Slashdot Mirror
OpenOffice UI Design Proposals Published
An anonymous reader writes "Various members of the OpenOffice.org community have been submitting their first revisions of proposals to the OpenOffice.org Call for Design Proposals to redesign the user interface of Open Office. As part of Project Renaissance, attention is being drawn to the OpenOffice user interface, and it's 'user-friendliness.' Among the designs, is FLUX UI, which won an award at the Sun Microsystems Community Innovation Awards Program. Anyone can, and is encouraged, to check out the proposals (scroll to bottom of page) and leave your comments so that the designers can improve their designs for the final deadline for proposal submissions to the community."
First, do away with the standard File menu bar. Put the most common actions (Create new file, Save file, Print file, etc) in a big button in the corner. Then create a tabbed menu "strip" separated logically by function. Have something like a Format strip and an Insert strip with all the actions you'd expect included there.
As computers become more touch-panel oriented, bigger buttons will be mandatory. The old File Edit Options Help bar is going to be a millstone.
Hey, that's a great idea! But I think we should call the strips "ribbons"...sounds way more sophisticated that way.
... make everything available via hotkeys (emacs and vi mappings should be provided) and change the arrow keys to print the particular arrow typed. This would be a significant improvement over the current design and would encourage users to work instead of playing with their mice.
;)
Don't forget to make sure it's difficult for the visually impaired to use, and impossible for those relying on screen readers to explore the interface as a sighted person could do! You're 99% of the way there already, I'm sure you can come up with the remaining 1%
Self proclaimed typo king, and inventor of the bear destroying coffee table (patent not pending).
Shouldn't these common actions (Save, Print, New) be presented in a standard way across all applications? I don't think it would help ease-of-use if OpenOffice implemented its own cutesy button bar that's different to all other apps. But if most programs on the system could change at the same time, it might be worth a try.
-- Ed Avis ed@membled.com
Isn't this kind of ironic, Oracle?
I'm surprised far too often by great and useful features I didn't know OO.o had.
It has never been hard to learn and is pretty ubiquitous. I think it all works pretty well.
While I am sure that all this additional exploration of new ideas and concepts is a good exercise in generating new ideas and all, I think gone are the days when "new" means "better."
It turns out that "circle" is the best shape for most applications of the wheel. (Some exceptions exist, you don't need to point them out.) For 2D information formatting and arrangement, the menu bar and tool bars do a pretty good job of making it as easy as possible even though other paradigms exist and the menu/tool bar doesn't cut it well enough for other things.
(grammar corrected)
I wouldn't consider this post Flamebait. The parent pretty much described Office 2007's interface, which everyone was complaining about when it first came out.
On the subject of ribbons and tabs, I would favor the ribbon interface similar to Photoshop. For example, when you choose the cropping tool, there is a ribbon with all the options for cropping. However, I'm not too much of a fan of the Office 2007 interface. I think they did a poor job in the organization of the functions, and didn't even offer an option to switch interfaces.
You've been subjected to "The Beta Index." Go in to Help/Preferences, and uncheck "Use Beta Index" in the Beta Index options.
OpenOffice.org Wiki has a problem Sorry! This site is experiencing technical difficulties. Try waiting a few minutes and reloading. (Can't contact the database server: Too many connections (localhost))
It's sleek, informative and minimalist. 2-thumbs up, would buy again!
Oh god, that woman is John Romero!
Please, please, please.
You can have it both ways. Do your Flux/Ribbon thing, but leave a standard mapping shortcut for those of us who don't like to spend 10 seconds mousing around when we can perform the same command in three keystrokes. Allow us to turn off the ribbon doodads, show both at once, or just the legacy menu.
You don't want to turn us into this, now do you?
Difficult for the Visually impaired? How so?
I actually have an eyesight problem myself, it's nothing MAJOR in the sense that I can't do everyday tasks (I can't, I just can't see clearly very far). I blow up the font a little bit and I'm all right and for me, personally, the ribbon interface that people seem to hate so much is a godsend. I can easily tell what every button does without squinting, but then again, I never feel the need to use an on-screen reader or whatever. However my first inclination is that the ribbon interface isn't at fault for "breaking" them, but rather the on-screen reader programs just haven't been updated to handle the new interface.
