| Date: | October 12, 2004 / year-entry #364 |
| Tags: | other |
| Orig Link: | https://blogs.msdn.microsoft.com/oldnewthing/20041012-00/?p=37593 |
| Comments: | 81 |
| Summary: | Philip Su's discussion of early versions of Microsoft Money triggered a topic that had been sitting in the back of my mind for a while: That people lie on surveys and focus groups, often unwittingly. I can think of three types of lies offhand. (I'm not counting malicious lying; that is, intentional lying for the... |
Philip Su's discussion of early versions of Microsoft Money triggered a topic that had been sitting in the back of my mind for a while: That people lie on surveys and focus groups, often unwittingly. I can think of three types of lies offhand. (I'm not counting malicious lying; that is, intentional lying for the purpose of undermining the results of the survey or focus group.) First, people lie about the reasons why they do things.
This is one of those unconscious lies. People claim that they want a computer to do their personal finances, to organize their recipes, to mail-merge their Christmas card labels. They are lying. Those are the things people wish they would use their computer for. That's before the reality hits them of how much work it is to track every expenditure, transcribe every recipe, type in every address. In reality, they end up using the computer to play video games, surf the web, and email jokes to each other. Just because people say they would do something doesn't mean they will. That leads to the second class of focus group lying: The polite lie, also known as "say what the sponsor wants to hear". The following story is true, but the names have been changed. A company conducted focus groups for their Product X, which had as its main competitor Product Q. They asked people who were using Product Q, "Why do you use Product Q instead of Product X?" The respondents gave their reasons: "Because Product Q has feature F," "Because Product Q performs G faster," "Because Product Q lets me do activity H." They added, "If Product X did all that and was cheaper, we'd switch to it." Armed with this valuable insight, the company expended time, effort, and money in adding feature F to Product X, making Product X do G faster, and adding the ability to do activity H. They lowered the price and sat back and waited for the customers to beat a path to their door. But the customers didn't come. Why not? Because the customers were lying. In reality, they had no intention of switching from Product Q to Product X at all. They grew up with Product Q, they were used to the way Product Q worked, they simply liked Product Q. Product Q had what in the hot bubble-days was called "mindshare", but what in older days was called "brand loyalty" or just "inertia". When asked to justify why they preferred Product Q, the people in the focus group couldn't say, "I don't know; I just like it." That would be perceived as an "unhelpful" answer, and besides it would be subconsciously admitting that they were being manipulated by Product Q's marketing! Instead, they made up reasons to justify their preference to themselves and consequently to the sponsor of the focus group. Result: Company wastes tremendous effort on the wrong thing. (Closely related to this is the phenomenon of saying—and even believing—"I'd pay ten bucks for that!" Yet when the opportunity arises to buy it for $10, you decline. I do this myself.) The third example of lying that occurred to me is the one where you don't even realize that you are contradicting yourself. My favorite example of this was a poll on the subject of congestion charging on highways in the United States. The idea behind congestion charging is to create a toll road and vary the cost of driving on the road depending on how heavy traffic is. Respondents were asked two questions:
Surprisingly, most people answered "No" to the first question and "Yes" to the second. But if you stop and think about it, if people avoid driving during peak traffic hours, then congestion would be reduced because there are fewer cars on the road. An answer of "Yes" to the second question logically implies an answer of "Yes" to the first question. (One may be able to explain this by arguing that, "Well, sure congestion charging would be effective for influencing my driving behavior, but I don't see how it would affect enough other people to make it worthwhile. I'm special." Sort of how most people rate themselves as above-average drivers.) What I believe happened was that people reacted by saying to themselves, "I am opposed to congestion charging," and concluding, "Therefore, I must do what I can to prevent it from happening." Proclaiming on surveys that it would never work is one way of accomplishing this. When I shared my brilliant theories with some of my colleagues, one of them, a program manager on the Office team, added his own observation (which I have edited slightly):
I'm sure there are other ways people subconsciously lie on surveys and focus groups, but those are the ones that came to mind. [Insignificant typos fixed, October 13.] |
Comments (81)
Comments are closed. |
I’ve seen that too. I used to lie at the Pepsi Taste Challenge when I was a kid, because I thought that choosing Pepsi meant I got a prize.
Smart surveys will repeat the same questions over and over in different wording, while smart focus groups should observe what is being done, not what the user says he/she is doing.
Raymond, that isn’t lying (for the most part). Aside from telling someone what they’d like to hear, all that stuff is simply people being inconsistent.
If you buy a PC because it’s going to organize your finances, but give up because it’s so much work, that’s honest. Ever get a gym membership and then not go?
If you claim that toll roads won’t reduce congestion, but will influence your driving, that’s not lying. That’s an example of a failure to generalize, or an assumption that it’ll be different people on the toll road. I avoid I66 (DC metro) due to the crack-induced HOV restrictions, but it sure doesn’t help the congestion.
People are not rational. You of all people should know that. They hold multiple conflicting views and see nothing wrong with it.
Raymond,
This post says something I have been trying to say to others for along time. Thank you for wording it in a far better way than I ever could.
One day, Raymond, one day, I will be as cool as you.
James
Hi! I have been lurking around for a while. First post here, though.
Just on the Office team program manager’s first observation, I just recently experienced this while installing a SATA PCI card. What a pain! It seems the driver refuses to work if you add it as a new hardware device, as opposed to for a device that windows knows about but doesn’t have a driver for. Spent an hour trying to figure out why I was getting a "Code 10" error. Somewhat annoying.
Anyway and interesting point you raise in this article. I’m afraid I don’t remember justifying why I got a computer. I’m sure it had to do with learning to programme – which I did :)
re: the Office usability manager’s comment — the switch in the user’s perception may be a shift in gestalt, a well-known psychological phenomenon. Good trainers do this all the time: first create the mind-set from which everything makes sense, then (and only then) do the instruction. Nothing "makes sense" outside of a context, or way of looking at things.
I don’t think that’s a very strong basis for determining what people actually use their computers for. Anyone with a web browser can partake in online banking – in the case of my bank, the bank website has better analysis tools than MS Money or Quicken does, not to mention that its free, platform independant and more trusted – the only parties involved in my financial life are myself and my bank.
It doesn’t matter if people "lie". Most market research is based on how people historically answer. As long as they lie the same way as other people it makes no difference usually. Quant research (focus groups) is usually tested with Qual research (surveys).
