1. Borrowing now from a thread in in discuss@ux
2. with reference to three or more relevant mentions of style in ui@ux
3. in particular <http://ux.openoffice.org/servlets/ReadMsg?listName=ui&msgNo=31>, an extract from which I'll place at the foot of this message
4. focusing on part of the goal of Renaissance,
5. expanding that goal a little: what I would _most_ like from a Renaissance UI -- in any form -- is:
* for users to finally _realise_ the
previously unrealised value of features.
(In other words, without insulting users: what they *think* they want to accomplish is twisted out of shape by past experience of poor UIs; in many cases, a best accomplishment may be significantly different from their first imagined accomplishment. This paragraph may not make sense on the moment, I might sit on it for a few days/weeks!)
In my experience, observing a near-universal lack of appreciation of related features, the prime candidate is:
* styles, in particular paragraph styles.
So: without aligning myself with any particular Renassiance UI notions or mock-ups that have arisen to date, I'd like the focus of this topic to be:
+++ how to actively help users to
+++ realise the value of paragraph styles.
Approaches to realisation should probably _not_ include pop up dogs, dancing paper clips or intrusion/pop-up of any kind.
Approaches in Renaissance probably mean arranging the UI, the whole caboodle, to subtly demonstrate the value of applying styles to paragraphs. And/or the value of allowing OOo to apply a style for you.
I'm fairly certain that the designers amongst us have this goal, or something similar in mind but still, I'd like to express this as probably my #1 wish.
Approaches beyond Renaissance might mean (shudder!) a future, more graceful (less potentially irritating) auto-correction of double paragraph marks.
The 'styles' aspect of Jaron's paragraph is IMO implicit within <http://wiki.services.openoffice.org/wiki/Renaissance:The_Challenge>.
(That URL is added to my Diigo for sooner rather than later reading.)
Le 04.01.2009 08:47, Graham Perrin a écrit :
>> help them accomplish what they want to, by providing efficient access to
>> valuable functionality
> 5. expanding that goal a little: what I would _most_ like from a Renaissance
> UI -- in any form -- is:
> * for users to finally _realise_ the
> previously unrealised value of features.
> In my experience, observing a near-universal lack of appreciation of related
> features, the prime candidate is:
> * styles, in particular paragraph styles.
> So: without aligning myself with any particular Renassiance UI notions or
> mock-ups that have arisen to date, I'd like the focus of this topic to be:
> +++ how to actively help users to
> +++ realise the value of paragraph styles.
> Approaches to realisation should probably _not_ include pop up dogs, dancing
> paper clips or intrusion/pop-up of any kind.
> Approaches in Renaissance probably mean arranging the UI, the whole
> caboodle, to subtly demonstrate the value of applying styles to paragraphs.
> And/or the value of allowing OOo to apply a style for you.
> I'm fairly certain that the designers amongst us have this goal, or
> something similar in mind but still, I'd like to express this as probably my
> #1 wish.
In reply to this post by Graham Perrin
Le 4 janv. 09 à 08:47, Graham Perrin a écrit :
> So: without aligning myself with any particular Renassiance UI
> notions or mock-ups that have arisen to date, I'd like the focus of
> this topic to be:
> +++ how to actively help users to
> +++ realise the value of paragraph styles.
I have already expressed some ideas on how we could technically
manage styles, even if it seems that I could not make clear how it
could "help users to realize the value of styles." I will probably
come back on it later, but for now I would like to discuss how users
are (or could be) evolving from "novice" to "master of styles." I
will use many steps, and please notice that for every step, there
will be many users who will never make the following step.
= First step: expectations =
First, I think that nobody will ever use an application such as
Writer or Word without first knowing what it is made for. Probably
every new user of such an application had faced before some printed
texts made with this kind of tools, and somebody told them that it
was made with (most likely) Word.
