Top Ten Persistent Design Flaws 1067
jlouderb writes "Bruce Tognazzini former human interface evangalist at Apple, and currently a principal at web design firm Neilsen Norman Group has begun cataloging the top ten design computing flaws that we just live with with, but shouldn't have to. Only seven are found at his article, and (not surprisingly) three are Mac related. My favorite: the mysteriously dimmed menu options. Why are those darned things grey anyway?"
Some of these things are valid... (Score:4, Informative)
Like the thing about disk removal. The only thing Windows handles being removed "gracefully" is a floppy (and I'd hardly say "gracefully", if you had a file open on the disk). And Mac OS could have done that, but the idea was to prevent the user from removing the disk until, say, its contents have been properly saved. So Windows let you remove a floppy. So what? What if you hadn't saved the file on it that you "meant" to? Then what? At least Mac OS enforced the proper order of operations, i.e., finish what you're doing with the disk first, then eject. To insinuate that Windows gracefully handles the unexpected removal of USB and/or FireWire external volumes is crap. Since Macs don't even have floppies anymore, and this argument doesn't apply to FireWire/USB volumes (though he implies that it does), this argument is somewhat moot.
And I can categorically say that his "computer not booting" story after he removed a FireWire drive is bullshit. If you remove the drive while it's asleep, yeah, it won't like that when it wakes up; usually, it will say a FireWire device has been removed before being unmounted. Worst case scenario would be rebooting the computer. But there is no way the computer just "wouldn't work" until the drive is plugged back in. That's just bollocks. Sounds like he had one bad/erratic experience that he thought was related to disk removal, and created this entire issue around it.
Other observations are kind of generic wishlists for the behavior of various features and functions. Some of them are frankly good ideas.
But when I read "Principle: The user is in charge and should be free to carry out any activity at any time without fear of reprisals" I just about lost my lunch.
Re:Some of these things are valid... (Score:3, Informative)
Actually Windows (XP) doesn't nag if i just yank my USB thumb drive out without doing the "Safely remove hardware" thing.
Yeah, it doesn't "nag"... (Score:5, Insightful)
What if a user has an open file, and yanks the drive? How does Windows "gracefully" deal with that? Answer: it can't.
You can pull the drive on a Mac, too - the difference is that the Mac will say, hey, you should have unmounted this first...hope you saved everything. And instead of doing something like auto-unmounting-without-nagging-when-no-files-are
Re:Yeah, it doesn't "nag"... (Score:4, Insightful)
Something along the lines of:
"Windows failed to write files to a volume. These files may have been lost."
No applications crashed, no nothing but that error.
That's about as gracefully as is necessary when it comes to the user purposely (or accidentally) abusing the computer.
Yeah - and that's fine. And Mac OS X does essentially the same thing. But Tog is somehow asserting that Windows does/did it "better", because it used to let you remove a floppy without doing something in the OS to unmount the volume. Huh? So what? A user could still screw up their data; they have LESS of a chance doing that when they're at least warned BEFORE they unexpectedly remove a volume.
Re:Some of these things are valid... (Score:4, Insightful)
Answer? Nothing about computers is 'intuitive' it's all learned behaviour. The fact that people actually whine and bitch about something that small makes me laugh, expecially now that in OS X the Trash turns into the Eject icon when you grab and move a removable disk.
Bruce has always been the ultimate whiner, in and amongst some of his valid critiques, and he still wants a computer to be a mindreading typewriter at the end of the day.
Re:Some of these things are valid... (Score:3, Funny)
Re:Some of these things are valid... (Score:5, Funny)
"The nipple is the ONLY intuitive interface. After that, it's all learned"
Re:Some of these things are valid... (Score:5, Interesting)
Re:Some of these things are valid... (Score:5, Funny)
Re:Some of these things are valid... (Score:5, Insightful)
The one I found funny was the continuous save. Computers "used" to do things that way (in the 70's) and if the power went out not only was your in memory copy bad, so was the one on disk because it was saving when the power went down (well back then it was on casette but the damage was the same) and is corrupted. Thats not even thinking about the fact that writing to disk all the time would slow the application down to the speed of molasses flowing uphill in January. This isn't to say that there is no happy medium. I find that 5 minute saves are plenty for me and I prefer them to go into a "backup" file that the application can handle instead of being saved in my actual document.
Re:Some of these things are valid... (Score:5, Insightful)
I doubt many applications would cause noticable performance degradations these days just by doing automatic saving. Save for a few specialty applications, there are more than enough idle cycles hanging around to do that work while the user picks his nose.
Re:Some of these things are valid... (Score:4, Informative)
Re:Some of these things are valid... (Score:4, Insightful)
If a computer crashed a dozen times a day and then always came back right where it stopped with all open documents fully recoverable, it would merely be an annoyance. Most people wouldn't care that the system was unstable. Those crashes would just give them a chance to stretch their legs for a minute while the computer comes back up. But instead, their computer crashes once every 3 months and they all too often wind up with documents that are completely unrecoverable, or a totally unbootable computer. Half a day's wasted work that must be rebuilt from scratch. That's the kind of thing that makes a guy pick up his keyboard and start beating on his monitor until it falls off his desk. We've all seen the video, and we've all felt exactly like that guy at least once in our computing career.
If someone would just take the time to come up with properly implemented full-data journaling for some common applications, they would make a fortune the likes of which Microsoft has never seen. I don't understand why common data loss is still acceptable. This is the 21st century after all. Computers have been around for half a century. Yet the closest I've seen is Word's auto-save and recover feature, which more often than not seems to fail to recover your file. Many times I've seen it "recover" on line or even nothing from a document that was many pages long. Not cool.
I tried to pitch an idea for application-level journaling on a BeOS developers' mailing list a few years back and got nothing but blank stares. As far as regular users are concerned, it would be the ultimate advancement in desktop computing, yet they (the developers) couldn't conceive any reason you'd want to do such a thing. "Get rid of one of the biggest annoyances of the whole computing experience? Why would we want to do that?"
Oh, well. Maybe in another 30 years, eh?
Re:Some of these things are valid... (Score:5, Insightful)
Answer? Nothing about computers is 'intuitive' it's all learned behaviour. The fact that people actually whine and bitch about something that small makes me laugh, expecially now that in OS X the Trash turns into the Eject icon when you grab and move a removable disk.
