The theory behind the GUI
While watching different idéas for the GUI-design I often think "That's beautiful, but is it functional?". My opinion is that the GUI should be functional /usable and beautiful. Are there any special effects they should have a function of usability.
Transparency in the terminal windows: Do you have to know where the trash is when using the terminal?
Aqua dock imitations or Dockbert variations: They look very nice, but are they really usable? You don't even see the name of the window (the icon represent) unless you hold the pointer over it. The classic BeOS dock is very good to use. Windows are grouped by application, and you don't see more than you need to.
There are many more examples on good and bad GUI designs.
I recommend all GUI designers to read a book by Donald Norman, for example "The Design of Everyday Things". You will get more comprehension of what's usable and not usable.
I can also recommend this webpage about GUI-design, guuui.com.

Comments
The theory behind the GUI
hi,
there are several books out there about useable interfaces.
each operating system/window manager has it's own guidelines.
another one called "the humane interface" by jeff raskin (used by macs).
amazon has it here
an interface guideline for gnome. gnome looks nice but it's becoming more and more like windows. i like gnome more than kde. but kde i think is more advanced and smaller.
eclipse has a very very nice interface. guidelines are here here
also, for some reason, a be's website still exists http://www.beatjapan.org/mirror/www.be.com/index.html. some can scavange a few remains from there. the date is 1998
i think APP/Interface team may profit from reading a couple of books or articles about guis while building the gui.
i hope my suggestion helps.
Not to worry
As long as I'm around, Haiku will have someone determined to have a usable interface and who has a clue what usability is really all about. I'm no expert, but I like to think that I do OK. Stephan is also aware of such things - WonderBrush is good evidence of that. The GE list and the forums are a good place to toss out ideas (particularly in the Suggestion Box), no matter how wacky they might be, but the members of the project will make the final decisions when the proper time comes. :)
--DarkWyrm
The theory behind the GUI
Large grey concrete buildings might be effective and resource effective, but would you want to live in one? ;)
I've always, well atleast for my first ten years in computing, been a fan of extreme minimalism and utilism. However, with my constant decline in computer interest so has my desire for effectivity decreased. I know my computer has alot of processing power and ram to spare, so while not use a small percentage of to make things a little prettier?
Ofcourse it should be an option for the user to decide, but something isn't worthless because it's only decoration. Doesn't have to be like Windows Vista (moving a window uses 30% cpu on my Athlon 165 btw) but appearance does matter!
@fhein
I'm a huge proponent of usability, but I also like to have a GUI that looks nice. Stephan Assmus' new icon format and theme are IMO look phenomenal and are so small as to be theoretically faster than any other OS out there. While it has been changed some since I last worked on it, the decorator engine in the app_server is something I designed, along with the default Haiku decorator itself. I like to think that my work still hearkens back to the classic BeOS while still looking new and yet easy on the eyes. You can have both -- it just requires lots of careful thought. :)
The theory behind the GUI
thanks to firefox, the biggest thing is tabs. perhaps soon we'll see email with tabs, word processor with tabs. This is the next step from the Be concept of grouping. It's better because with multiple windows and tabs, I can group things by task. For example, I can have a window with all the haiku sites in tabs, window with all my firefox tabs, etc.
So with workspaces, deskbar grouping and tab gouping we have enough hierarchy for all kinds of things. This would allow me to keep track of a lot of websites, a lot of email-a lot of everything at the same time!
The theory behind the GUI
I think the general theme of this OS should be "we're not going to patronize you with baby things like My Computer. You are smart and have a lot of stuff to deal with but you don't have time to deal with computerisms. We're going to work on make managing all that stuff much easier. Simple but not simplistic"
The theory behind the GUI
totally agree.."you're smart enough than to have us baby you. make things simple but not idiotic." just like that.. make thigs be useful and productive NOT be stupid
The theory behind the GUI
That's right. oh boy big icons....that's only good for a kiosk at disney world.
The theory behind the GUI
Not quite.