So new users comes with expectations about what the application can
perform: it can put text in bold, italics, underlined, change font
and font size, adjust margins, make bulleted and ordered lists,
tables, graphics, headers, footers, page numbering, footnotes, tables
of content, etc…
When using the software, they are then driven by formating-related
goals: "I want to put that in bold face", "this is a title, it should
be big, how do I change the font size?" And because the corresponding
buttons in the toolbar are easy to find and understand, they learn
quite easily how to do basic formating.
= Second step: first encounter =
The user then realize that some operations are not available directly
from the toolbar, but he knows that it is possible because he have
seen documents which obviously made usage of that operation, such as
"changing the default bullet in bulleted lists."
He then finds that to achieve that, he has to use a dialog (which is
actually the 'list style' edition dialog). How he discovers that is
not the question here, even if making it easier is certainly one of
our goals. Similarly he discovers the other style edition dialogs,
still being driven by formating-related goals.
At this point he may think "Yeah I've discovered all that formating
options, I'm becoming a master!". But in fact he has a wrong
conceptual model, since he believe that the dialog is just some
interface to apply a formatting (action), while it is in fact an
interface to edit a style object's properties.
= Third step: applying styles =
I suspect that many users begin applying style because of the
semantical properties of some styles. For example, applying a title
style makes the paragraph appear in the table of contents.
But saying "hey, Writer, this is a title" and formating the title may
still be two distinct operations for many users: they apply a style
(without modifying it) for a bad reason, and without understanding it
(bad conceptual model).
= Fourth step: the deception =
Now that the user knows how to do all that formating with dialogs, he
will use it a lot. And he will then realize that it is a boring,
difficult, inefficient and repetitive way of performing formating.
I suspect (by the number of badly formated documents that I have
seen), that a lot of beginners format their titles by repetitively
formating each title. The more advanced ones even use the "copy
style" button for that (I have seen it with my own eyes!), because it
is an answer to the question "I want that text to look like this one,
how can I do that?"
And so, I suspect that most users will discover the use of styles
primarily as a way to avoid repetitive formating. Some kind of macro
(once again, bad conceptual model). And once again, I will not
discuss here how they discover the "manage" tab in the style edition
= Fifth step: separating shape and content =
Only a small fraction of the user may rich this last step, because it
needs to have been taught to think that way, or a long experience in
document editing, to understand the advantages of using styles to
separate shape and content.
Note that this last step is only a methodological one, it does not
need to have a better knowledge of the interface, but to have a
better knowledge in the process of using that interface.
= Conclusion: =
The first issue with styles seems to be discoverability.
In particular, there is no clear relation between the formating
toolbar and the style edition dialogs. So we should blur the
distinction between "formating" and "applying a style". We could for
example group styles together with the basic formating buttons in the
formating toolbar, and present them in a similar way, were the effect
of applying the style is illustrated visually (yes, this is some kind
of formating palette). Thus, we should ease the third step. To ease
the second step, the palette should also contains all possible
formating, including those that are actually only possible via the
style edition dialogs.
The second issue is a wrong conceptual model (actually, there are
many of them depending on the step), which seems to originate in the
fact that formating is perceived as changing the properties of the
selected text, while it actually change the properties of the style
applied to that text.
So, I think we should certainly highlight the fact that every
formating action acts on a style, not directly on the text. One way
of doing this is to have a representation of the definition of the
style applied to the text under the cursor. Formating a text (such as
setting it in bold face) would then have an effect not only on the
visual appearance of the text, but also as a modification of the
To make it more efficient, we should also highlight the fact that
styles are manipulable objects, and I think that reification is a
good way to do it. I we have both a formating palette and a current
style definition, it should be possible to drag styles from each
other, allowing to store the actually applied style in the palette
for latter reuse, or allowing to replace the currently applied style
with another one in the palette.
Of course, it may sound (and be) more complex, so we first have to
check if my concerns are valid, and then if the solution that I
presented effectively reduce it without bringing new issues.
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]
|Free forum by Nabble||Edit this page|