Protecting the guilty. A old former rock star wanted to do a comeback tour. A survey measured how much people would pay (most wanted to pay $2 or $5) even though no concert is anywhere near that cheap (a really cheap concert would be $35). But they liked the answers and the tour went ahead.
As he keeps coming on the TV I presume his comback is sucessful.
A twist on this is something that never ceases to amaze me. There are many things in the products I work on that are not intuitive to most savvy computer users (who haven’t been trained on our products). When I try to explain to our product managers and even other developers that we should attempt to make their use intuitive, the usual response is: ‘no, we just need to train the user better!’.
Granted, our target audience is significantly more narrow than something like Office, but still, it blows me away that this mindset is still affecting usability decisions.
The premise behind the intial point — that people are not truthful, consciously or otherwise, when asked a survey question — is probably flawed in itself, in that it presumes people are rational about things like computer purchases, or any purchase, for that matter. The science of retail and the art of advertising proves every day that there is a wide, wide gap between the rational brain and the wallet.
This is even more true in politics. In political surveys, when people are asked their opinion about some issue or other. As Louis Menand recently wrote in The New Yorker:
"When pollsters ask people for their opinion about an issue, people generally feel obliged to have one. Their answer is duly recorded, and it becomes a datum in a report on ‘public opinion.’ But, after analyzing the results of surveys conducted over time, in which people tended to give different and randomly inconsistent answers to the same questions, [a researcher] concluded that ‘very substantial portions of the public’ hold opinions that are essentially meaningless — off-the-top-of-the-head responses to questions they have never thought about, derived from no underlying set of principles."
I don’t see any logical inconsistencies in the answers to the traffic questions.
I would probably answer the same, and this is probably the reality.
You would expect that sometimes, when you can, you would change your ways a little bit, and so would other people, but the changes would not be dramatic enough to have any effect.
Raimond, Have you read the "Wheels" by Arthur Hailey? An easy read, but quite a lot of interesting info from the undercovers of car industry, including something like this (not an exact quote)
…So they send a dame on the streets to ask people about what car they wanted to buy. If a guy were with a girl, he would try to impress her … If a guy was by himself he would try to make himself feel good and would tell things about fuel economy and safety … But in the end of the day when a customer walks onto a dealer lot what he buys is the car’s sexy looks…
;-))
Steve, I agree 110% on the new MS natural keyboard; I recently bought 4 MS Natural Keyboard Pro keyboards on ebay just to have some decent keyboards for backup.
I disagree with the contradiction. One might simply think "I won’t be driving if the tolls are higher but I suspect most people will drive anyway". With this in mind, it makes sense to answer No to the 1st question and Yes to the 2nd one.
Doh, serves me right for not reading the article entirely before commenting. However, I do disagree that this kind of thinking means that the person considers themselves "better" than the other pollees.
I heard the rearrangement of the arrow keys and insert/delete/pg up/pg dn/ cluster was so that MS could avoid patent royalties on that keyboard arrangement that someone else held.
Is there any truth to that?
The upside down T is far superior to the "new" layout.
Andy: "The upside down T is far superior to the "new" layout."
Nah, there is a better solution, where left and right buttons are double hight, so you can press left/up or left/down or right/up or right/down having the same distance between the keys in each pair. I had a keyboard like this 15 years ago, so great for games ;)
By the way, about the usability: somewhere on the way from Word 6.0 Office team screwed up font/paragraph formatting behavior when I delete a formatted paragraph. Previously it worked great, the paragraph mark characher was like a formatting holder, so its deletion would remove formatting. Now it works kinda different, I backspace to a paragraph, and another paragraph after the one I am working now changes formatting. I cannot even describe it ;) Anyhow, Word is horrible now, with strange selection behavior and weird style application.
Merle, that’s my story about the floppy on a Mac.
http://groups.google.com.au/groups?q=%22david+candy%22+format+floppy+mac&hl=en&lr=&safe=off&selm=Ou6cazKLEHA.1312%40TK2MSFTNGP12.phx.gbl&rnum=1
and
http://groups.google.com.au/groups?q=%22david+candy%22+format+floppy+mac&hl=en&lr=&safe=off&selm=O9qAa%23N%23BHA.372%40tkmsftngp04&rnum=2
I can license my story to you for a peppercorn.
Steve,
I regard MS having STOLEN money from me for a non PC Compatable keyboard. Compatable keyboards cost from $10. A sub standard and faulty designed keyboard costs $200. What a fool I was for thinking that good mouse design (in the mechanisms) would translate into a good keyboard. They should come with warning saying they are not suitable for computer use.
I use CUA keystrokes. I’ve asked raymond to talk about it but he always ignores my questions. AFAIS (S is for Suspect) Win 3.1 was CUA complient. This means Shift + Insert, Ctrl + Insert, and Shift + Delete are the keys I use for Paste, Copy, & Cut. But on these faulty fraudelent keyboards Insert shares a key with Prt Screen, plus it (the Fn key) defaults to off so when I try to paste I overwrite the clipboard with a screenshot.
Where was the usuability testing on this.
Starting Help & Support puts the cursor into the search box for typing. Opening google the caret is blinking in the textbox but it doesn’t have focus. So typing is just thrown away. MS decided to remove beep if keys aren’t going anywhere from XP. So 100 times a day I type things then has windows throws it away I have to click the textbox, click away, but not too far (else focused is lost), to get rid of the autocomplete, then retype (very often very long obscure words – I would give some examples from today but none of the searches I did today are in History – only relly old ones and ones done in last 20 minutes – one was Kantian).
This is not usuability.
"I’ve asked raymond to talk about [CUA] but he always ignores my questions."
There’s not much to say about CUA. That’s why I haven’t written about it.
I have no connection with the keyboard team; if you don’t like their keyboards complaining to me isn’t going to accomplish much (aside from giving you a place to vent). Personally I’m not a fan of any of the "advancements" in keyboard or mouse design since 1996. (I have three mice in my office. None has a wheel.)
I too object to the term "lie" used in this sense; to lie is to intentionally mislead someone, but to utter an untruth you can attribute to many things (ignorance being the common one).
It was the relationship between MS’s current and past UI design and CUA. Is there one? Is windows supposed to be CUA complient, was 3.1 CUA complient. Was there a design decision to be non CUA complient assuming that 3.1 was.
I thought that would fit in with the history of windows. And OS/2 UI can also get a look in as well.