As the saying goes; the only intuitive interface is the nipple (and even that barely qualifies, some babies have a hard time coming to grips with it). But at least a user interface can be consistent. Dragging the floppy to the trash would suggest wiping the entire floppy disk, but it doesn't do that; in fact, it makes sure your files aren't deleted!
In fact, good graphical user interfaces are user-friendly (to neophytes at least) not just because they're consistent, but because they are modeless - vim is pretty consistant, but not modeless.
I think this is a justified gripe, now matter how easily it is learnt. Other user interfaces might have more deficiencies, and ones that are harder to overcome, but mac ain't perfect either.
MOD UP Re:Some of these things are valid... (Score:5, Interesting)
Re:MOD UP Re:Some of these things are valid... (Score:5, Insightful)
Re:MOD UP Re:Some of these things are valid... (Score:5, Funny)
Re:Some of these things are valid... (Score:3, Insightful)
Not only is it not intuitive, it's counter-intuitive. Can you comprehend the difference?
Only stupid and careless people can figure out the mac interface by themselves - intelligent, careful people won't perform certain experiments. Example: Nobody with any sense will ever drag a mounted network drive to the trash can, because that would erase their network drive. So
Re:Some of these things are valid... (Score:5, Informative)
And what will it do for all other objects? There is a difference between an intuitive interface and one that takes a metaphor to damn literal.
And finaly [kernelthread.com]:
Re:Some of these things are valid... (Score:4, Interesting)
Oh, look, right-clicking or ctrl-clicking* gives an option to "Eject [Volume Name]"!
Oh, and look what else--in Mac OS X 10.3, which came out over a year ago, most users are navigating their filesystems in a window like this [apple.com], where each ejectable volume has an "Eject button" right next to its icon and name (even in Open/Save dialogs!) One friggin' click! Or is that counter-intuitive for a Windows XP user, who has to locate and click the "Safely remove hardware" icon on the taskbar (which is represented by a tiny 3-D rendered grey rectangle and green left-pointing arrow, and may or may not be hidden as "inactive"), click the USB/1394 drive, click stop, confirm it by clicking stop again, then close that window.
Oh, and look what else, under every previous version of Mac OS/Mac OS X you could eject a disk by just hitting File->Eject (or Command-E on the keyboard). Network or removable.
The fact is, the trash-can eject is an old shortcut (whose origins have already been explained here) and which is still supported if you choose to use it, but which NO ONE EVER NEEDS TO KNOW OR USE ANYMORE. Just because it's a possible way to do it isn't reason to bitch, because there are at least three more intuitive ways to do it. Bitching about that would be just like bitching about the fact that you could open a terminal under linux and type "umount -f
____
* which is what I think you meant--Control, not Command, is the context menu key on Mac OS/Mac OS X.
Re:Some of these things are valid... (Score:4, Insightful)
This complaint is crap. You don't have to drag the disk to the Trash to eject it.
In Mac OS X you can also eject a disk by clicking the eject button in the Finder. Which makes good sense as a UI operation, especially since you "eject" other mediums (shares, usb disk, iPods,
ejecting disks (Score:4, Insightful)
But that's pretty irrelevant. Dragging the disk to the trash is a quick shortcut for skilled users, but has never been the primary method. The primary, normal method of ejecting a disk has always been the same way you perform actions on other icons: select it, then choose "Eject" from the "File" menu. No voodoo, no risk, no inconsistency.
Re:ejecting disks (Score:5, Funny)
Re:Some of these things are valid... (Score:5, Insightful)
As for the above, swap user with driver and you may see my point.
Agreed... (Score:3, Insightful)
Principles and lunches (Score:5, Funny)
It appears that everyone is guilty of having a framework. This guy, you, me, everyone. We think that what we experience in the world, and what we think about it, is all there is. We're all pretty small, even the wisest of us.
In this case, a Mac guy says the user is in charge, and thinks it's a law of nature.
Microsoft treats users as a renewable resource, to be used and reused as needed.
We Unix types, on the other hand, know that users are an unfortunate side effect.
Re:Some of these things are valid... (Score:3, Insightful)
And people wonder why software engineers get testy with designers sometimes. We're supposed to engineer systems that let users do whatever they want without reprisal. I can't think of anything else I use where I have that guarantee... even something as simple as using of humankind's oldest tools, the knife.
Re:Duh! Award Nominee (Score:5, Funny)
The #1 Design Flaw (Score:5, Funny)
Coral Cache Link (Score:5, Informative)
In My Book... (Score:5, Interesting)
Tight security where it doesn't matter and sloppy security where it does.
Inexplicable configuration. This is broad a broad item and includes buried preference settings where you'd never think to look, default settings to most frustrating (think Word), system settings under inappropriate categories and items with more than one relevence only found under one.
Pop-Up windows which steal focus immediately from whatever task has focus (active rather than passive bulletins) Ever been typing something, and hit ENTER just as something pops up? Gee, what the heck was that about?
Re:In My Book... (Score:5, Interesting)
And the corollary, applications that don't steal focus and don't create an entry in the taskbar - so they just sit there in behind your windows - like Winzip at its license screen. Did it finish loading? Where is it? Idunno. Or even worse, windows properties windows and the way they pile up back there.
But the biggest one: apps which can have a subform that disables access to the rest of the app, but if you move to another window and then move back, you can obscure the active subform with the disabled forms, leaving you with a missing form and a curiously locked application.
you can tell I'm a windozer can't you. And anyone who complains about mounting/unmounting should find out what an excruciating pani the old 3.5" drive is on a win box.
Re:And related... (Score:5, Insightful)
But, to be fair, many X apps do the same crap. Here's one thing about X-Windows (or Gnome maybe) that drives me nuts: Let's say I have four workspaces...I like to use workspace one for Internet-related activities, workspace two is development-related activities, workspace three is productivity-app hell, and workspace four is terminals. Now, let's say I go to workspace one and launch Mozilla...(really any app will do), then, while it's loading, I switch back to workspace two to continue debugging an app while Mozilla loads...then, BING! Mozilla pops on workspace two. Why won't an app stay on the workspace it was originally launched from? Does it have to follow me to the current active workspace?
