|
What about letting users configure it themselves? A simple routine to set the button text and dialog return value for all OK/Cancel buttons would do the trick. That way older users could keep doing things the way they are comfortable with, while new users could do things the "Microsoft Way" if they want.
Or you might even get rid of the Cancel buttons. I never saw much use in them myself. The "X" in the upper right corner can be used for the same functionality. I suppose it might make some users more secure having a specific button for abandoning changes, though, just so they could be sure their exit is done in a controlled manner. Although you can code the X to do exactly the same as a Cancel button, it might seem more unsafe to users.
|
|
|
|
|
What kind of weird design decision was it to have it the opposite way round to, probably, 99% of every other application out there?
.-.
|o,o|
,| _\=/_ .-""-.
||/_/_\_\ /[] _ _\
|_/|(_)|\\ _|_o_LII|_
\._. |\_/|"` |_| ==== |_|
|_|_| ||" || ||
|-|-| ||LI o ||
|_|_| ||'----'||
/_/ \_\ /__| |__\
|
|
|
|
|
You're not getting much joy from this one are you! I'm afraid I disagree with your stand on this one as well, standards and consistency are what make a user experience easy.
Having said that I can sympathise with you when YOUR standard pre-dates Microsofts latest set of decisions. I still use a DAL that pre dates Entity Framework by a decade 
Never underestimate the power of human stupidity
RAH
|
|
|
|
|
Standards and consistency should be handled in the design phase - NOT the test phase. The tester should just STFU, especially if this issue has been previously marked as "not a bug", or moved to the product backlog.
".45 ACP - because shooting twice is just silly" - JSOP, 2010
-----
You can never have too much ammo - unless you're swimming, or on fire. - JSOP, 2010
-----
When you pry the gun from my cold dead hands, be careful - the barrel will be very hot. - JSOP, 2013
|
|
|
|
|
John Simmons / outlaw programmer wrote: The tester should just STFU, especially if this issue has been previously marked as "not a bug"
Oh I agree with you there, especially if the original design pre dates the flavour of UI design. I just think the original design is flawed as I am pretty sure the dialog layout was set in the late 80s.
Never underestimate the power of human stupidity
RAH
|
|
|
|
|
When you consider the intended role of the tester, and his operational charter, my point renders the question of button order irrelevant.
".45 ACP - because shooting twice is just silly" - JSOP, 2010
-----
You can never have too much ammo - unless you're swimming, or on fire. - JSOP, 2010
-----
When you pry the gun from my cold dead hands, be careful - the barrel will be very hot. - JSOP, 2013
|
|
|
|
|
Mycroft Holmes wrote: I still use a DAL that pre dates Entity Framework by a decade
 As do I and probably most experienced practitioners.
You'll never get very far if all you do is follow instructions.
|
|
|
|
|
Colin Mullikin wrote: (OK in bottom right corner, Cancel to the left of it)
Which is the sane way to do it. When I first started using Windows in the 3.1 days, I kept clicking on the the Cancel button by accident because that intuitively seemed to be the most logical place for "continue on to the next dialog." I've always that that the way Microsoft decided to do it was bassackwards. And no, I don't believe Microsoft "studied" the problem or knows anything more about good UI design than I do. Just look at W8 for proof.
Colin Mullikin wrote: The next time he brings it up I might punch him in the face.
Nah, swat him on the butt and ask him if you broke his nose. Obviously he's got his head where his arse is, just like his idea for where the OK and Cancel buttons should be.
Marc
|
|
|
|
|
Consistency is the key...
But I'm curious. Am I the only person who uses .net msgbox's and hence to achieve consistency you have to use OK-Cancel.
|
|
|
|
|
Simple solution is to have a OkayCancelButtons composite control that you can use a style setting to choose which way round they appear. Default is Cancel/Okay and QA mode is Okay/Cancel
|
|
|
|
|
Doing this is entirely situational, switching the OK/Cancel buttons on a few dialogue boxes now and then so that the end user actually has to pay attention as to what they are doing rather than just clicking OK on everything while being a zoned out drone. Alternately you can randomise the OK/Cancel button so that on some dialogue boxes they are switched and on others not, but when the program is used a second time, the OK/Cancel buttons may be in the reverse order.
But if you don't mind your users being zoned out drones then there is no legitimate reason to switch them.
|
|
|
|
|
Does seem kinda dumb to have them the wrong way around
Windows has had them the same way around for way more than a decade
|
|
|
|
|
Change all the 'OK' buttons to 'This is what I want to do'
and 'Cancel' to 'I do not wish to change my mind'
The advantage of this is that you will have no room for any other buttons,
so you can't change their test.!
|
|
|
|
|
Colin Mullikin wrote: That is his one and only reason
Unfortunately it is a good reason.