I’ve searched the web for the standard but cannot find it (can find lots written about it but not the actual standard), ditto at IBM’s web site.
Anyway change the subject.
People say they buy computers for recipies and finance because that’s all newspapoers used to say why people should buy them. So if one hasn’t already bought one (and most companies were NOT computerised so most people had no direct experience with any type of computing) the propaganda was the person’s only source of information on what computers could be used for.
I didn’t use (or see) a business computer till the early 80s (except I used to break into a bank’s computer room on weekends and play with their mainframes in the 70s) but I remember clearly the scorn I had for those two suggested uses. Even 5 years ago I remember pondering how difficult it would be to cook in a kitchen while running back and forth to the computer room (not that I’ve ever read a recipie from any media). So the same old recipie/finance (and finance doesn’t include online banking – that new) was still being peddled 5 years ago. I no longer read computer magazines but they probably still say it.
PS Back to keyboards. It’s wireless and doesn’t work if the cat’s inside unless I hold it in front of the monitor. Some MS staff member offered to buy me a new keyboard from the company store, some mvp had a theory of antenna orientation (I’ve designed one commercially sucessfull antenna – I had already tried that).The neighbours cat only comes in with me in winter so I’ve got 6 months before he’ll be in again. In winter I often use the onscreen keyboard rather than a real one. Yet the mouse works ok. Could probably take it 30 miles away and it would work (3 metres anyway).
David:
Regarding the CUA standard, you can buy it from:
http://www.digitaldinos.com/Pages/ForSale/PCCompatible/docPCCompatible.htm
IBM Doc #: SC26-4583-0
(I have no idea if this place is reputable or not…)
As it was to answer some newsgroup question about Alt + Backspace (same as Ctrl + Z in office standard ie undo) I not going to pay money to answer someone’s else problem. But it made me realise my assumptions that Windows is CUA complient is probably wrong. But yet is is at least partially complient.
As I said in passing on Larry’s blog today
"… first it was the move to Document Centric computing and I believe we are now in Task Centred computing (Media Player, My Pics, Help & Support, etc)."
But I only think the above is true. It mightn’t be because MS doesn’t really list it’s design philosophy. The UI group probably believes that Windows standards are user discoverable or something like that. If one wants to know how many Dialog Units (size of)to make a dialog then one can find the info easy (even though it’s useless info as displays are pixel based and my programming language [VB] is twips based and the default sizes are wrong and are still wrong in version 6), but the underlying concepts seem to be a mystery.
I have a copy of the CUA, it is on my bookself of obsolete UI guidelines, between OpenLook and PenPoint :)
If you read CUA, it is clearly IBM’s attempt to make OS/2 apps feel like mainframe terminal sessions; not that relevant to today’s problems. But the key bindings are and should be sacred.
David: ah, but it took you "hours". It was only about 20mins for me. ;-) I don’t have web confirmation, though, as this was back from the days when gopher and veronica were cool and new technologies…
I never could remember the CUA cut/copy/paste keys, and hated moving my hands over to those keys — especially painful if you switch between keyboard layouts a lot. I still remember discovering Ctrl-XCV for those, and thinking "Ah! Now I’ll be able to use those features!". (previously I would always use the menus. With the keyboard, of course, but it was slower by far)
It’s known also in the field of statistics that people have a desire to be consistent in their answers to a survey (see the first quote at http://www.yes-minister.com/ypmseas1a.htm, under the second episode). See also http://anthonywells.typepad.com/anthony_wells/2004/09/anthony_wells_g.html – a guide to political opinion polling. He says:
"Another issue in framing questions is where in a survey a voting intention question appears. In their main public polls the pollsters will always ask the voting intention question before any questions about attitudes to issues or personalities, less the earlier questions affect people’s stating voting intention."
I could never remember the CUA keypresses – Windows 3.1 and later borrowed the Mac’s shortcut keys which to me are far more memorable. Ctrl+X = Cut, Ctrl+C = Copy, Ctrl+V = Paste, Ctrl+Z = Undo. I remember the first three by noticing that the X looks a bit like a pair of scissors, V is a sort of insertion point, and C, er, is the first letter of Copy. Did Apple do usability studies on this? I don’t know.
Here I thought that PenPoint was the Tablet PC.
I belive Ctrl + C et al was first introduced in Word (probably ver 6). FYI Alt Backspace (Undo) works in office but not in the shell. But CUA Cut/Copy/Paste work in the shell, just not undo.
It all came about from some question about Edit.com, I thought that the CUA keystroke shouild do whatever the person wanted but then couldn’t tell if Edit was CUA complient, went searching to test it against the standard, suddenly my assumptions started looking wrong, tested in the Shell and tried to compare to my memory of 3.1 (9 years ago).
I still can’t believe it’s not on the IBM web site.
In omnibus surveys (that’s where you buy a question in a larger survey – it’s cheap) political questions are asked last for cost reasons (some people will talk about their toilet paper use but not their voting). After being asked an eclectic range of questions they are more likely to complete the survey if asked political questions last.
I’m actually worried the Ctrl + Insert keys will disappear in a future version. I’ve been using Ctrl + Insert for just over 10 years. The transition from IBM mainframes (not CUA) with 24 PF keys and WordPerfect style F keys was hard enough (and just had me wasting months customising programs to make them more mainframe/WP like).
Still I’d like a cognitive map of just what the design is TODAY (I know what it was at various times in the past).
Beat to the punch again. This made me think of the New Yorker piece Mike mentioned. Very interesting, but at least somewhat depressing read: http://www.newyorker.com/critics/atlarge/?040830crat_atlarge
Raymond, you’re lying when you say they lie. Or were you mistaken?
I was pretty young when I got (was bought) my first computer. I recall arguing that I could program on it (I did) and use it to assist with homework (I wrote programs as "special" projects, so it was kind of true).
But in the back of my mind I *knew* I wanted the computer so I could play games on it (and write games when the cartridge offerings got to be so pathetic).
But hey, I was a kid then…
—
I am not quite convinced that your third form of lying (contradictory beliefs) really is lying at all. Most people cannot reason their way out of an open cardboard box. There is no particular reason to assume that their belief systems are internally logically consistent.
I have encountered several cases where my own beliefs seem to logically contradict. Some of this is that few statements one encounters are clear cut; most have biases that affect several different things.