I would think any app should be smart enough to do two things: (1) know where it is when it's launched and stay there; and, (2) know if it loses focus during start up and NOT re-take it. How hard could that be?
Power Failure Crash... (Score:5, Insightful)
Re:Power Failure Crash... (Score:5, Insightful)
UPS is ok to weather the power shortage, a battery inside the power supply would allow for clean shutdown.
Re:Power Failure Crash... (Score:4, Informative)
Trivial? Not really. Your power supply is probably at least 300 watts maximum output, right?
300 watts @ 12 volts = 25 amps. And that's assuming perfect efficiency (impossible).
You can get that from a lead acid battery, sure. You'll only quintuple the price of a power supply. Oh, and then there's the disposal issues and other environmental laws. Let's make that octuple.
Yeah, there's other batteries. No, almost none of them can be tossed, and they're all more expensive, too.
I've seen these supplies where the UPS is built in. They usually start at about $150 US...
Re:Power Failure Crash... (Score:4, Insightful)
You don't need the whole computer to work.
Put a chip on the motherboard to manage the whole thing. The OS gets an alert from the power supply that everything is about to die. It immediately dumps whatever it was doing to RAM, and gets the CPU to flush the write cache. Now, everything we need to save is in RAM (And that took probably a microsecond - the power would probably last this long without any backup.) The OS sends a message to the hibernation chip on the motherboard. This chip immediately cuts all power to the CPU and all peripherals except a single hard drive, and the RAM refresh (no fans, PCI bus, etc). It then does a DMA transfer of RAM to the hard drive, and sets a flag in CMOS for the next power-on to indicate that it needs to restore.
So, you need full power for about 1ms or so, and then power for one hard drive and RAM (no CPU) for about 30-60 seconds. That can't be more than a watt or two. If you were really slick you'd design any extra hard drives to put power back into the system as they spin down (regenerative braking - but we don't really need it). The power could probably be generated by standard-sized (AA/9V/C/D/etc) batteries - which are a trivial expense compared to UPS batteries.
Even a desktop running full-speed doesn't pull 300W - that is a peak capacity which is probably only seen when drives are spinning up initially.
Embedded system solved this two decades ago (Score:5, Interesting)
For the last ten years at least companies like MAXIM have been shipping zillions of WDT chips for use in embedded systems. They have all sorts of functionality and cost a dime. There's at least one in about every laptop. But because we have grown to expect our computers to be flaky and unreliable, there's no demand for robustness in desktop systems.
E=1/2CV^2
Most every PC power supply uses a switching convertor jsut like the one in a TV set (some even use the same control chips). They don't use bigass iron core transformers and they don't directly regulate to 12V or 5V or whatever - they use a DC bulk supply that is directly rectified from the AC line (yes, that's right, no transformer) and switch it down at high frequency (thus smaller transformers). A cheapass PC power supply might have (if you're lucky) 2x330uF of storage on this bulk supply - at 170VDC that's less than 10 joules, at 70% efficiency that's enough to drive 140W load maybe 50 mS.
All it would take to increase that to seconds is to add capacity to the bulk DC supply that's already part of every system. This would require getting larger caps to replace the cheap low value caps and a twenty cent varistor to limit inrush current so you don't blow the internal fuse simply by plugging it in.
They could even go to fullwave rectification on the input, use a 350VDC bulk supply instead of 170, and use 1/4th the capacity - a 2000uF/340VDC supply would have enough reserve to keep the entire system running a couple of seconds under "panic load." Stick a single 4700uF/450VDC cap in the "premium" power supply and you'd have a system that would be rock stable through just about anything.
The caps would cost $2-$5 instead of the twenty cent crap that's in there now; the sleep signal is already there, but no one uses it. Figure ten bucks to the end user and you have a system that will perform flawlessly through those little glitches and would have time enough to perform a proper shutdown on those rare glitches when the power didn't come back a second later.
Ten bucks. Maybe. But there's no demand for it because no one knows they could expect better at an equally reasonable price. Reviews don't even test for such basic functionality - no one has a clue, and the industry doesn't want you to know better because they would rather keep those pennies of profit themselves.
And BTW if you are feeling particularly sporty all it takes is a parts order and courage with a soldering iron. I've installed photoflash caps in TV sets to bolster the power supply and it works wonders - the cheapass Philips in my living room had this treatment and it's outlived two others and has a rock solid picture. My ancient HP Vectra firewall PC with the 233mhz cpu and the mighty 100W internal power supply coasts right through brownouts that caused my "better" desktop system to restart... that is, before I fixed it, too.
Re:Power Failure Crash... (Score:5, Insightful)
> shortage, a battery inside the power supply would allow for clean shutdown.
It shouldn't even need to be enough to shutdown -- all it needs is to dump the
RAM and processor state (register contents and such) to a designated area on
the hard drive (or flash RAM dedicated to this purpose, or whatever) from which
the BIOS firmware can restore everything when power comes back. The OS would
not even need to know the power was ever out, except to fix the system time.
Re:DC supply in the case? It's been done. (Score:3, Funny)
Because it would cost money.
Seriously though, I wonder if this guy also bitches about the fact that his car stops when he runs out of gas.
Re:Power Failure Crash... (Score:3, Interesting)
I remember reading about a OS that they demo'ed
by kicking the plug out of the wall.
After plugging it back in, the machine would
replay its "journal", and continue as if nothing
had happened.
If someone remembers the name of this system, or
has a link, that would help.
Re:Power Failure Crash... (Score:3, Funny)
but I love the free-verse format of your post.
Really, all
and main article submissions!
should be done that way.
The editors might do something then.
Mel would've liked that.
Dimmed menus (Score:5, Funny)
Re:Dimmed menus (Score:5, Informative)
Right click on the Taskbar and open up Properties. Then uncheck the 'Use Personalized Menus' box to disable it.
Re:Dimmed menus (Score:4, Informative)
Re:Dimmed menus (Score:3, Insightful)
Comment removed (Score:5, Informative)
scroll bars with ADD (Score:5, Insightful)
Its all part of MS's policy of torturing their users until they buy "intellimouses" with scroll wheels.
