Hi, On Mon, Aug 24, 2009 at 7:32 PM, Anca Luca <ancapaula.luca@xwiki.com> wrote:
Guillaume Lerouge wrote:
Hi Fabio,
On Mon, Aug 24, 2009 at 6:54 PM, Fabio Mancinelli < fabio.mancinelli@xwiki.com> wrote:
On Aug 24, 2009, at 4:47 PM, Anca Luca wrote:
To help, imagine the following interaction model (which we envisage but it's not a priority right now): in a wizard step, if the step form is invalid, the "Next" button is disabled (for example if the user has to make a selection, the "Next" button will only become enabled after the selection has been made). Now, given that the "Finish" button would be present on _all_ steps but disabled, I think it can become a little confusing for the user, who wouldn't know if there is something he needs to select, fill in, etc to enable it.
I'd go for minimal UI (i.e. all buttons on the footer strip, but invalid buttons are invisible).
I am not an UI expert but as an Eclipse user this semantics would confuse me. When I create a project in Eclipse, for Example, I have the "Finish button" present on all the wizard forms but disabled (not that Eclipse is the reference for UIs though :))
It becomes enabled only when I can "finish" the creation process early or when I am at the end of the wizard and everything validates.
If you hide disabled buttons you end up with with situations where you have three kinds of "configurations" 1) Enabled buttons 2) Disabled buttons but not hidden (the next button in a wizard when the form is not complete) 3) Disabled and hidden buttons (like the finish button in you previous example)
I don't think you want to make the "next" button appear and disappear as the user edits the form, doesn't you? I think this would be weird imho.
That was also my initial opinion. What would actually happen is that the "Next" button would disappear when at the last step of the form ;-)
As I told you in our voice conversation, we can have a Next button in the last form of the wizard, which would do the very same thing as a finish (except that it would be confusing maybe for the user who wouldn't understand what's the difference between the two, would click a next and have the wizard closed and would think that he did smth wrong).
This means 2 buttons that do the same thing but are not located in the same place not named the same. I can see confusion in our users' eyes already ;-)
The drawback with that approach is that the user doesn't know whether doing an action (selecting a link for instance) will "un-disable" a button or not -> some buttons will get enabled (say, "next") while others won't ("finish").
Another drawback I see, although it might not be that bad because we're talking only about 3 buttons here, is that the user constantly sees things he cannot use (ah look a button "what does this button dooo?" ah cannot use it why?). Why would the "blinking" be a problem? buttons would only change configuration from one step to another. You click a button, you get a new form loaded, with new options to go further. Is this an issue? (I'm no expert, I'm just saying).
You don't get a "new form", it's still part of the same wizard experience. Imagine if Eclipse's buttons got hidden or shown at each step of every wizard, you would find that very disturbing.
Thanks, Anca
In practice there's indeed a "blinking" risk, we might be aiming at the
best
while the good would be enough (yes, that's a rough translation of a French saying).
P.S.: and yes, we might be over-engineering the 3 buttons...
We definitely are :-) Let's go for the basics !! That's what people are used to, it matches the user model... Why would we not want to match the user model? Guillaume
Guillaume
-Fabio _______________________________________________ devs mailing list devs@xwiki.org http://lists.xwiki.org/mailman/listinfo/devs
_______________________________________________ devs mailing list devs@xwiki.org http://lists.xwiki.org/mailman/listinfo/devs
-- Guillaume Lerouge Product Manager - XWiki Skype: wikibc Twitter: glerouge http://guillaumelerouge.com/