+1 IDisagreeSoHeMustBeATrollOrAnAstroturferOrAShill
What,
It needs is a significant amount of effort bringing the graphs in Calc up to a level that even approachs what was available way back in 1986 in Lotus 123.
Calc's graphs are a MAJOR stumbling block to my being able to push OO to clients as an alternative to XL.
Redesign graphs, enhance them, whatever you want to call it, fix them please....
I hated Office 2007's "ribbon" interface when I first saw it. However, after the first few days of using it, I found myself at least twice as productive when using it. Yeah, I know... it's a Microsoft idea, and therefore it's automatically bad. Except, it isn't. Everything I need is easier to get at with fewer clicks, and working properly with styles is finally a snap.
It's hard for me to take seriously people's snobbery toward the latest Microsoft UI designs, when so much of the open-source world is simply a direct rip-off of OLD Microsoft UI designs. OpenOffice is largely an MS Office 2000 clone, KDE started out as a beefier Windows 95 clone, and the new desktop menu in Gnome is a bastard stepchild of Vista and OSX. Up until very recently, innovation in UI design hasn't been an open-source strong point... and it would be nice to see more innovation rather than derivative work in this area. I look forward to seeing what the OOo community(*) comes up with.
(*) Just as I look forward to seeing what the "OOo community" IS under Oracle. Up until now, the community was basically "Sun".
Bullshit. Human interaction with computers is embarrassingly inadequate. I still have a calculator, despite having a computer capable of billions of FLOPS, why? I still sometimes write things on paper, even with my computer right in front of me, and being able to type faster than I write by hand--why do I do that?
I'll tell you why: because computer interfaces still pretty much suck. Getting what I want in front of me RIGHT NOW is an elusive thing in computerland.
Did I just goatse... myself?
No. In the event of an actual goatse, you would be quivering in the corner desperately searching for your innocence.
Don't show the user what he/she doesn't need at the moment (context sensitivity).
Oh dear, I really don't like this one. The UI will be in constant flux. I won't know where things are or what the application can do. Also, there's a chance I am about to do what they don't expect me to do.
Ever look at the difference between the old analog aircraft gauges and the newer strip-based instruments in glass cockpits? There's a massive difference. Looking at the Cessna 172S and comparing the analog to G1000 versions, the turn coordinator is different, the airspeed indicator and altimeter are now sliding strips, and all of them are overlaid on the horizon indicator, which is now essentially the size of the screen. There are enough differences that a pilot moving from analog to G1000 generally has to be checked out before being allowed to fly it.
Interface overhauls should be carefully considered, but suggesting that they should never happen is akin to announcing that nothing more can be invented. That's been declared by many people over time, and they are often proven incorrect.
You can never go home again... but I guess you can shop there.
You think that's bad? Try using MS Office 2007 on a eee. The bloody ribbon takes up a quarter of the freaking screen!
The latest version of OpenOffice is the first one on OS X where the spell checker actually uses the default, built in spell checker on OS X which is used by all the other programs and already programmed with all the preferences and words from my other work. I applaud the addition of this functionality.
Sadly, the UI by which it is accessed is clunky and nonstandard. In every other program, highlighting a word and right clicking on it brings up the context menu that lets me directly select the corrected version of the word. In OpenOffice I have to run the spellchecker which opens a separate window to provide suggestions which I then have to close once I'm done and go back to working. The only usable way to do spellchecking becomes to ignore all spelling errors until I'm done then go through and correct spelling mistakes at the end, a slow and tedious workflow.
Further, In every other program, the context menu that comes up when right clicking on a word allows me to use the dictionary/thesaurus service and to correct grammar mistakes using the universal grammar checker. OpenOffice still ignores the standard APIs and thus still does not have these freebie functions even basic text editors on OS X have. When I have to copy and paste my text out of my full fledged word processor and into a basic text editor in order to check grammar or apply any other text services, well something is wrong. Some of the features OpenOffice does present in their context menus are useful, but really I want to select the correct spelling for a word flagged as misspelled a lot more often than I want to change the font size of a word. The options presented to not reflect my needs and I doubt they reflect the needs of the average user.