Also, the person’s perception of the world might be such that those two answers are still consistent. Consider: the person believes, in their mind, that traffic is fixed. Every day they drive at the same time, and so they never see a traffic-free road. If they assume that congesion charging is not going to stop anyone from driving, then it will have no affect on their *perception* of traffic — and there will be no point for them to alter their routine since traffic is a constant.
—
Sounds like the Office usability participants are wusses. I rant and rave at my software almost every day. ;-) Of course, it’s hard to differentiate the "intuitive" and "learned" components of usability
I still remember the first time I had to format a floppy on a Mac. I had never used one before, but was in a Mac lab, and just wanted to be able to take a file from one machine to another. I hunted and searched for any sort of formatting utility, but could not find one. For twenty *minutes* I hunted. Finally, in frustration, I shoved the floppy into the machine, not really expecting anything to occur. Voila! It popped up a format dialog.
While this is probably "intuitive" to Mac users, it went against everything I had learned from years of non-Mac systems.
I agree 100% with user studies, and want to point to the latest MS-natural keyboard as an example of where it has gone wrong. they keyboard has changed some keys around (the insert/home/end) pad, presumably for "usability improvements". But the end result is that the change is actually negative for anyone used to the old keyboard as suddenly whey they hit end, they get delete instead.
Same for VS.net and its keyboard rebindings. Every version of Visual studio changes keystrokes *for no apparent reason*. There may be usability and consistency benefits for new starters, but for any experienced users, its a complete nightmare.
so, next time you meet the usability folk, as them about usability for existing users,
example 1 isn’s a lie at all.
"People claim that they want a computer to do their personal finances, to organize their recipes, to mail-merge their Christmas card labels."
They actually WANT to do these things, but they don’t end up doing them. It isn’t a lie in any sense of the word. You asked what they wanted, and they answered with what they wanted at the time of asking. If they don’t do it, it’s worth investigating why they don’t. Either because it was harder than they thought, or they found better uses of their time. EG, I WANT to finish my scholarship applications tomorrow, but I know that I won’t because my referees havn’t gotten back to me yet.
As a self evident argument maybe some consideration should be given to the use of the word "lie". I’m a big fan of usability and agree that a program that is difficult to use is just as bad as a buggy one (note that during the usability tests described above the difficult to use software was described as "broken"). The challenge of usability is that it describes computer-human interaction and therefore is very subjective. This is not how programmers think. Deterministic or at least objectively consistent behavior is the programmer’s mantra (side note: much appreciation recognized for Mr. Chen and others helping to make Windows perform in a consistent way release after release). So, in this objective meets subjective universe perhaps an end-user, when asked what would make he/she happy with the software, who then is unhappy when the software is changed to provide what he/she wants is, objectively, "lying". The question then follows "of what use is asking the end-user what they want"? This is alluded by the Office program manager who simply states that usability must be based on user interaction and not words.
As a valuable exercise I wonder how such software surveys or software usability tests could be worded to elicit actual useful spoken information?
It seems everybody went nuts on the word "lie". I was simply "punching" up the topic to make it more visceral. If I had said, "What people say and what people do do not always match", it wouldn’t be as juicy.
Agreed. I’m just suggesting that you used the word "lie" for the same reason that end users contradict themselves. We’re all looking at the problem from our own perspective.
It is truthful to say lie. Isn’t the definition of a ‘lie’ something that is contradictory of the truth – irrespective of whether it is intentional or not?
David Candy wrote:
> But I only think the above is true. It
> mightn’t be because MS doesn’t really list
> it’s design philosophy. The UI group
> probably believes that Windows standards are
> user discoverable or something like that. If
> one wants to know how many Dialog Units
> (size of)to make a dialog then one can find
> […] but the underlying concepts seem to be
> a mystery.
Erm… what are you having trouble finding? The book "Windows User Experience Guidelines" is where you should start, and there are the WOSA books if you want to dig back further. If you want to go really far back for fundamentals, check out the Apple Human User Interface guidelines (because you’re unlikely to come across a Xerox Star manual, but you’ve got a good chance of finding a used Apple one).
David Candy wrote:
> I belive Ctrl + C et al was first introduced
> in Word (probably ver 6). FYI Alt Backspace
> (Undo) works in office but not in the shell.
> But CUA Cut/Copy/Paste work in the shell,
> just not undo.
The CTRL + C/X/V shortcuts were in Windows 3.1 which came out in 1992, and presumably predate it.
Word 6 was 1993.
Apple had their Apple+C/X/Z/V in System 6 which was released in 1988. That was the first Mac os I used – so it may have been in earlier versions too.
I read all of Tog’s books (Apple’s UI designer – though he only does usuability on things that are US specific these days*). I understand the underlying philosophy of the Mac well despite not using it. It’s my base for understanding Windows. But that’s my point – the philosophy of UI is only propounded by the Mac – and Windows broke from that 20 years ago. But it is the only conceptual model available.
To understand windows one must infer everything. The UI Guidelines in the MSDN (and I have various versions of it from 3.1 on – though only the latest installed).
Take tree views.
On one hand Explorer (as in two pane mode) has had it’s Start Menu shortcut progressively put deeper and deeper into the Start Menu structure. What does this mean, especially as the Macs eschew tree views altogether. I surmise from this that MS has decided that treeviews are "bad", and so I haven’t used a two pane view for years. BUT we have had introduced the MMC thing that makes extensive use of treeviews.
So are treeviews generally good or generally bad. If the first why is the explorer shortcut being hidden from users, if the second how did MMC escape into the world. I have to infer these concepts from the Apple UI designer and MS behaviour. Note BeOS also doesn’t have a treeview and I can’t recall for OS/2 (but sure it did have treeviews).
You can go to Tog’s website at http://www.asktog.com/
Back to the subject of "Lies". I understood the meaning Raymond intended. Do you want passion or droning monotones.
As Tog repeatedly points out in his books, UI testing shows mouse is quicker than keyboard. Users maintain the opposite but stopwatches show mousing is quicker (at least with a mac style top menubar).
There’s a physicists called Fenyman (or something – he was on the shuttle challenger inquiry). In one of his books he talks about being at university and become facinated with the human (dis)ability to tell elasped time. What he discovered was people count in different way (visually, audiolly, kinesthicly – pardon my spelling). If one prefers to count visually (see items or numbers in one’s mind), doing other visual tasks will slow down the sense of elasped time (in our terms of reference the brain is a specific purposes multiprocessor time sliced OS that counts time in a timing loop – two processes in that part of the brain means each get 1/2 the available time so twice the time goes by).