One thing I learned is that users complain a lot, but very often they adapt within a few hours of the rollout. Those that can't adapt are those that couldn't work with it in the first place.
Besides 200 dialogs is not thát much. Yes it is a bloody annoying and tedious job. it will keep you busy for the better part of the day, but that's about it.
But I can imagine it's not much fun to be told to do something when you did it differently for so long.
|
|
|
|
|
But before they adapt, your highest paying customer forces the account manager to tell you to put it back as it was in the first place. Mean while, the lowest paying customers have already adapted to the change and complain when you change it back.
|
|
|
|
|
|
Well, you should have thought of that before you did such a piss poor design... 
Anything that is unrelated to elephants is irrelephant
Anonymous
-----
The problem with quotes on the internet is that you can never tell if they're genuine
Winston Churchill, 1944
-----
I'd just like a chance to prove that money can't make me happy.
Me, all the time
|
|
|
|
|
I came to Windows from RISC OS, where OK was in the bottom right, two decades ago, and I still think that it's the correct layout.
If MS thinks it's right for the order to be OK - Cancel, why don't they use Next - Prev too?
|
|
|
|
|
Way back, in the old days when MS and IBM were friends, was developed a standard called the Common User Access, CUA, giving the common guidelines for Windows, OS/2 and Motif user interfaces. The CUA was published as an IBM document, but was endorsed by MS (and I believe several other companies). Windows 3 was developed in close to 100% adherence to the CUA rules.
The CUA stated clearly that normal completion of a dialog is done by clicking the button in the lower right corner. In other words, OK to the right.
The first CUA rule (at least among the essential ones) were the location of the Help menu: CUA requires it to be pushed to the very right on the menu line. I don't remember when MS decided to move it together with the other pulldown menus; that could be in Win 3.11. We started out with consistency where you by instinct clicked the bottom right button to complete normally, to a transition period where an increaing fraction of the applications made you follow your instincts, swear, and redo the entire dialog, this time reading the button texts closely, to the current situation where everything is so inconsistent and free of rules that you cannot rely on instincts but must read all the buttons anyway. ("But OUR software is consistent" - sure, the good thing about standards is that there are so many to choose from. Most customers buy software from several different vendors.)
Whenever I make user interfaces, I follow the CUA rule of normal termination being the bottom right button. Nowadays, all user must read the button labels anyway, so this is as good a choice as any other. Or even better, since I can justify my choice by referring to a user interface standard document. (True enough: It was published more than 25 years ago, but it is far better than nothing).
|
|
|
|
|
Member 7989122 wrote: that could be in Win 3.11
No, Win 3.11 was still good.
Member 7989122 wrote: the location of the Help menu: CUA requires it to be pushed to the very right on the menu line
I remember that and I wish I could still put it there, but WinForms doesn't seem to allow it. 
You'll never get very far if all you do is follow instructions.
|
|
|
|
|
|
The period in my life when I had my highest frequency of swearing loudly at my office desk was when we used an editor that at exit might display a warning (this was before the mouse was common, so you responded through the keyboard:
"You have not saved the last modifications to the file. Do you want to save it before exiting? (Y/N)"
Of course you would immediately hit the "Y" key. Then we switched to another editor that had very much of the same style user interface, but it would give the warning:
"You have not saved the last modifications to the file. Do you really want to exit without saving? (Y/N)"
Obviously, from old habit, you hit the "Y" key. Then you would swear loudly and be grumpy for the rest of the day.
Moral: Don't play games with the user's old habits.
|
|
|
|
|
Exactly.
But your example is quite different from the subject.
|
|
|
|
|
Oh sure - there is no need to repeat exactly the same example.
I certainly would insist that it is not very different; it has to do with old habits. The CUA prescribed that the button to click to termminate a dialog normally should be the bottom rightmost one, and in the old days, all applications followed that recommendation. Then some software turned up where following the habit of clicking the bottom right button would NOT confirm all your entries, but cancel them.
That is very much the same as following your old habit of responding "Y" to a yes/no question, without reading the label. You loose your modifications is both cases.
|
|
|
|
|
His reasoning is valid. It may not be sufficient to justify changing it, either because user annoyance more than counters it or simply because it's a reasonably large cost for a reasonably small gain, so he shouldn't keep pushing it if he's been rebuffed. But inconsistency with standard OS practices is a valid thing for a tester to complain about.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