So basically my complaint with OpenOffice's user interface is that it does not conform to the standards of the OS on which it is running and instead dumbs down functionality to the level of the lowest common denominator OS.
Yeah I guessed so, because, well, I assumed you couldn't proofread your comment very well.
(ouch ouch! karma burns!)
But... the future refused to change.
Why is it that developers think they need to move crap around or redesign so that they frustrate users? Is it some kind of sick game?
Because that's what Microsoft does. See, if you're going to release a new application or operating system and ask customers to pay for the upgrade, it helps if you offer new functionality that the customers can use. However, developing new functionality that's actually useful is difficult, and if you aren't able or aren't willing to do that, then the next best thing is to make it look very different. Ideally it will look much better, but as Windows XP's Luna theme has proven, "different" is enough.
Because if something looks different enough, human psychology makes people think that it is actually different. Also, culturally, we're disposed to think that newer+different=better. Therefore, people will pay to upgrade even if they don't stand to benefit.
So with each new version of Microsoft Windows or Microsoft Office, they always change the UI even if the functionality is the same. At the very least, they apply a new skin and shuffle around which controls go where.
I'm not sure, though, what the benefit is for FOSS. It's not like people are going to pay to upgrade to the new version of OpenOffice.
Try either double-clicking on one of the ribbon tab titles, or right-click on the ribbon and check "Minimize the ribbon" or using the keyboard: , , n
Don't forget to make sure it's difficult for the visually impaired to use, and impossible for those relying on screen readers to explore the interface as a sighted person could do! You're 99% of the way there already, I'm sure you can come up with the remaining 1%
Oh, theres a lot more than 1% of the way to go to make a totally useless GUI.
How about using unintelligible icons? That way you can make it impossible to teach anyone how to use it verbally, makes it only possible to describe operations visually. "now right click on the second icon from the left that looks like a squashed centipede, obviously everyone who centers text thinks of squishing a centipede". Bonus points if the icon is could be interpreted obscenely in a Freudian manner or is a swear word in some obscure ideographic script. After all, all of your users are experts at learning ideographic scripts like Egyptian hieroglyphics, so instead of typing "load" or "open" on a command line, make them memorize that a clovis arrowhead means open in this program, but a little star trek shuttle means open in this program.
Then too, make it graphically as utterly modal as possible. Pop up screens that come from pop up screens that come from menu bars on pop up screens. Make it as challenging as memorizing the knot and overlap structure of a bowl of spaghetti. Organize the pop ups and menus solely by programing team or by how the marketing gang declared how the tool would be used. Bonus points if its possible to open multiple different config windows simultanously, but only change things in one window at a time. And try to lock the screen so the user can't look at other windows (like a cheatsheet or notes or whatever) while a config window is open.
Don't ever use threads and don't worry about responsiveness. If clicking on the "wrong" thing appears to lock the machine up for seconds, even minutes, with no way to quickly stop it or go back, thats OK. You know you've succeeded if the user forums describe the best roll back technique as "quit and reload" or "easiest just to reboot and try again". If they complain that is slow, tell them to get a faster PC.
Can't get here from there... Lets say there is 20 step procedure to get from here to there. Make sure that the rollback procedure is a totally and utterly different 40 step procedure. Whatever you do, don't make a global "undo" button that works, or at least works reliably (its OK if it only works on 75% of the operations, then no one will expect it to ever work and thus will never use it). Forward should never equal or be equivalent to backward.
Everyone whom uses the program only wants to see your glorious program, right? Not their little data or whatever it is they are working on. So FLOOD the workspace with an infinite array of tool bars and buttons covering almost the entire workspace. After all, if they paid $500 for a bigger monitor, your program should get that screen area, not their data.
"Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
Modal dialog boxes interrupt workflow. We need to make most dialog boxes modeless and dockable.
The only thing worse than computers is phones with computers in them.