Re:scroll bars with ADD (Score:4, Informative)
Re:Dimmed menus (Score:5, Insightful)
And in Word it's not a case of 'least used menus'; I'm using word this very minute, and menu items that I've used, seconds apart, are always hidden ('minimal menu' mode for lack of a better or official term). So I'm wasting more time searching for menus than I should, and it's just totally annoying.
On the Written Word (Score:5, Interesting)
SOATYPICALSENTENCEWOULDREADLIKETHIS!
Oberammergaueralpenkräuterdelikatessenfr.. (Score:3, Funny)
(borrowed from here [serve.com])
grey doubt (Score:4, Interesting)
Oh well
Re:On the Written Word (Score:5, Informative)
Re:On the Written Word (Score:4, Funny)
Re:On the Written Word (Score:3, Insightful)
Lists (Score:5, Informative)
Or how about non-resizable dialogs with a set number of items in a list which displays all of the items minus one. WTF!?!
10 persisting people design flaws (Score:5, Funny)
#2-Thinking that computers do "magic", or at least should do to not have design flaws
#3
#4
#5
#6
#7
#8
#9
#10- Making top ten list without having 10 things to list
Re:10 persisting people design flaws (Score:5, Funny)
Design flaw # 11 (Score:5, Funny)
Using very large golden gradient shadowed GIFs each worth over 4K to represent the numbers 1 - 10 in a "Top Ten Persistent Design Flaws" webpage. It not only looks ugly, makes the website slower consuming more bandwidth, but it also takes away a good chunk of the left side of the page.
Cheers,
Adolfo
GUI design - favorite site (Score:5, Informative)
http://digilander.libero.it/chiediloapippo/Engine
Check out the hall of shame section, it's hilarious!
PS - this link is a mirror of the original site
Stealing Focus (Score:5, Insightful)
You lose your thread of thought AND the computer decides you said "OK" to "do you want to email your credit cards around the world" while you sit there wondering what just happened.
Reverse dates (Score:5, Insightful)
Re:Reverse dates (Score:5, Interesting)
Ah, someone else that agrees with me on that!
The US style of writing dates (and I live in the US) drive me completely batty. MM/DD/YY? No! That makes no sense. YYYY-MM-DD makes the meaning far more clear, and you can even extend it arbitrarily... YYYY-MM-DD-HH-MM-SS-uu.
As an aside, how often do you have secretaries and public clerk type people (ie, the DMV) freak out on you because you write dates like that?
I often get "How long did you serve", since apparently the military (only some branches? no clue, just speculation) encourages that date format.
I have learned that any answer involving the phrase "lexical order" will only result in blank stares.
Re:Reverse dates (Score:5, Informative)
So you people who still insist on MM/DD/YY, you are OLD AND BUSTED.
YYYY-MM-DD = NEW HOTNESS.
MM/DD/YY = OLD AND BUSTED.
Re:Reverse dates (Score:4, Insightful)
YYYY-MM-DD makes sense because it's listed in descending order of unit times. It's like a numbering system, with most significant digits first.
MM/DD/(YY)YY makes sense because many people write their dates like October 11, 2004.
If you have to communicate with people, don't be a lazy ass and write out the name of the month, to remove ambiguity. If you have to communicate with machines (or if you like to think this way, like me) then use the YYYY-MM-DD form.
Article not quite right... (Score:3, Informative)
Um, hate to burst your bubble, but MS GUI does not recover smoothly from such events, unless one considers a BSOD smooth recovery. Since Windows 95, and still today in Windows XP, removing a CD or floppy from the drive before Windows is finished with it will result in the system hanging at best, and BSOD at worst. Not exactly what most people would consider smooth operation.
Neither Linux nor Apple nor Microsoft correctly address the problem of removable media:
He has such a hard on about the Dock (Score:5, Funny)
Duration: [in years]: seems like a thousand centuries ago...
Supplier: Tog
Alias: "I have no concept of the difference between objective and subjective usability complaints."
Product: Tog's parents.
Bug: Tog's perceptual abilities.
Class of error: Intellectualy density.
Principle: "My opinions are holy."
Proposed Fix: Zoloft
Discussion: Some of the things he lists as flaws in the Dock are things that I acutally like about the Dock. It's a very subjective thing. Some of the things he laments losing from Mac OS 9 were not the bee's knees he seems to imagine they were. He was just used to them, is all.
Bug first observed: Can't check the date on the original Dock whinefest because his site is slashdotted. It happened some time after Tog ceased to be relevant.
Observer: Harvey Birdman, Attorney At Law
Bug reported to supplier: No. No point. You cannot argue with self-proclaimed learned wisemen.
Bug on list since: Whinefest first published.
Re:He has such a hard on about the Dock (Score:5, Insightful)
Or another Tog bug, based on 5.
Bug Name: Tog knows nothing about the history of the web.
Duration: Just discovered, but probably years.
Supplier: Tog
Alias: "I'm trying to impress you because I used the web WAY before you chowderheads did."
Product: Tog's Design Flaws list
Bug: Tog's incorrect memory of history.
Principle: "I will spout off knowing nothing about what I'm talking about."
Proposed Fix: Lateral Cranial Impact Enhancer of your choice.
Discussion: Claims to have reported URL space bugs to Netscape in 1991 and Microsoft in 1992. However, Microsoft didn't have a web browser until 1995 and Netscape didn't even exist in 1991.
Bug First Observed: Today.
Observer: Hopefully, the greater part of the Slashdot readership.
Bug reported to supplier: Ha!
Bug on list since: about now.
Not Bugs, Maybe Not Really a Problem, Either (Score:5, Insightful)
In my experience, this affected Macs much less in a brownout situation than PCs. The Macs (at the time, desktop G3 systems) stayed up after a power blink of 0.5 sec, losing no data. I think current Mac OS hardware is more robust in this area, but this is not really a fault of the computer or the OS. No power, no computer worky. Sorry.
Workaround in a mission-critical area: Buy an uninterruptable power supply, petition Apple to make a computer with very expensive (but non-volatile) flash RAM, or use an Apple laptop, which has its own battery that makes it resistant to brownouts and blackouts.
Issue 2: The Dock in Mac OS X.
Grousing. In the old Mac OS 9 days, there was a Dock analogue called the Launcher. It was ugly, and I rarely set it up for users, but it worked. Some people still use it for their Classic apps in OS X.