Apple found that users go blank, aren’t aware of the passage of time when they try to recall a keystroke (for about 2 secs).
Yet MS has always been a keyboard company (even before accessability became the reason) yet they don’t justify these design decisions.
So were these users in Apple’s labs lying when they said that using a keyboard is quicker than a mouse? As many that read here also make the same assertions (but note windows has windows menubars not screen menubars – Apple’s research may be different if run on Windows platform) are they lying or telling the truth. They think they are telling the truth. They ARE telling THEIR truth (post modernism).
MMC’s target audience is soooo different from Explorer’s.
"Never attribute to malice that which can be adequately explained by stupidity."
http://www.jargon.net/jargonfile/h/HanlonsRazor.html
Another example is MDI/SDI. Explorer is SDI and Fileman/Progman were MDI applications. Word was MDI, became SDI, then became MDI or SDI (but like everything about office it simulates MDI/SDI).
I infer from this that MDI is "bad". Users are resistant to moving away from an MDI style (and I think this shows they are application centric).
Tabbed browsing is MDI by another name, and is a hotly requested feature. Where does MS stand on the MDI vs SDI debate. Should I write MDI or SDI application.
This is what I mean about the underlying concepts need to be articulated and the reason for these concepts so I can make decisions on specific circumstances.
In the military they have the concept of doctrine. One thing it tells you to keep a reserve. The military force uses doctrine so all soldiers can predict the behaviour of their mates nearby so everyone understands the drill. But in particular circumstances it might be appropiate to not have a reserve (capturing a single downed pilot with no enemy nearby). They can make this decision knowing from their doctrine that this is usually a very bad idea but might be required.
What I’m advocating is MS to have UI doctrine. How things should work unless their is a particular reason to override it.
Doctrine is usually philosophical rather than a list of rules or ckecklists.
David – I think you’re looking at it from the wrong point of view here. Don’t write MDI/SDI because that’s where MS ‘stands’ on it – pick the interface style that makes the most sense from the point of view of your application.
Why DID MSFT decide to rearrange the Insert/Home/Etc. cluster? Why can’t I have F-Lock on by default? Why are the F-Keys now 4 groups of 3 instead of 3 groups of 4? There must be a reason why they did it, but for the life of me I can’t work it out…
WRT mouse-faster-than-keyboard, once you’ve learned the shortcuts and don’t have to spend 2 seconds remembering them they’re FAR quicker. Ctrl-X is <1 second, but "remove hand from keyboard to mouse, locate moue pointer on screen, move to menu, click, click" takes me a good 3 or four seconds. (Admittedly the Mac "menubar at the top" dealy removes the "find the mouse pointer" part; it’s a shame the draggable menus in Office won’t dock to screen edges.)
And David — MDI all the way!
Talk of surveys reminds me of an episode of a British TV series "Yes Prime Minister". Sir Humphrey Appleby is explaining Bernard Woolley how to ‘rig’ an opinion poll on the reintroduction of national service
(taken from the excellent http://www.yes-minister.com look there for more quotes from the series)
Sir Humphrey: "You know what happens: nice young lady comes up to you. Obviously you want to create a good impression, you don’t want to look a fool, do you? So she starts asking you some questions: Mr. Woolley, are you worried about the number of young people without jobs?"
Bernard Woolley: "Yes"
Sir Humphrey: "Are you worried about the rise in crime among teenagers?"
Bernard Woolley: "Yes"
Sir Humphrey: "Do you think there is a lack of discipline in our Comprehensive schools?"
Bernard Woolley: "Yes"
Sir Humphrey: "Do you think young people welcome some authority and leadership in their lives?"
Bernard Woolley: "Yes"
Sir Humphrey: "Do you think they respond to a challenge?"
Bernard Woolley: "Yes"
Sir Humphrey: "Would you be in favour of reintroducing National Service?"
Bernard Woolley: "Oh…well, I suppose I might be."
Sir Humphrey: "Yes or no?"
Bernard Woolley: "Yes"
Sir Humphrey: "Of course you would, Bernard. After all you told you can’t say no to that. So they don’t mention the first five questions and they publish the last one."
Bernard Woolley: "Is that really what they do?"
Sir Humphrey: "Well, not the reputable ones no, but there aren’t many of those. So alternatively the young lady can get the opposite result."
Bernard Woolley: "How?"
Sir Humphrey: "Mr. Woolley, are you worried about the danger of war?"
Bernard Woolley: "Yes"
Sir Humphrey: "Are you worried about the growth of armaments?"
Bernard Woolley: "Yes"
Sir Humphrey: "Do you think there is a danger in giving young people guns and teaching them how to kill?"
Bernard Woolley: "Yes"
Sir Humphrey: "Do you think it is wrong to force people to take up arms against their will?"
Bernard Woolley: "Yes"
Sir Humphrey: "Would you oppose the reintroduction of National Service?"
Bernard Woolley: "Yes"
Sir Humphrey: "There you are, you see Bernard. The perfect balanced sample."
Ah, but surely it should be up to highly trained professionals to interpret what people say and deduce the wish from the want.
Afterall, it’s a hidden skill in most programmers to receive a spec and to realise what is essential and what should be in version 2.
My experience with personal finance software is that it is horrible. I’m back to Excel sheet and way happier. Some reasons in money 2004:
1. On setup you can do max 9 accounts. – We have way more than that.
2. It insist that transactions between two accounts are done in the same date on both – which is simply not true. I often ahve my cc payment show first at the cc and be debited a day ot two later. I write checks on certain dates but they get chased often weeks later and the bank shows them on that date. I can get tons fo more exmaples.
3. We have two discover cards but Money can’t handle two logins for one Merchant.
4. It claims to support ING Direct yet it tells you both "Your apssword is 0 characters" and "Your password can not be empty". In reality ING Direct uses an accoutn number, a password(PIN) AND a third revolving question like SSN, ZIP, etc. It has always been like that since day one.
5. The stupid browser interface that doesn’t look any normal application.
6. Ear piercing help that I can listen to perfectly if I plug my headphones and leave them 2-3 feet away form me. No option to control this.
…
these are on top of my head and I haven’t touched it in many months.
Also many banks give Money or Quicken for free so people don’t buyt it. It also comes often bundled in new computers.