My IP phone here at work helpfully keeps a log of missed calls, and their numbers. It helpfully displays these as a list on the screen, and helpfully lets you highlight one and press the "dial" button to instantly re-dial the call. But, uh... it's too fucking stupid dial the "9" for an outside line, meaning the call either fails or goes to some random extension in the company.
As an added bonus, there's no way to have the missed call list and dialing interface both on-screen at the same time. So my phone is surrounded by a pile of Post-It notes containing numbers I had to write down, only to dial mere seconds later.
In short: computer interfaces suck, ignore anybody who says "leave it the way it is" because the way it is sucks.
Comment of the year
I suggested something similar on ubuntuforums a couple of years back and got shot down instantly. But the idea behind flux/ribbon is actually really good. Hide buttons that you arn't using at the moment and give the document more space.
Menubar:Replace the main menubar with a menubutton, use this to show all menu bar buttons that aren't shown by menu buttons that are spread out at the appropriate ends of the main toolbar (help)
Buttons: you are likely only interacting with one thing at a time, if define usage cases narrowly enough to put all the relevant tools on a toolbar but widely enough that there are only a few settings, then you can save space (or give more space to just the relevant tools). /print/save/open/new) and read (auto-hide the entire toolbar, giving 100% of screen estate to the document)
*Some Actions can be done from any of the states Copy/undo stick this outside of a "container"
*Bind keys&buttons (automatically based on selection?) to toggle whats in the "container"
*Imo the container should be editing text/ editing pictures/layout(including columns & tables)/document(changing setting
|Menubutton(s)|permanent buttons|toggles:relevant buttons|help|
Customizations:Make the whole thing customizable (if the relevant buttons are to big to fit in the provided space that section should be the first to loose space (be it only showing the 1st few and adding an arrow or allowing scrolling though the relevant buttons)) and allow users to define thier own usage cases, with repeated buttons (looking at you kde3) and thier own triggers (some people want to go straight to the text editing menu as soon as they select text others dont).
Make the whole look changeable (companies may want to replace the default menu button with a company logo? or make the whole thing bright pink?)
Allow the different sections to be separated (so you could give the relevant buttons an entire tool bar underneath)
Allow different toolbars to use different sized icons
Providing too much customization is not a bad thing as long as most people can use the defaults.
|Menubuttons|_____toggles_____|help|
|permanent buttons|relevant buttons|
Themes:Provide an easy&safe way to save/share a theme.
Provide sane defaults and get it out there, during the next release cycle look at which themes popular (you'll probably find there to be a few popular themes, classic, ribbon, office, geek, flashy) and ship them with the next release. There is no point in doing research designing what you thing is a good compromise for the work load that office/home users put their suite through, when you can just put a version out there and see what people do with it.
IranAir Flight 655 never forget!
Well on the other hand if new users are trained to use ribbons coping that interface might be beneficial.
Now some aspects in OO.o are just horrible, the color picker looks *and works* like something straight from 1994.
I think they should ask for help or at least inspiration from AbiWord.
Abiword has a wonderful UI, minimalistic yet does 99% of what you want. It's colorful and even cute yet still looks professional. The icons represent exactly what they mean, the menu structure is very intuitive etc.
The problem of Abiworld is that it can't do *every fukken thing imaginable*, you can --for instance-- underline and/or overline and/or strike anytext, but always using the same color of the font. AFAIK in OO.o 3 you will be able to use 3 different colors for that. I'd personally stab anyone who sends me a document with multiple colors per font.
In principle you could apply Abiword's elegant GUI design to all the features in OO.o, it just will be a lot of work. But it would still be intuitive and standard.
This FLUX is both nonstandard and a cheap rip off.
But... the future refused to change.
I'm really starting to think that all these people that need extensive training to use office software should be out back shoveling sand over a wall. I mean, if they can't get the basics in 10 minutes of clickly-clicky, I shudder to think the pearls of wisdom that will emerge once they get down to 'work'.
Nerd rage is the funniest rage.
THANK YOU
There are already more important problems with OO.o anyways.
In Writer, the image scaling looks like crap (they print out OK, but on screen they look horribly pixelated).
