Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 9/5/84; site unicus.UUCP Path: utzoo!mnetor!yetti!unicus!craig From: craig@unicus.UUCP (Craig D. Hubley) Newsgroups: comp.sys.amiga Subject: Clipboards and the Amiga UI Message-ID: <668@unicus.UUCP> Date: Sat, 20-Jun-87 00:31:46 EDT Article-I.D.: unicus.668 Posted: Sat Jun 20 00:31:46 1987 Date-Received: Sat, 20-Jun-87 20:21:40 EDT Reply-To: craig@unicus.UUCP (Craig D. Hubley) Organization: Unicus Software Inc., Toronto, Ont. Lines: 116 Here follows a very long Amiga UI discussion that may not interest you, and includes one man's long-considered opinions on the issue of Amiga desktops. mjw@f.gp.cs.cmu.edu (Michael Witbrock) asks: > When is a between windows, within windows cut/paste buffer going to be > implemented. I realise that there is a clipboard device. Why isn't it ever > used? Even a simple minded, X type text cut/paste buffer would be nice. In my opinion, one of the most serious errors in the Amiga UI design was the brain-dead imitation of unfriendly Macintoshisms, from whatever origin. The clipboard I believe to be one such "ism". When I take something from one window, I prefer to have it "stay around", IN SIGHT on the desktop. So, selecting text or graphics from one window and "cutting" it should place it ON THE MOUSE and allow it to be placed on the desktop somewhere, as is or shrunk to an icon. Many of these could lie around the desktop at once, and could be picked up by double-clicking on them (since there is no other meaningful action that could apply). Any directory could thus become the equivalent of a "scrapbook". Once it is carried on the mouse, it could be moved into the application window and placed there. Much of this could also be implemented by making a multi-item clipboard visible, with any item currently in it selectable. The whole concept of an off-screen clipboard for something you are CURRENTLY WORKING ON is anti-WYSIWIG and against the spirit of desktops. It also forces you to either keep only one item in the clipboard, or REMEMBER how many items and in what order are there. Or check same, which is the same as an optional visible clipboard. Mac gets away with this metaphor only by being monolithic and singletasking. I first used this sort of capability in Xerox Interlisp, and I despise having to select Cuts and Pastes from menus sitting at the top of the screen, far from my real work. Pick up a pencil and see how much mouse movement is required for a typical operation. The research community (Lispmachines, Suns, etc.), definitely power users, LOVE MICE. This is in contrast to power users who move to the Mac, who often hate the "rat". Quite simply, it has to be moved too much. Thankfully Apple is now moving the other way, supporting submenus, pop-up menus (but only to fill in forms, not execute actions). The removal of Steve Jobs and a refreshing new approach to basic research seem to have triggered this. The one-button mouse was another such Job-ism, even though Star proved the two-button mouse made more sense. I would hate to see the Amiga, with the best hardware, end up with the worst software. Also in this vein, pop-up menus quite desperately need to be supported. They cut down on screen movements considerably (read 90%) and genuinely support power users while greatly aiding naive users. Think, you can say "Exactly what will happen with this particular item" rather than "The kind of thing that will happen", as in: "Discard Hack" rather than selecting Hack and then the generic operation "Discard". Languages like Smalltalk display the workability of mixing generalized protocols with specialized commands. It is essential to mix CLI-style and icon-style work, which are two differing "modes" on the machine now that need to be better integrated. It seems, from the Star work, that putting a couple of operations on the mouse makes life much easier than having just one "point" operation. After all, people do more than point at things on desktops. They also pick things up, move them, and perhaps crumple them. All with the hand, without reference to other objects. Why do we have two buttons, anyway, if not to perform at least two operations? The present right-button protocol does nothing but show you what the top level menus are. This saves almost no screen space, since the line is still there when the menus are off,usually. We might want to extend the present mouse protocol to be: POINT - left-button = select it (same as now) USE - double-left = use it (same as now, window to top, icon to move) CHANGE - right-button = POP-UP menu of operations (if small, like window operations or doing something to an icon) or pull-down if large (like a whole document, etc) And what about: - double-right = use it in some other way? move it? see info? So far as I know, this is not really contrary to the Amiga UI definition now. This would go a long way towards removing reliance on the top row of the window or screen (with its required mouse movement) and moving towards a real object-oriented desktop, which I believe C-A tried to do first (and failed for lack of funds, talent, management enthusiasm, ???) before porting TriPOS. Comments? Flames? Agreement? I would particularly like to hear comment from those who have worked extensively (as users and/or programmers) on lots of icon/window/pointingdevice systems (I've used 5 extensively : Icon (Bionic Beaver with the world's fastest - trackball-based screen editor), Xerox Star, Xerox Interlisp, Mac and Amiga). Need comment from Sun and Smalltalk folks. The Amiga is THE IDEAL MICRO to implement a very high-powered UI since: - it has IFF - it has sprites for quick-moving and arbitrarily shaped icons, and small popup menus. - it has multiple screens analogous to a "rooms" metaphor, to support multiple classes of user tasks. That is, Workbench for doing file handling, VT100 for talking to big boxes, Hack while waiting for downloads, a fake spreadsheet to fool the boss, etc. All with a very convenient way to move them around. This approach is JUST NOW being picked up by the research community, and the Mac II missed it by going to a large "virtual desktop" which Xerox studied (as "Bigscreen" I believe) and discarded because "rooms" was cleaner. - it has a dedicated and interested user community who know what they like and what will work. By contrast, check out GEM sometime. - it has the blitter for moving large things around quickly. - it has the potential for high-power 68020 processing. - it has IFF. You know, its only good to have THE hackers' machine if it gets hacked up the way hackers like it, by hacks of course. (background) Hack Hack. Hack. P.S. When I have time, next century or so, I'd be glad to help write a good "clipper" and make it PD. I seriously believe that a better desktop, and a better commitment to desktop video, could boost the machine to absurd heights. And yes, I'll buy a 2000 and put a 68020 in it. Soon. Really. -- Craig Hubley, Unicus Corporation, Toronto, Ont. craig@Unicus.COM (Internet) {seismo!mnetor, utzoo!yetti}!unicus!craig (dumb uucp) mnetor!unicus!craig@seismo.css.gov (dumb arpa)