Workaround: Many, most third-party. Apple's interface, until OS X was icon-centric for launching apps, rather than menu-centric (in Windows Start menu). The Dock is no more perfect than the Start menu, but at least it provides a consistent launcher for common apps, instead of having the user search through folders for the right app icon to launch.
Better: Have installers ask user to add icon for applications to the Dock, which isn't done most of the time, forcing users to search about in the Applications folder.
Issue 3: Dimmed menus.
A bit of a grouse, but logical. Some OS X apps by third parties HAVE shown info in the greyed out menu as to why the option is not available. I believe it is more programming efficient to leave a greyed out menu than to attempt to hide it (affecting where and the order of menus on the menu bar from one moment to the next, which would confuse the hell out of me).
I believe Tog's thought, of adding a special option in a greyed-out menu as to why this command is dimmed, could be useful. Otherwise I think he is blowing the issue up. Of course, the more complex the app (especially with palettes and THEIR commands, the more weight his argument holds.
Issue Seven: Disk Drive Nazi.
Not a problem, at least until removeable drives arrived.
The Mac OS has always been intelligent, preventing you as the user from accidentally ejecting or formatting a disk you are using, including network devices. This is a Good Thing. Compare this to the behavior in Windows, which will still allow you to eject media in use, causing All Kinds of Hell.
Workaround: His point seems more specific to USB and FireWire drives. Unless Apple creates a hardware lock that physically locks a device, preventing the thing from being removed, then there's not a lot to do there, except Apple making the OS more robust in screaming at people to tell the OS that the drive is to be disconnected before they physically remove it.
Good so far (Score:3, Insightful)
And on that note, why can't the BIOS battery be rechargable? Why should my computer *ever* think it's 1969, or 1980, or 1984?
Mirrordot Link (Score:3, Informative)
text in full (Score:3, Informative)
In some cases, the bugs have outlasted the original developers, persisting so long that their successors may not even realize they are bugs--they seem the result of "natural laws." In other cases, the developers know these bugs full well, but refuse to address them. These all need to be addressed, and that address should be far out of town.
Bug Name: Power Failure Crash
Duration: >30 years
Supplier: Desktop computer manufacturers
Alias: "Oh, Sh--!"
Product: Desktop computers worldwide
Bug: If the computer loses power for more than a few thousanths of a second, it throws everything away.
Class of error: "That's the way Grandpa did it..."
Principle: Protect the User's Work
Discussion: Somehow, the most destructive act a computer can carry out, other than destroying the contents of a hard disk, got "grandfathered in." Somehow it became OK for computers to just die if the power fails.
If cars modelled this behavior, you might drive your car from New York to Miami, run out of gas in Fort Lauderdale, 10 miles from your destination, and suddenly find yourself back in New York.
Immediate Fix: Web Developers
Store (encrypted) information in cookies even before transfer to the server, so information is preserved from all but the most serious "melt-downs."
Proposed Fix: Application Developers
Convert your existing software and write new software to perform Continuous Save, so users cannot lose more than the last few characters typed or gestures entered. Do not fail to provide sufficient Undo and Revert facilities enabling users to get back to where they were before they started doing the wrong thing.
For all the drawbacks of the crude system most applications have had until now, one advantage was that new drafts did not take the place of old until we said so.
Oh, and by the way, a dialog saying, "This action cannot be undone. OK Cancel," is not a suitable substitute for a Revert facility for anything at any time.
Proposed Fix: OS's
Build support for Continous Save and Revert into the toolbox.
Proposed Fix: Computer Hardware
Add very short term batteries or tantalum capacitors to systems with volatile memory with enough power to dump the memory to disk and go into hibernation, perhaps 30 to 45 second worth.
Bug first observed: 1976
Observer: Tog
Bug reported to Apple: 5 Mar 1985. Quote from that memo:
The age of computers that die when the power goes off will fade to an interesting footnote in history, just as radio gave way to TV. The question is not whether Apple will [address the problem], but when. I believe the time is now....We
have the opportunity to add another dimension to computers; let us take it.
Should happen any day now...
Bug on list since:List inception: 1 Dec 2004
Bug Name:The Macintosh Dock
Duration:Four and counting
Supplier:Apple Computer, Inc.
Alias:"The Cool Demo"
Product:Mac OS X
Bug:There are actually nine separate and distinct design bugs in the Dock, probably a record for a single object. You can read about them all in my Article, "The Top 9 Reasons the Dock Still Sucks."
Class of error:Confusing a demo with a product
Principle:Demos and products are two separate entities. The Demo's purpose in life is to help sell the product. The product's purpose is to serve the user.
Proposed Fix:Leave the Dock just as is. It looks great on stage durin
Eight (Score:5, Insightful)
Bug name: PDF
Duration: 10+?
Supplier: Adobe et al
Alias: Why-is-it-so-hard-to-write-decent-software?
Product: Various PDF viewers, primarily for Windows
Bugs: One: Acrobat kills Mozilla. Two: Hidden "check for updates?" box locks up IE.
Class of error: Poorly written software
Principle: Simple software shouldn't hog resources or kill other apps.
Discussion:
Why is it so hard to write a decent PDF reader? Preview for Mac is fast and doesn't crash anything. And yet Acrobat for Windows (and maybe for Mac--I haven't tried it) is slow, a resource hog, locks up Mozilla/Firefox until the file is done loading, hides its "check for updates" window (without a tab on the XP app bar), and locks up the PDF-viewing window in IE until the "check for updates" box is dealt with.
Re:Eight (Score:5, Interesting)
I think FireFox would be very well suited to build it's own PDF viewer into the core code (or at least promote a module for Win32 users that ISN'T Adobe's).
Another big bug in Acrobat reader is that if you're in FireFox and try to issue a keyboard command while reading a
If the Mozilla foundation included sane PDF capability, it would end up in even BETTER perception of improved response. Uninformed users automatically make the psychological connection of "poor Acrobat PDF performance"+"IE"="poor IE performance".
I think it would be wise for the FireFox crew to capitalize on this because it would give the user an [even] better browser experience (on Win32).
menus are grey because they're disabled, get help. (Score:3, Insightful)
Uh, they're grey because they're disabled ?
