sylvantech.com/~talin/projects/ui_design.html
User Interface Design for Programmers "Focused on programmers and Windows…and why consistency and familiarity is so important for users." http://www.joelonsoftware.com/uibook/fog0000000249.html
Golden Rules for Bad User Interfaces "Good to think about…and for fun too." http://www.sapdesignguild.org/community/design/golden_rules.asp
Making a Better Open Source CMS "Some insightful comments by a usability expert." http://www.veen.com/jeff/archives/000622.html
Don’t Make Me Think (Book) "One of the best books on usability for the web. Underlying principle is that a website should be so easy and intuitive for the user that they don’t need to think about it. While the book focuses on Front-end – many principles can apply to Backend." http://www.sensible.com/chapter.html
Designing Interfaces (Book) "Excerpts from book that shows a lot of patterns for “effective interaction design”" http://designinginterfaces.com/Introduction [not available anymore]
uxmatters website A good resource of UX articles not many but good. Aims to produce a volunteer-driven, nonprofit Web magazine that delivers compelling content about * developing effective user experience (UX) strategies * designing digital-product user experiences that optimally serve people’s needs and satisfy their desires http://dev.uxmatters.com/MT/archives/000107.php [not available anymore]
Why usability in Open Source Software is poor I would like to recommend this read to everyone, I have started a small discussion on the HCI list about it as well: http://www.cs.waikato.ac.nz/~daven/docs/oss-wp.html - kasper
OS-X HCI Guidelines Grate Examples! >> http://developer.apple.com/documentation/UserExperience/Conceptual/OSXHIGuidelines/index.html [not available anymore] - Jens Hoffmann
Interface Design patterns "Best guidelines for designers for creating interface design patterns" http://www.techved.com/blog/interface-design-patterns-a-designers-guide
What competences should the HCI team consist of:
- Roles:
- Leader (can organize and lead the team)
- Usability experts/testers (have experience with and can perform usability tests pre and post production)
- Programmers (can implement features)
- Mailing list subscribers (can suggest freely)
- Just members (what can you do?)
Recruitment of test users / process: See "Online survey" thread on HCI list.
Being a member of the HCI team is easy: Just sign yourself up here and subscribe to the mailing list. If you enter your name in this list, be prepared that you are asked to take responsibility on the team.
Enter your [name] - [email(at)domain(dot)tld] - [skills and interest related to usability/HCI team]
- Kasper Skårhøj - kasper2006(at)typo3(dott)com - Initiator, Interim team leader, usability terrorist for years (;-), interest in catalysing on the team
- Cate Deans Smith - cdeans(at)diess(dot)com(dot)au - interested in proofreading documentation so that it is easily understood by the new and inexperienced user, and in monkey testing systems for the same purpose
- Sebastian Erlhofer - erlhofer(at)mindshape(dott).de - interested in getting TYPO3 more intuitive for everybody and especially easy to use for editors (working on Usability_Survey [deprecated wiki link] at the moment)
- Andreas Balzer - typo3(at)andreas-balzer(dot)de - "interested in proofreading documentation so that it is easily understood by the new and inexperienced user[s]" (by Cate Deans Smith, some lines above :) ), and in performing usuability tests with people who want to take part.
- Andreas Foerthner - Andreas(dot)Foerthner(at)netlogix(dot)de -
interest in creating a WebDAV library to primary improve the
usability of file handling and extending that to templates, pagetree,
content or whatever in the long run. So let's say I'm a
programmer ;-)
- Kuckuck [deprecated wiki
link] 10:55, 19 June 2006 (CEST) pleas don't, realy don't do one
at your own, there are two existing libarys, one from ezPublish
and one from PEAR...
- Andreas: We used the one from PEAR ;-) I meant a libray specially adopted for TYPO3
- Kuckuck [deprecated wiki
link] 10:55, 19 June 2006 (CEST) pleas don't, realy don't do one
at your own, there are two existing libarys, one from ezPublish
and one from PEAR...
- Patrick Gaumond - patrick(at)typo3quebec(dot)org - Newbie-friendly approach, wanting an even better TYPO3
- Quentin Dewhurst - quentin(at)gest8(dot)com - LAMP Programmer and Typo3 user with HCI and usability skills, willing to contribute some time and effort.
- Daniel Pötzinger info(-a-t-)poetz-media.de Interested in refacoring of Listview / Help in Discussions of Systemdesign
- Oliver Wand - wand(at)itaw(dot)de - Due to a lot of experience with adaption of large companies to TYPO3 a lot of knowledge about the "daily problems" when it comes to handle content and Backend usability in general. Programming: Looking for the total PHP nerd, you should continue looking :-) Surely I know what PHP looks like and I wrote some customer specific extensions and still willing to deepen my knowledge.
- Zach Davis - [email protected] - PHP Programmer, long-time TYPO3 user, not bad with XHTML / CSS, English editing / writing skills -- looking to help however I can.
- Marion Seitz - seitz(at)cvjm(dot)de - usability testing / proofreading any documents, editing them and translating them into German / defining different levels and needs of BE users
- David Slayback - dave(at)webempoweredchurch(dot)org -- extension developer using TYPO3 for 2 years with strong interest in user interface design and improvement. I desire to help make TYPO3 much easier to use, and am willing to do whatever necessary to help - usability testing, prototype interfaces, thinking through issues, and discussion.
- Lori Feiler - lorifeiler(at)comcast(dot)net - fairly new TYPO3 user (not too tainted by successful use ;-)), 16 years end-user doc writer where I focused on the user experience and improving the user interface. I would like to help as needed identifying FE & BE user issues, usability testing (small samples), navigation in BE, and context sensitive help requirements (for the ?'s)
- Daniel Brüßler - info>>danielbruessler-dot-de - I help with the documentation of some projects and support documentation-writers.
- Jens Hoffmann - jens.hoffmann(@)baiz(.)org - Designer / CSS Coding / providing Usability Ideas
- Patrick Broens - [email protected] - Extension developer / Graphic designer, Experience with interface design. Personal itch to redesign the whole framework user interface, prototyping
- Sudara Williams - [email protected] - Usability counter-terrorist ( ;) and a stirrer of the pot, inspired by the Rails world. Lets rip it out, put it back in, and call it a day.
- Aassim El-Bouchtaoui - [email protected] - Interaction Designer,Enterprise CMS consultant. Why are the bigger commercial CMS even worse if it comes to usability? :-) 'd like to do expert reviews, usability testing and some cognitive walktroughs