Various functions in Calc are cumbersome to use compared to Excel. They need to take some time with an accountant, hear all their complaints and streamline the UI - they're things that most users won't be bothered by, but when you're working with spreadsheets literally all day long it throws a big wrench in the works (accountants using spreadsheets where they should be using databases - NOT Access BTW - is another topic).
I don't have any serious problems with OO.o but lots of people who use office apps more often have complaints, they should fix those instead of trying to make it look nice. I think offering an "MS Office compatibility mode" in the installer to change the default file formats would go a long way to reducing complaints - that's the most common one, they hit save, don't look at what they're saving it as and then other people can't open the OpenDocument file and the whining starts.
"When information is power, privacy is freedom" - Jah-Wren Ryel
I agree (about it looking old, anyway). I think its layout should remain approximately the same, but it needs a facelift.
Hey, that's a great idea! But I think we should call the strips "ribbons"...sounds way more sophisticated that way.
Oh come on, no one would be stupid enough to fall for that!
Sounds like CS4.
Canada: The US's more awesome sibling.
The issue is, it's only needless to *you*. Are we to assume that standards and abilities apply across the entire user base? For others, it might be a total clusterf*ck in the old version, and the new one clears up every single issue they ever had. I gotta tell you, I tried OO.org, and it felt like I was using Win95 again. I have no idea if it was actually a bad program or not, but I absolutely couldn't use it. I have it as an emergency app on my flash drive, but I pray I never have to use it. Being OSS, I think the strength would be that we can BOTH be satisfied. I cannot see any reason that they wouldn't be able to make a redesigned UI, and then someone can find a way to make a plug-in to emulate the old UI.
Canada: The US's more awesome sibling.
Double-click the tabs at the top of the ribbon, and the ribbon itself vanishes. Single click a tab to show it temporarily, like a menu. Double-click to get it to show permanently again.
This has existed since it was beta software called "Office 12." It's surprisingly poorly advertised; the behavior is logical but a little un-intuitive since most UIs don't do things like that, so a little user education would be smart.
There's no place I could be, since I've found Serenity...
Difficult for the Visually impaired?
This has the unfortunate implication of those who are not so great to look at. Perhaps the parent means vision-impaired.
I happen to Like Microsoft's 2008 ribbon. It got rid of the redundancy of having a toolbar and menus. It was a good idea on their part to simplify the interface in such a way. I'd hate to say it Microsoft devs once and a while do some things right.
RES PUBLICA NON DOMINETUR
Or for that matter, go to Customize, pull up "Commands Not in the Ribbon," and assign "ToggleFull" to the F11 key (the same key that is used in Firefox). Voila! No wasted screen real estate.
Breakfast served all day!
Simpsons did it
Haha... mercifully I haven't been exposed to Office 2007 so I didn't spot the joke.
-- Ed Avis ed@membled.com
You think that's bad? Try using MS Office 2007 on a eee. The bloody ribbon takes up a quarter of the freaking screen!
If you actually measure the height of menubar + default set of toolbars in Office 2003, and the height of ribbon in Office 2007, you'll see that ribbon is, in fact, slightly smaller.
The idea is NOT to give people choices.
Consistency is key to a UI. By giving a choice to switch you muddy that up. I'm STILL having to check when doing phone support if the control panel is in classic or catagory view. If MS had gone with just ONE then I would know exactly what they should be seeing. By not giving a choice they made it standard and now when I work with somebody on an office problem I know what interface to expect.
I suspect you'll be in a fairly small minority. The current UI not only looks dated (not a serious problem) but also has basic usability flaws all over it (which is). It's visually cluttered but inconsistent in where related commands appear on menus and toolbars, and there are numerous problems with interfaces for specific functionality (the styles and numbering features in Writer, formula editing in Calc, etc.).
If you follow the link to TFA and note the various priorities they describe there, I reckon they've got a pretty good handle on how a replacement UI should be designed, and if you consider how each point applies to the current UI, it's a pretty good checklist to identify the shortcomings.
If you disagree, post your argument. (-1, Overrated) isn't your personal censorship tool for views you don't like.