I'm sorry, I don't understand how this is a design flaw. You'd rather the menu in question _always_ do something? What do you want copied when nothing is selected? Would you rather the menu was enabled always, but just beeped or did something else ( i.e. not the desired action ) when clicked ?
The menu is grey to let you know it can't do anything until some other action is taken. It doesn't just disappear because location/muscle memory is how we remember where that menu is. What would be the better design? How is a disabled menu a flaw, again? You'd rather get a dialog box telling you that you need to do something before clicking here... how could you have known ? Why is clicking "copy" bringing up a dialog box ?
TSFA says :
The software "knows" why it has dimmed the item. Some decision or decisions led to the flag being set. At the same time as the flag is set, the reason why should be made available. If the user clicks on a grayed-out option, the reason or reasons should be made known. And none of those, "Gosh, Oh, Gee, it could be any one of these 14 reasons or maybe something else" messages. The message needs to be intelligent, responsive, and accurate. This one is important. This one needs to be done right.
Ok, so the issue is that you want to know why the menu is disabled. So, which of 20 different on-screen objects do you want a message to indicate could be selected to enable "copy" ? Even if you manage to get the message "right", how useful is the message "You must select something to copy." going to be after the second time you see it? At that point the greyed menu tells me everything I needed to know.
Gee, I wonder why that one hasn't been fixed. Yea, that's a real design bug, right there. Just like the dock, which even my mother-in-law can use, with it's 9 bugs and all...
Now, ASCII sort and reasonably flexible data entry ( aka Bug Name: Let's you save me some work ) now, those are real design bugs. Design bugs which are usually there ( as the article notes ) doe to lazyness of the software designers/creators.
A few of these design problems I can agree with, but IMHO, if you're troubled by a disabled menu, that's a clear sign you don't understand the function of that menu, and you might want to try a menu item that isn't greyed out, like that one labeled "Help".
Re:menus are grey because they're disabled, get he (Score:4, Insightful)
This argument doesn't seem very consistent. You're suggesting that the first time you see something and it's not obvious to you, it should tell you so you'll know, but at that point, it's not useful to you. What if someone else is using your computer and has not seen that message? Would it be useful then?
Do you realize that there are more menu items than just ``copy''? I use a lot of applications with a lot of menu items (i.e. Final Cut HD) that will occasionally have something that sounds like what I want, but it's greyed out. Why would I not want immediately contextual information describing what I need to do? Is it really practical to suggest that I pull out the manuals and try to figure out what all is required to use something when I could just hit the brief contextual help?
A more concrete example: I'm in gimp which is giving me the option to scale my image, but not crop it. Why is that? Why can I move this layer down once, but not twice? I happen to know these answers, but it wasn't very long ago that I did not, and it was frustrating to want to bring a layer to the bottom and having gimp just refuse to do so with no explanation as to why (which was added in 2.something...but not on the menu).
Comments (Score:5, Insightful)
This is due to the save file paradigm. Changes only get saved if you tell the computer to. People have long realized this is bad; this is why some programs have autosave. I am all for saving changes continuously - and forking a file if you want to have distinct versions.
The Macintosh Dock
I guess this is more of a personal thing. Personally, I think the Dock is great, although I prefer separate launch icons and open window icons (aligned at separate edges of the screen), a la NEXTSTEP. The Mac doc certainly kicks the Windows taskbar (and imitations') ass.
Mysteriously dimmed menu items
I don't necessarily agree these are bad. The alternatives are removing them (bad because menu structure changes), not disabling them (makes no sense - they are disabled because they aren't meaningful right now), or not dimming them (bad because you don't signal the action is unavailable).
The proposed fix is a good idea, though.
ASCII Sort
This issue has never affected me much. The alternative is is having lots of black magic exceptions to get items sorted the way humans might sort them. To me, it seems these exceptions are hard to deal with for machines, but for humans as well. I don't think it's worth the trouble.
What is good, though, is having proper metadata support, so that we can sort not just by filename, but also by author, project, modification time, etc. Add in a search function, and you don't even notice the asciibetical sorting anymore.
URL Naming Bug
Some browsers already convert spaces in URLs to '%20' or '+'. I think this is the way to go. I'm not sure if stripping spaces (as the author suggests) is a good idea. Does he mean to make "my birthday pictures" internally translate to "mybirthdaypictures"? Why? My filesystem can deal with spaces just fine. Perhaps stripping all spaces after the first (i.e. removing errorneous spaces) is a better option.
Let's you save me some work
So, not accepting multiple formats for the same data is bad. I have to ask why the multiple formats exist in the first place. If we're talking about SSN, library card number, etc. there's always an authority issuing these numbers. Why not use the same format they use, everywhere? If users want to be inconsistent, they must be prepared to deal with the consequences.
The Disk Drive Nazi
I, too, hate that machines don't let me have my device back. Linux and BSD (and probably other unices) can be particularly annoying in this respect. Someone once tripped over the USB cable of my webcam, unplugging it. Nothing but a reboot would let me kill the program (which was in uniterruptible sleep), reload the (confused) driver, plug the cam back in, and start streaming video again. Grrr. Isn't this what exceptions are for?
Spaces in filenames (Score:3, Insightful)
> People separated written words with spaces from the time writing was invented up until around 30 years ago whenaspacebecameavaluableobjectnottobewasted
Tog, that whistling sound is the point going over your head.
30 years ago, we took spaces out of filenames not because we needed to save characters, but because we were all using a CLI, and we did it because we were using spaces to separate words.
then: vi ~fredfoo/stupidapp/stupid.cfg
now: vi C:\Documents an
Drenched in irony (Score:5, Insightful)
1) No alt tags - you've used images to number your list, yet no alternative text for blind users (or those with images off), this is a very well established as bad usability
2) Confusing title - you say top 10, but don't actually have 10 items on your list, an important aspect of usability is clarity, which your title lacks.
3) Consistency - you've divided each item into sub-sections, yet the sub-sections are inconsistent with from one item on the list to the next. If a sub-section is not applicable, I suggest you add, for instance: "History: N/A," this will save readers scrolling back and forth for a section they might believe to have missed
4) No submission form - You provide an option for people to submit suggestions for your list, yet fail to provide a basic HTML form for them to do this, instead you opt to let them do the work.
