Uwe> Is it in principle possible to have below the main toolbars Uwe> (like file edit etc) two or more lines of icons (instead of Uwe> the one with open, dired etc) Yes and no. No, there's no provision for multiple rows in the toolbar. Yes, you can have multiple toolbars, up to 4, one on each edge of the frame. You can also switch toolbars in place. The VM mailreader actually uses both techniques, providing a second toolbar (by default at the left edge) and switching the *menu*bar in place, providing a button to toggle between the main menubar and the VM menubar. A similar technique could be used to swap toolbars. It's not what you asked for, but there's a fair amount of flexibility there. Also, you could do something really kludgey with the tab control. I believe it's straightforward to have multiple tab controls in the gutter. But the tabs don't behave like toolbar buttons. -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Ask not how you can "do" free software business; ask what your business can "do for" free software.

On Tue, Oct 14, 2003 at 01:39:52PM +0000, Uwe Brauer wrote: And the point of copying the windows UI disasters into XEmacs is? OG.

Uwe> On 14 Oct 2003, galibert(a)pobox.com wrote: Uwe> Why disasters? It's a matter of personal preference, tradition, and to some extent the user population. Emacsen traditionally have very powerful keyboard-oriented interfaces with learning curves that are not terribly steep but very deep. They provide lots of facilities to customize Emacs's behavior to your habits. It's always possible to recursively use your previous customizations to do something more powerful, or to conform more accurate to your needs. Windows applications, on the other hand, are designed to have a very steep learning curve that bottoms out immediately, like toasters, blenders, and other modern kitchen appliances. My usual experience with a Windows program is that it has tons of buttons, none of which does quite what I want it to. (Freecell and Minesweeper are exceptions. Minesweeper is especially nice since it can be configured to have hundreds of buttons, all of which are essential to its use. :) In less prejudiced terms, in a text editor toolbars for most functions are an extremely inefficient UI compared to the keyboard, for a practiced user. On the other hand, with suggestive icons a toolbar can help an occasional user find important common functions quickly, while the menus (again, if well-organized) can provide a hierarchical index to less frequently used functions. Emacsen have historically not been for occasional users; they appeal to users who use them as their primary application, and even their "operating environment". So I wouldn't call it a "UI disaster" (150 million AOL users can't all be 100% wrong ;-), but I question whether we're doing either the existing users or potential users a big favor by targeting the number of buttons in the default configuration of XEmacs. On a 1024x768 display, it should easily be possible to get 20 buttons in the toolbar on a reasonably-sized frame. A better idea, IMO, is to make sure that there are no more than about 10, and that they are useful. I use the Info, Print, and News buttons in the default toolbar, and could dispense with the News and Info buttons---I use News exactly once per session, and most of the time I get to Info via C-h C-f. The other 12 I only use if someone reports a bug. In fact, verifying bug reports is the main reason I build with a toolbar any more. OK, clearly I'm unusual, but really I have to wonder what good another 15 buttons would do. Especially when a bit of work would make the tabs a significantly better interface to application switching than a toolbar is. Uwe> would Xemacs become more instable because of more icons? I Uwe> was pointed to kile /kile.sourceforge.net/ which comes with a Uwe> lot of integrated icons. Does this make the application Uwe> unstable, if so, why? No, it has nothing to do with instability of the application. Excessive iconization has, however, been known to drive Emacs developers into a certain degree of mental instability. -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Ask not how you can "do" free software business; ask what your business can "do for" free software.

On 14 Oct 2003, listmailxemacs(a)kyllingstad.com wrote: I rarely use the tool bar by myself, it is just I feel this is something the average users misses most. yeah I remember this, but first David is not a Xemacs user himself and then the intention of my question was not only: is it possible now, but it is worth to implement something like this in the future? Ben's todo list at least let the possibility open. Uwe

On 14 Oct 2003, stephen(a)xemacs.org wrote: [snip] Well, I agree with most what you have said, when I started to use computers most of them did even had a graphical interface [snip] Well (X)emacs is not only very customizable but also has features which to my knowledge no other editor processes. For example for editing scientific documents the combination: auctex+reftex+x-symbols+preview-latex has not surpassed by any free or non free application. It is however surprising to see, that most users do not want even to consider using it, since it has no user friendly interface such as icons. I feel the whole affair as a huge waste of energy in some sense, great effort, but then not the amount of users one would expect. Of course if, one hand the implementation of such a feature would be a very difficult task, realized may be in the release 33, and on the other side (X)emacs consider itself as a product written by hackers for hackers, then let's forget it. [snip] This of course is the last thing we want.

On 14 Oct 2003, james(a)xemacs.org wrote: [snip] I had made similar, --but not to this extend--, experience with students using matlab. My naive hope is, that once you use something with a lot of features + icons after a while you drop using the icons because the other way is so much faster, but may be not: perhaps the group of users is divided and these who want icons and nothing else, and those who want to configure and enhance.

On 14 Oct 2003, galibert(a)pobox.com wrote: Listen you don't have to convince me, I barely use icons at all, if I find a useful command I bind it to the key. I don't use kile either, I used once for a while winedit, thanks. I just want to point out that there is feature rich application like Xemacs and most beginners are not willing to use it because it does not feel right. It is pointless to discuss this with them. You can list up all the nice features of Xemacs but the answer is: too complicated, not for me. Now there are 2 possibilities, we ignore these people (Xemacs is for hackers or advanced users written by hackers.) Or we try to make Xemacs more attractive for the public. If this too difficult, requires too much effort, will happen the next century, etc then let's forget it. Uwe Brauer