More important is that every program works the same way. Explorer is SDI so every application also needs to be SDI. I love running the CE emulators – despite having absolutely nothing to do in them, the joy of using something, to no purpose, that seems designed is overwhelming. Two years ago I installed 3.1 somewhere – the joy was almost painful in using what 32 bit OS don’t have – design.
We can go on. Try selecting the following source displayed in IE (as a text file or html)
<p><input parameters=something></p>
without taking the paragraph marks. Now this feature was copied from Office, but the office team worked out that you needed to override potentially (context specific) stupid behaviour, so it can be turned off or overridden for a particular select operation. This is not possible in IE or even this text box. Why? Why copy office in a half hearted fashion. Not to mention how to select without a mouse.
Plus spell checking doesn’;t work here either. Now of course I can write some hack that starts word as an automation server, pass it to word to spell check and write it back – but why should I need to. Why doesn’t it work properly. It’s been years.
Perhaps Raymond can show this to Bill G as it seems noone at MS cares (or has any power to care) about the lack of coordination. Bill might. These are now 9 year old problems.
T thought the purpose of having API services like menus, text boxes, checkboxes, toolbars, status bars, titlebars, common dialogs, a dialog engine, common controls, string tables was to provide a common user interface (just like CUA). But it has failed.
I don’t use Linux or Unix because
a/ I’m a snob and looked down on toy OSs like unix in my mainframe days, and
b/ If Linux can’t decide if Gnome or KDE is it’s shell then obviously it not a suitable product. The reason – there is no design (if I had to use it I like KDE as they copy MS).
I want to be convinced that MS isn’t incompentent. Since reading your blog I’ve come to the conclusion that these aren’t smalll overlooked things but evidence of systemic problems in MS. So I can’t expect that it will be fixed.
Computer + Operating System + Application should = an integrated package. That’s why doctrine both within and outside MS is important. My mum payed for adobe products, I remove them – why – because Adobe doesn’t want to do things the windows proper way and as they have convinced weak minded idiots that PDFs are a smart way to present online data (rather than it’s only legitimite use – printing). I read up to several hundred PDFs a month – that’s thousands of pages – scrolling sideways, down a column, then up to top up the page, then down the secobd column, etc.
Users are microsoft’s customers not developers. There needs to be normative standards (ie Expert defined standards) for the users benefit. Developers generally want to write good products but they need normative standards.
For the second type of lying, that’s why it’s important to ask:
"What would product <X> need to do to get you to switch from product <Y>?" rather than "Why is product <Y> better?"
If you ask the first one, you may find out that there are other reasons why people prefer <Y>, or that they can’t think of things that would make them switch.
"(…) People claim that they want a computer to do their personal finances, to organize their recipes, to mail-merge their Christmas card labels. (…)"
Actually using computers to organize the recipes has a surprisingly long history:
http://www.old-computers.com/museum/computer.asp?st=1&c=927
;-)
David, why bother with treeview in Explorer? Why use Explorer at all? It is a totally unfriendly unusable tool. From the DOS times Norton Commander and its clones did the job much better. Windows Commander is the right tool for the job.
I still cannot understand why MS have not introduced normal two-pane source-target paradigm for its file manager. They were close with Windows 3.x File Manager but decided to ditch it in favor of "cool" MDI Explorer. Well, Explorer sucks, its usability is below freezing point. MS thought that users would use cut-n-paste for moving files around in the same manner they move text around the Office apps. What about comparing or synchronizing directories? What about detailed file info including size in bytes? What about attributes handling? What about not to waste screen estate for stupid pictures, but to show only the list of files? And do not point me back to the command line, it sucks too.
I respect MS for trying to provide keyboard interface for every feature which is accessible using mouse. But sometimes providing the interface is not enough, sometimes it just can be made better and simpler. Being known for copying good stuff, it is amazing why MS did not copy Norton/Total/Midnight Commander. Well, at least it gives Christian Ghisler some revenue source.
Your average user doesn’t care that a file is 185,141 bytes in size (as opposed to 185,140 bytes). Let’s face it – nobody who reads this blog is a typical user.
David Candy suggests:
"As Tog repeatedly points out in his books, UI testing shows mouse is quicker than keyboard. Users maintain the opposite but stopwatches show mousing is quicker (at least with a mac style top menubar)."
Jef Raskin ("The Humane Interface") contradicts that in his book (repeatedly).
It depends on the activity, obviously. Using a mouse to select every other file in an Explorer pane to copy all of them to three separate folders is nigh impossible with a keyboard. Bitmap editing is another place where mice shine.
Things like cut/copy/paste, closing windows, switching between windows… slow. Especially if you’re doing any tolerable amount of data entry.
If you only type two words a minute, the cost of moving your hand between keyboard and mouse is negligible. But for quick typists it’s interminably painful. I learn keyboard shortcuts as soon as I can and use them for just about everything. It’s why I use Opera to surf around. Yes, a lot of things are "broken" (sites that are IE-specific). But I don’t need to touch the mouse.
Switching back and forth is the main latency. It can take a second or two to move hand, find cursor, move it, and click. I can activate three or four keyboard shortcuts in that time.
If you’re a one-handed typist who leaves a hand on the mouse at all times… perhaps it’s better for you. But it is not for me.
[quote]It depends on the activity, obviously. Using a mouse to select every other file in an Explorer pane to copy all of them to three separate folders is nigh impossible with a keyboard.[/quote]
Even in crippled Explorer it can be done with just few keypresses: Ctrl+A, Ctrl+C, Ctrl+V, Ctrl+V, Ctrl+V. If you do not need all files, hold Ctrl while selecting files with spacebar. Ergonomics for three-handed mutants, heh. You need to handle three buttons keeping one constantly depressed, while in Windows Commander the same is achieved with only two. If you use classic Norton Commander keymap, files are selected using Ins key. Which is great with numpad arrow keys. Files can be quickly selected with only one hand, and no keys should be held in the process.
I don’t know what usabilily studies MS conducted when they were inventing Explorer, but apparently they were comparing their own File Manager with Explorer.
Raymond: 2/3 of those whom I shown Total Commander stuck with it. Now they cannot understand how they lived with Explorer. But right, they are not your average users. Does it mean that Windows should be dumbed down? Why not to provide "Novice" and "Expert" modes? XP does something about that with its Task-oriented and App-oriented Control Panel. But this is just a tiny step. Are couple of Registry settings and default screen theme the only difference between Professional and Home edition? MS could do better than that.
http://weblogs.asp.net/oldnewthing/archive/2003/07/28/54583.aspx
2. Home and Pro are intentionally similar so that people can upgrade from Home to Pro and not have to relearn everything.
