Re: [Usability] Tab implementation review

[Jef Driesen <jefdriesen hotmail com>]

Kalle Vahlman wrote:
> 2009/2/9 Allan Day <allanpday gmail com>:
> > I've been working (on and off) on reviewing all the bugs and other
> > discussions on tabs in GNOME [1]. It's reached a fairly mature stage
> > now. I think the review contains useful information about how tabs could
> > be improved in GNOME. Any comments or suggestions?
>
> Not to encourage innovation or anything, but according to this study:
>
>   http://scienceblogs.com/cognitivedaily/2008/12/casual_fridays_whos_tabhappy_a.php
>
> most people have less than 7-8 tabs open when they browse the net (IMO
> the number one tab use scenario really).
>
> And there's a simple reason for that; tabs just don't work when
> there's too many of them (ie. more than what can be reasonably
> accomodated by the window size).
>
> So rather than reinventing tabs and figuring out clever ways to make
> tabs work, in my personal opinion the same effort would do much more
> good in designing and implementing a better way to handle multiple
> views (eg. improving window management or using a canvas approach)
> than what tabs can offer...

When doing software development, I usually have many files open in my 
texteditor (gedit). With such a large amount of tabs open, the tabbar is 
almost useless, except for tabs that are located closely together. But I 
have to admit that I really like the gedit sidebar that shows a vertical 
list with all open files.

Compared with the tabbar, this sidebar can show a much longer list and 
its width is usually larger (and can be enlarged as necessary). On top 
of that, it supports search-as-you-type. For many documents, this is a 
much better user interface than a tabbar. I think other applications 
could benefit from using a similar approach too.