There are more, but I'll stop here, since I expect this to be modded down anyway. I hope you see the irony.
Sin number 0. (Score:3, Insightful)
If you are not playing quake or starcraft, a mouse is just a luxury. Design to avoid its use.
--
Wiki de Ciencia Ficcion y Fantasia [uchile.cl]
Re:Sin number 0. (Score:4, Insightful)
Actually, in the early days of the Mac, the rule was in reverse. That is, everything should have been possible without a keyboard, without having to emulate one. Keyboardless Macs were actually common during the 68k era; they were used for kiosks, printing stations, status checking and other tasks which didn't require data entry.
For every user who has trouble manipulating a mouse, there is a user who has trouble dealing with a keyboard. This notion that 2-D manipulators are a inconvenient UI concept boggles my mind; I just don't see how you can use software like graphics editing, page layout, or reality simulation effectively without some form of input from a mouse, trackball, or tablet...
My take on their take. And I'm not too kind... (Score:3, Interesting)
Dynamic electronics tend to like power. That's in their nature. If that upsets you, use static RAM (which doesn't need refreshing, so has a retention level that can handle power spikes and stuff) or FLASH RAM (which retains data indefinitely without power). You get a performance hit, but you can't always get it all ways. If you absolutely must have the performance, then use full log journalling for all transactions. If you can't afford to be down, then use hot-standby High Availability. So what's your excuse for ignoring what is already out there?
Dimmed menus work a hell of a lot better than not having the options at all, which is a popular alternative. (It's popular, because you don't end up with a gazillion greyed options cluttering your menus.) The problem is not the dimming, the problem is that menus are too big. The dimming has nothing to do with it. Keep It Simple, Stupid is the only bug you can legitimately claim here.
DOS and Windows' DOS shell don't sort at all. Windows' GUI doesn't auto-place unless you tell it to (and even then it can require a crowbar and the suitable application of threats). Unix and all derivatives allow you to pipe ls through any text processing code you like, and GNU's ls has so many sorting options built in that you almost don't need to do that. If this is an Apple bug, don't blame everyone else.
I know of no browser on Earth that doesn't allow you to escape a space. What, that's extra typing? That's not the bug described. The bug described is that spaces "aren't allowed". You know what? Yes they are. Even if your browser won't let you enter spaces directly (and I know of none like that), there are ways round it. All you need is something that swaps spaces for a %20. What, you can't do macros? Don't blame software engineers. Maybe blame your browser, but most likely you need to take a good look in the mirror and blame that person instead.
This is one of the few genuine bugs I've seen on the list. And it's not exactly unique to computers. It's also nearly unsolvable. Let's take the date 01/02/03. Is that European format? (February 1st, 2003), American format? (January 2nd, 2003), or International format? (February 3rd, 2001). You can't tell from that information, as it's ambiguous. That's a good word to learn, in computing. Computers don't like ambiguity. You'd need an additional drop-down menu, from which you would pick the format. For EVERY data entry panel. The format list would be between 4-2,000 entries long, depending on the type of data. You don't think that would confuse the users a hell of a lot more?
Many problems fall in this list.
As for problems with docking bars, the Windows GUI, etc, that's what FTP sites are for. Prefer a UNIX-like environment to MS Windows' desktop? Just download Afterstep for Windows, or use the X11 package from Cygwin.
If solutions exist, but you persist in living in the problem, why the hell should anyone feel sorry for you? This list is about as valuable as a Windows user complaining about all the security holes and speed issues, KNOWING there's plenty of free alternatives, but CHOOSING to ignore them, because only by staying with Windows can they continue to feel sorry for themselves.
When living in misery is a choice, the misery ceases to be a defect of other people and their work.
modal dialog error message (Score:3, Insightful)
Rebuttal (Score:5, Insightful)
1) Power Failure Crash
-- A "Continuous Save" is unpractical. Committing every action to permanent storage, aka a hard drive, would both kill performance and shorten the drive life. It would also increase the risk of hard drive failure during the crash by increasing the likelihood that the drive would be in use.
2) The Macintosh Dock
-- "It's not that the Dock sucks so much as a productivity tool as it is that Apple threw away so many more powerful, useful objects in its favor." So it works, but there were better options in his opinion? You'd be hard pressed to find anything that couldn't be described in this way.
3) Mysteriously dimmed menu items
-- I can see the point of wanting them to say why, but it is very short sighted to say the message must be exact. A much better solution is that in Help, every menu option should be searchable and explain exactly when it can be used and how. (Though I miss the Apple Help Balloons. Heck, now that I think about it I think they worked and could explain disabled Menu Options but no one bothered to fill them out.)
4) ASCII Sort
-- This is a consistent extension of alphabetic sorts, and will likely never change in standard file system listings. The example of iTunes is a specific application with a specific data set, and any application should organize data as appropriate for the use. Part of the point of iTunes IS to organize files in a way that makes sense for what they are, whereas the operating system must treat all file names equally and not make assumptions about what they represent.
5) URL Naming Bug
-- Correct history: filenames didn't have spaces because the early command line parsers separated tokens by spaces. Even today, command line parsers need help either by quoting the entire name or escaping the spaces. (The Apple II worked because the parser was even simpler; every command was only one word and everything afterwards named the object to be acted upon.) The problem with the proposed fix is that the only place spaces are not allowed is in the machine address part; spaces are allowed willy nilly in the directory portion as per the server's setup. There is no consistent way to know whether spaces in that portion should be dropped. While the browser could be written to automatically remove spaces in the first portion doing so in the directory portion would be disastrous for many web sights. Having it do both would seem to be a blatant inconsistency.
6) Let's you save me some work
-- This is actually reasonable, and as a programmer it's a pet peeve of mine that the computer should do the work. It's not always possible though, and sometimes compromises must be made. I prefer if the field only wants numbers it would say so and prevent numbers from being typed without beeping or anything. I think it's a good compromise between getting a clean entry and not interfering with the user, since any spaces/dashes would just be ignored.
7) The Disk Drive Nazi
-- This was a feature. It prevented floppy or system corruption. (The System was on a floppy and could otherwise be ejected.) OS X is much more dynamic in recovering from these incidents, having to deal with USB, Firewire, and Network drives. The incident with the Powerbook described is most likely the result of using a non-Apple drive with a bad driver. Booting from an emergency CD would eliminate. Given the author's history it's even possible he was using OS 9, increasing the likelihood of a driver problem.