David Candy:
re: the explorer/treeview/explorer being hidden comments…
Erm… you do know that you don’t NEED the Explorer shortcut any more, don’t you?
Either hit Win+E – gives you the "explorer" view, or from any open folder, click on the "Folders" toolbar button.
All the Explorer.exe shortcut gives you these days is a shell window with the Folders button activated. That’s why it disappeared – it’s much easier to just go to My Computer and make sure that the Folders button is clicked in if you want that.
I have an hta that lists nearly all shell shortcuts (even wierd ones like Winkey + B [sets focus to 1st icon in tray – I won’t get in trouble for calling it a tray I hope] or Right control + Scroll Lock [Causes a memory dump – need a reg tweak to enable]). It grows weekly (I’m about to add the mousekey keys).
You can see a version in google
http://groups.google.com.au/groups?q=%22david+candy%22+key+shortcut+shell&hl=en&lr=lang_en|lang_es&safe=off&scoring=d&selm=%23hjpnNjqEHA.3464%40tk2msftngp13.phx.gbl&rnum=4
I had been thinking of a way to get this on Larry’s or Raymond’s blog as I suspect really wierd keystrokes are builtin to Windows and the person who did it might read one of their blogs so I can add more to my list. If you want it as an hta mail me at david@mvps.org
Dosshell works on XP if you don’t mind short file names only. My Dos file manager was Dirmagic by PCMag (you had to subscribe to get it), a enhanced version of one of their free utilities. It is function key based. I have Norton’s too and XTree but don’t like either (and had PCtools in the past).
I refused to use long file names (LFN are for girly boys) in 95 so I wrote a program that crashed if it came up to a long file name. I started doing things windows way then, even though it is a secret way.
I treat Explorer (folder view) as a namespace browser. I have no problems using it. It’s not a file manager and if one conceptualise it as one then one will always be disappointed. I find explorer very powerful.
Some MS development team sent a commandline question for usuability testing a couple of months ago. I asked "how could you test this" and got an answer that showed they had no idea. My suspious mind thought "it’s an ISO9000 thing, they have to tick a box". I bitch there too about mixing Dos, Unix, CMD, and the NT5.x (statefull like nslookup or multipurpose commands like netsh) conventions.
If MS cares about my mum (a typical user) why not me?
I wonder if there is a Tog equiv at MS. And if so who is s/he.
Tog spent a lot of pages arguing about this 2 secs. I would need to know more and watch my own experiments to form a firm opinion. But I know the top of screen traps the mouse and Tog thought that very important. I know fatitgue makes it really hard to be accurate with a mouse.
I only use the Windows navigation keys so I would go Alt + F, S rather than Ctrl + S. I don’t use the winkey at all (Ctrl + Escape), and I use Shift + F10 instead of the context key. I loathe wordstar conventions and Ctrl anything makes me remember Ctrl + K, M. While I’ve been weaned, very painfully, from Function Keys. I read a lot and my keyboard is out of reach normally so I often use the onscreen keyboard. And I work in any of three modes, keyboard only, keyboard and mouse, or mouse only depending on how much typing I’m doing (I’m K&M at the moment). I’m about to read and will be mouse only and on screen keyboard if needed (basically typing stuff in Encarta or google so see what a word means in the context I’m reading).
May back when I was an undergraduate student, I took an introductory psychology course (highly recommended for anyone who deal with UI issues, BTW)
One of the problems discovered early when getting rigorous on the experimental side is that people will do pretty much *anything* when asked. This makes directly experimenting on any topic difficult.
One experiment went to see how far they had to push subjects before they would say ‘no’. They specially hired offices in disrepuatble parts of town, and ran very dubious sounding experiments. Some favourites were asking people to balance ball bearings, one on top of the other. They would do this for hours, so long as you politely asked them to try ‘just a little longer’ for the experiment. Another I remember was filling out a multiple choice questionnaire full of offensive questions, such as "How many men (apart from your father) has your mother slept with in the last 2 years?" and the lowest answer available would be 3! People still completed the questionnaire, filling in the least objectionable answers.
We have a natural tendency to want to please others, whether it is for acceptance, respect, whatever. We will always lean towards what we think the interviewer wants to hear, and always find nice ways to avoid saying bad things.
Field testing UIs is HARD!!
10/13/2004 12:08 PM Raymond Chen
> Your average user doesn’t care that a file
> is 185,141 bytes in size (as opposed to
> 185,140 bytes).
100% true. Your average user doesn’t want detailed view, and fortunately for them the default is icon view. The only users who care that a file is 185,141 bytes in size (as opposed to 185,140 bytes) are those weird users who use detailed view.
Icon view vs list view, Bush vs Kerry. People just don’t have much of the choice. Those who saw two-panel file managers with clear source-target concept, usually do not come back to clunky Explorer.
You do realize that you can do the two-panel thingie with Explorer? Ctrl+Click two buttons on the taskbar, then right-click and select one of the "Tile" options.
<quote>You do realize that you can do the two-panel thingie with Explorer? Ctrl+Click two buttons on the taskbar, then right-click and select one of the "Tile" options.</quote>
Riiight, know about that trick, someone shown it to me, but I forgot about it. By the way, not the most obvious thing to do, and definetely not the kind of the thing a user can get to by himself.
Anyway, these are two different windows, behaving more like two different applications. There is no "source-target" concept, the best a user can do is to drag files over. There is no speed key combination to copy or move or compare or synchronize files (had you have said Briefcase in one of your previous posts?) simply because source window is totally unaware about target window. both windows cannot be resized together, so they would be, say, equal size and side by side but would occupy only half of the screen. If a user switches to another application, then switching back means switching to only one window out of this pair. Each window allows at least two, sometimes three panes which can be focused, so it is a pain to tab over them in order to navigate directories or to create files. And definetely not the least: why on Earth the stupid picture with something like "Local disk", occupying a whole column worth at least 150 pixels??? Does MS think that large stupid pictures make things easier? Nah, they just clutter the interface and draw the consumers to buy larger monitors. I am sure I can turn this thing off in some .htt file… Is this something an average user is supposed to know?