As for My Computer being babying, it may not be the metaphor you prefer for your computer, but it does not mean that it is babying. In anycase, with regards to your comment about the big icons, simplicity is not a lack of visually pleasing elements, it is ensuring that those elements operate correctly, productively, and pleasingly. Using a computer should not be like reading a flyer from a Microsoft Publisher '97 template. Like all things, interfaces should not be assembled backwards from function to form, they should be consciously designed with the principles of form and function in a harmony that creates a productive and pleasant experience for the end user. There are cases, where big icons do that. There are cases, in my opinion, where big icons could be appropriate for Haiku.
The theory behind the GUI
honestly i think gui comes down to 1. usability 2. flexibility to do all the shit your heart desires.
1. note that usability comes first. for example selecting folders with wildcards would make my life so much easy!!!! tiny touches that helps you perform better. any type of list should need a search with wildcard option
2. and make it look cool..please. i dont ask for apple puffy explosions or twirly magical minimizing windows. but some neat icons, good colours that dont bother me, more than 256, background picture. its about making the space mine..
these would make me a happy camper =)
[for both of these imagination is the limit..but too much is always bad. its good having a lot of flexiblity..but not cool having a milllion of these shipped with R1..just choose the best 17 and thats it..let the user murder himself with more. but do provide the hooks to put more..]
so there..i solved your gui riddle..simple as 1,2 ..umm 17
The theory behind the GUI
The mentality I'm against is the idea that all you need to do is 5 things represented by 5 big icons on the desktop. It gives the illusion that things are all easy and simple. But then if you want to do more than that, it becomes very hard.
For example, Gnome looks very simple but if you want to configure the UI you go to a config file. If you like the apps in ubuntu you're fine but if you want to download something else that's not part of the package, it becomes very very complicated.
So I guess what i'm trying to say is I have no real problem with big icons but don't make the apparent simplicity a facade over a very bizarre foundation
I want ease of use to go all the way down to the command line and the API's.
Re: The theory behind the GUI
The one case where you could use transparency are those modal dialogs that have to block your window for something important. Even then they shouldn't be transparent by default but if you need to see what they are blocking you should be able to turn on the transparency.
The theory behind the GUI
True simplicity isn't a façade.
The theory behind the GUI
I hate to say this, but there really isn't any simplicity in an operating system. There is just shifting around the complexity. The key is to place the least amount on the user
[user==][dev======][OS==============]
The problem is that programmers are lazy, resulting in cranky users (and rightfully so).
[user======][dev==][OS==============]
I can't remember offhand, but it's a usability rule called the Law of Conservation of Complexity.
The theory behind the GUI
i agree with dark wyrm..hes right...but in reality the users and developers meet around halfway..because users try to use the os..and the developers try to program it..both get lazy and both try to find the easy way out..its just human nature...right? but if developers would try harder..definitely an os would be more usable..it depends on usser feedback too! so its more organic.
The theory behind the GUI
Maybe not in the one you are using now. But, when a conscious effort is but not just into the programming of an interface, but into the design on an interface, it is necessary and proper for it to have simplicity. Unfortunately, it happens to be that this element is often not present in UIs that we use today.
The theory behind the GUI
I would certainly agree. Regardless of the code under the hood, the actual GUI of a program needs to be as simple and helpful as possible. I'd say that the reasons why there are so many complicated programs is lazy programmers, poor design procedures, or a combination of both. Design should come first and then the code, not the other way around, which is the common case.
Not all of the fault of bad BeOS apps lies with the developers, though. BeOS doesn't have nearly the features that, say, OS X or Windows have, but it has enough to make lots of regular programs possible. The burden falls on the developer to make up for what the OS doesn't provide. For some kinds of programs, this isn't a big deal. For others, though, a BeOS developer has more work to do than developers for other OSes. Rich text editing (complete with tables and pictures) comes to mind, for example.
The theory behind the GUI
Latest Opentracker, alt+shift+a. Does RegExps, too...
The theory behind the GUI
The OS should do more so that users don't have to work so hard. It shouldn't make advanced actions more complex just because most people don't use them. We want it to be easy to really take advantage of the computer and not be stuck at one maximized window at a time.
The UI should be ergonomic. It shouldn't distract you from what you are doing but it shouldn't be dull. Jumping on the latest 'in' style just looks dated after a while.