8) 9) 10)
Apparently, he's counting in base 7.
Rebus icons (Score:5, Interesting)
Duration: 15+ years
Supplier: Eudora, Rational (now part of IBM)
Alias: "Let's play a game - can you guess what this means?"
Product: Eudora mail reader, Atria/Rational/IBM ClearCase revision
control system.
Bug: Eudora: The 'check mail' icon has a picture of an envelope and a
"check" mark. ClearCase: The 'check in' icon has a picture of a
document, an arrow (in a direction arbitrarily ordained to be 'in') and
a "check" mark.
Notice the use of the "check" mark to imply the English word "check".
Not only is this going to be completely opaque to every non-English
speaker, it is very murky to about half of the world's English speaking
population also. "Check" is the American name for this mark, in British
(and Australian, New Zealand...) English it is a "tick" mark. It took me
two years before I realized why it was on Eudora's "check mail" button.
Discussion:
Icons are supposed to transcend language barriers - not to limit
themselves to one dialect. A related bug are the highly stylized icons
found on Swedish home appliances: circles, crosses, dotted arcs etc.
These are quite incomprehensible without a manual, which likely has been
lost. If they just wrote Swedish words, at least I can find a
Swedish/English dictionary in my local library.
Bug first observed: c1987, "Eudora" mail reader, c2000, Rational (now
IBM) ClearCase.
Bug reported to supplier: Reported to Rational c2000. They told me where
I could find the bitmap file for the icon so I could edit it myself.
---------------------
As an aside: I expect this one has long since been fixed. Macintosh,
c1990, in a shared computer room environment: You'd start using a
computer, and at some point the computer would demand that you insert
some floppy disk. Said disk belonged to the previous user of the
computer, who has left. The computer would refuse to do anything at all
until you supplied the (unavailable) missing disk. The only solution I
knew of to this was a reboot.
My top-10 design flaws (Score:4, Insightful)
2. Expecting that users will (or should have to) read anything.
3. Expecting that users will (or should have to) possess technical expertise or jargon.
4. Expecting that users will (or should have to) configure it before using it.
5. Guessing or questioning the user's intent.
6. Neglecting to handle all possible failure cases gracefully.
7. Neglecting to save state frequently enough or at all.
8. Pointless rearchitectures (if it ain't broke, don't fix it).
9. Avoiding necessary rearchitectures (you have to break a few eggs to make an omelette).
10. Designing based on your own motives (in-product advertising, etc) rather than on the user's needs.
From that mysterious text called the article: (Score:3, Informative)
That's a solution? (Score:4, Insightful)
Edit -> Undo
"You don't have anything to undo"
Edit -> Redo
"You don't have anything to redo"
Edit -> Cut
"You haven't selected any text to cut"
Edit -> Copy
"You haven't selected any text to copy"
Edit -> Paste
"You haven't copied any text to paste"
Great, one more way for my computer to treat me like a complete imbecile.
If an option is greyed out, it's usually because -- shocking -- you can't use it right now. This is Common Sense. If it's not Common Sense, it's because that application's UI designer made their menus too complicated to begin with, and in my experience software programmers who do that sort of thing would also make their pop-up help even more useless, something like: "This option is disabled because you can't use it right now."
Rule #1 in UI design: if you have to explain something to your user, you're doing it wrong. Or at least you're doing it inconsistently, which is the same thing in this business. I shouldn't need to wonder WHY an option is disabled, at if for some reason I should, it shouldn't be disabled at all.
Re:That's a solution? (Score:4, Insightful)
Re:I agree on the dimmed menus (Score:4, Interesting)
Most of the guys other items were just kind of "blah" to me - the dock, removal of hard drives from the powerbook, but the "grayed out for no reason" at least made some sense.
Re:I agree on the dimmed menus (Score:4, Funny)
Slashdot Design Flaw (Score:5, Insightful)
Example: A post of "Often it is difficult to figure out why certain options are dimmed and under what context they will become active. I don't see a better alternative though other than better documentation..." attached to a story containing the solution of "Make grayed-out objects clickable, revealing what has caused the object to be dimmed and what the user can do about it."
First Noticed: 1996
Proposed Solution: Require the user to read the article. This could be implemented in a number of ways: either the referring home page to the message board should BE the article, or a page between the story and the article should contain some sort of code permitting posting. Or a mod of "-9999999, RTFA" should be added.
Re:I agree on the dimmed menus (Score:3, Funny)
Re:I agree on the dimmed menus (Score:5, Funny)
No, if you really want to confuse the user, simply create the menu dynamically, picking 6 items apparently at random to put on the menu. Microsoft can't be wrong here, they have user-interface guidelines and everything. After a few seconds, when the user has had time to read most of the menu items, change them again, this time picking 12 items at random.
If you can, use two columns, and put an animation in so that the menu takes half a second to appear.
Re:/. ed already? (Score:3, Informative)
10 Bugs [nyud.net]
Re:Number 5 (Score:3, Insightful)
Firefox will not convert www.barnes and no
Re:Number 5 (Score:3, Funny)
And what if you wanted a picture tour of the barns of british nobility?
Re:Number 5 (Score:3, Funny)
NAME: newyorktimes.com
ADDRESS: 199.239.137.217
go to http://newyorktimes.com and it redirects you to nytimes.com...
I am being pulled in by a troll, aren't I?
Re:Dim consistency (Score:4, Informative)
Long ago, Balloon Help on the Mac did something like what he's suggesting. When you'd hover over a menu item it would pop up a balloon (tooltip) explaining what the item did. If you hovered over a dimmed item, it explained what the item did and also went on to explain why it was not available at the moment.
I don't believe that dimmed items are inherently confusing -- I know perfectly well why Firefox has dimmed my Cut and Copy commands right now -- it's because I don't have anything selected. On the other hand, I have no idea why Outlook Express has "Block Sender" (under the Message menu) dimmed while I've got a message selected in my Inbox. It'd be nice if I could easily find out ("This command is disabled because you don't have message filtering enabled" or "You must read the message first" or whatever the reason may be).