If you used Total Commander you would know that two-window mode in Explorer does not work nearly as well as two panel concept of commanders. Granted, Explorer allows to create several windows. A feature needed very rarely. It is like building a chair with infinite legs. Everything else is just a case of it. But sometimes having two panels are better than having infinite number of windows.
And how this two-window mode helps, for example, to decrease number of keys needed to select file one by one? Three keys, one of them kept hold all the way, this is just ridiculous. In Total Commander I just use arrows and Ins, arrows and Ins, arrows and Ins… All by a palm of one hand. Easy ;-)
"Does MS think that large stupid pictures make things easier?"
Studies have shown that large stupid pictures do make things easier. You can always turn off the large stupid pictures if you want – siwtch to Details mode, then you just get a small stupid picture.
I concede that Explorer doesn’t handle these advanced operations like "compare two directories". But that’s not something a typical user does. (And it generates lots of false positives, too. Open a document in a word processor, delete a word, reinsert the word, then save. The result will likely be different when treated as a byte-by-byte comparison, but most people would consier the files identical.)
About the selection in Explorer: I understand that this is how multiselect works in standard lists and combo boxes. But should the legacy stop the progress?
"You can always turn off the large stupid pictures if you want – siwtch to Details mode, then you just get a small stupid picture."
I was talking about the whole column on the left, which contains information about the parent object. Like when listing a root directory the columns reads "Local disk (C:)". One can indeed remove this column by customizing the window to "Classic" mode. How many users do that I wonder?
I don’t want to argue if Explorer is what an average user wants. I am pretty happy with my commander-style file manager. And it is small enough so I can put it on every Windows machine I happen to work on.
By the way, back in DOS days Norton Commander was preinstalled on virtually every machine which was sold in Russia. So people considered two blue panels to be part of the OS. Maybe this is how I got stuck to it too ;)
Interesting, I have read an opinion that Norton Commander on the one hand, and Explorer or DOS Shell tree on the other hand, represent different psycological state. Russians as more button-closed, tight-lipped, prefer the default Norton Commander view, which shows the content of only current directory. This is considered claustrophobic by some people, but offers some kind of safe small box where a user can reside. And then he would go to another box. Americans, on the other hand, are opened and free, so they prefer large open tree structure which they could browse with ease. Sounds funny, but I wonder does it hold any water? :)
As per the third "lie," though… What about the inverse? What if you really were a different driver, and so would drive even if you had to pay more, without it having any impact. Then you could answer yes for the first and no for the second and be perfectly correct, assuming most other persons answered the opposite way on the second question… So maybe it’s not really that simple?
Definitely, I can’t tell you how many focus groups I have participated in where I was coached by the RECRUITER on what answers I should give to different types of questions. If you answer the questions "correctly" you get more money from the survey.
It’s a completely broken system, my friends and family always laugh when we leave because no one told the truth.
Who knows, maybe the recruiter was instructed to tell us the answers so that the survey comes out in the favor of the company that paid for it.
However, for the whole inconsistency thing. It’s a well known fact that many people don’t’ know what they want. Just look at the entire female species and all the new wave sensitive indecisive men for example. Wouldn’t want to hurt anyones feelings now by giving an opinion, they might get mad at you or feel bad about themselves!
Forgot to mention, in my opinion the focus groups are a complete waste of time and money. I personally like them because $30 – $50 here and there for answering some questions is great for a college student. However, truthfully the only system I can see working is one where a person puts their money where their mouth is and that’s already available by purchasing products. Maybe companies need to have suggestion methods where you pay a certain amount to make a suggestion. Yeah it would cut down enormously the amount of suggestions received, but you would know that the suggestions you received are truly desired and it wouldn’t necessarily cut down on sales of your product.
There is an old wisdom: Only fools and children are telling the truth. I admit, if company X would ask me, what it takes to switch to their product and I don’t like company X, I would tell some shit, so they have something to do:-)
Interesting post on The Old New Thing: "People lie on surveys and focus groups, often unwittingly." Software engineers, usability reports and marketing groups often interpret what people say too literally. As humans, our words are imperfect approximations of our perceptions…
I never knew you could control-click taskbar buttons. Shame that I discovered it after it already became useless (because there’s no way to do it when your taskbar buttons are grouped…)
People lie on surveys and focus groups, often unwittingly. (This is from Raymond Chen’s blog, "The Old New Thing", which is a fascinating insight into the quirks and foibles of an environment where backwards-compatibility has to be preserved at all costs. Of course, not everyone at Microsoft thinks that way.)…
I suppose I wouldn’t be too far off the mark in suggesting that `Product Q’ is the big Q of Desktop Publishing?
Well, even if I am – I assure you it’s a perfect match. InDesign is superior in basically every way to Quark 4, which the majority of the industry is on. Staff will fight strongly to keep their old Quark versions. In my opinion, it’s simply because they’re afraid of change. I’m sure we’re all very familiar with the attidude.
Of course, they won’t say that – they’ll give a long list of missing features, and as each is shown to be a non issue more issues will be found to further hold up the process. At least if they’d simply say "We don’t want to change" that could be handled up-front and honestly. Grr.
I have responded to many e-surveys in the past, and have noticed a disturbing trend.
When asked for my job title, responding correctly ("I am a worthless peon") usually got me kicked out of the survey with a "you are not what we are looking for" message, while passing myself off as a Senior Manager or higher _never _had that effect.
Likewise answering the "Do you have puchasing authority?" and "May we contact you afterwards?" questions with anything other then "Yes" was a swift ticket out of the survey… and the promised reward.
If that is not an incentive to lie, I don’t know what is…
The number of people out to hurt the survey or focus group people, added to those out avoid hurting them outnumber the objective people that don’t give a rats either way.
The respondent that mentioned asking the same question multiple times (and unexpectedly) is correct, I believe. I worked with a psychologist (doing the analysis of her questionairre) who apparently knew this. The effort was to determine why the turnover rate for eligibility interviewers for welfare services was so high (1/3 AFTER training). She concluded that it was fear of the interviewee. There was no question that asked that directly.
The Old New Thing: Practical Development Throughout the Evolution of Windows
PingBack from http://www.copywritersboard.com/marketing-discussion/7672-disappointed-dan-kennedy-phenomenon-forced-continuity-4.html#post63538
PingBack from http://www.webproworld.com/marketing-strategies-discussion-forum/84761-online-retail-business-decisions-information-intuition-both.html#post439621