I've actually "asked tog" about the global menus on large monitors. he says his studies show they take less time overall even taking into account bring the cursor back to the app. he also said that pie menus could work to replace regular context menus even at the edge of the screen.
maybe we should take a look at next style vertical menus? You can move them around and tear parts off
My idea is to just have one word in the upper left corner "Menu" and when you move your mouse to it, it would pop up a vertical menu. This would save more space than the mac menu and especially the windows/beos approach which has menus for every window on the desktop.
or maybe give a choice between mac, next, beos/windows and my minimalist idea. After all, we give a choice of scrollbars.
The theory behind the GUI
Perhaps it's because I come primarily from a Windows / BeOS background, but I actually prefer to NOT have a global menu at the top like Mac OS. While it may help with Fitts' Law because the menus are effectively infinitely tall, it also introduces a mode, especially combined with MacOS' choice in allowing a program to be open even if there are no open windows.
At work, we use a Jabber server and the Psi client on all the desktops. I have a Mac mini in my classroom, and when the client starts, I close the window to reduce screen clutter. The only problem is that I have click on the desktop or look at the menu itself when I come back to find out what app I'm using because technically I may still be "using" Psi even though all I can see is the Finder. I can't count the number of times I've used a keyboard shortcut only to find that it didn't work because I was working with a windowless app at the time. Slower? Definitely, but also easier for anyone to know context when working with more than one app at a time.
Not in Haiku, we don't -- the sources for that preferences app are still in the tree, but it is no longer maintained or included in the image. I can't figure out why Be even had it.
The theory behind the GUI
if I have app A and B open and using A, on windows i can see B's menu so I know right away what to click on in B. With a mac, b's menu is a surprise before you switch to it. So right now it doesn't look good for mac and all I have is tog's word that his studies show it's faster.
I use tiny menu in firefox though it makes fitt's even worse. it saves screen space since all the other buttons are on the same row. I wouldn't like it for other apps because I'm very familiar with firefox as I use it all the time.
I wonder what you think about the Office 2007 ribbon? I saw 2 problems right away: it throws off any hope for a consistent Windows UI as you still need menus for all the other apps.
Also, it looks like it was designed for use of a single window that's always maximized instead of multiple windows. I think a floating or docked palette would make more sense and probably would make it easier to do the same action on multiple open documents such as search and replace.
of course if we did that, microsoft would have to change everything once again and that would be pretty funny. :)
The theory behind the GUI
I believe that the mac style menu is faster just like the Fitt's law predicts, but it also introduces confusing modality just like arieb describes. What I would like to see in Haiku is that when application uses full screen it would have its menu on the top of the screen just like in mac's. In other situations, all applications would have their own menu under their title bars/tabs as usual in Windows and others.
This would allow fast usage of full screen programs, without any modality confusion when using two or more programs side by side and should also fit well with the Gestalt laws (i.e. no visual disconnection between application's main window and menu).
As for the MS Office 2007 ribbon, I believe it is a huge mistake that will spread like a virus to other programs. I have seen how easily typical office workers freeze when something looks just a little out of the ordinary in the program they try to use. I bet they will have problems every time they have to switch to a strange ribbon tab from the one they are used to. I doubt there is any usability justification to the ribbon. It is just a chance for the sake of marketing and training consulting.
Quote: There are many more
There are many more examples on good and bad GUI designs.
Well, in the The Chronicles of Shell Replacement (Part I and Part II) by Mike Holland, he claims
All in all, I think SharpE is an ideal replacement for "Bill's" idea of what a desktop environment should be. SharpE has, in many ways, increased my efficiency in doing work, allowed me to make the desktop MY desktop and SharpE looks good doing it.
CLICK HERE and here to get just two ideas. :O
In other words, involving SharpE devs/artists could be useful !
(and a C/C++ port of #E would be great too)
>>Forward Agency
In progress we (always) trust.
I'd rather not click
CLICK HERE and here to get just two ideas. :O
In other words, involving SharpE devs/artists could be useful !
(and a C/C++ port of #E would be great too)
I don't see anything interesting in the screenshots. They try to look cool, but is there anything cool about the way we interact?
When I read statements like
Every pixel of visible space can be personalized...
I believe that they are creating a geek-only system. We want a system that is free of unnecessary complexities, not one that is full of complexity.
yay.
I believe that they are creating a geek-only system. We want a system that is free of unnecessary complexities, not one that is full of complexity.
yay.
Austin B.