As I said, I admire passion, I no longer admire that as the single most important characteristic because since I made that statement in September 2006 I have learned more and expanded my knowledge and personal viewpoints. I try to use new knowledge and new experiences to constantly re-evaluate my position and views as compared to other people who notoriously ignore or misrepresent facts to support antiquated untenable positions or constatnly reshash biased rhetoric without empricial evidence to substantiate claims.

Your absolutely correct about a pattern. You have a pattern of talking about psychological theories and speculating about emotional states rather than staying within the context of software testing and computer science.

I am not an expert in philosophy (which is good because Microsoft didn’t hire me to philosophize; they hired me to test software and they now rely on me to teach software testing to new testers and developers in the company), but I suspect you might be a good philosopher. However, based on your endless curiosity regarding someone’s emotional state at particular times, and the lack of technical accuracy in some of your previous comments it makes it hard for me to trust what you are saying within certain contexts.

Tuesday, March 04, 2008 3:41 PM by I.M.Testy

Research into cognition suggests very strongly that it did, in fact. Look up the work of Antonio Damasio, for example; read Karl Weick on the process of sensemaking. I’m finding it difficult to reconcile your earlier language and your earlier escription of your reactions with your denials here. It makes it hard for me to trust what you’re saying from moment to moment. But maybe there’s a pattern.

>I do not agree with the statement “it is fervent passion for the discipline that I admire most in testers.”

http://blogs.msdn.com/imtesty/archive/2006/09/08/746954.aspx

—Michael B.

Tuesday, March 04, 2008 2:25 AM by Michael A Bolton

WRT your percieved problem, when one right clicks on the image in the post in either Vista or Xp a popup context menu appears. On the popup context menu you can select the Open Link, Open Link in New Tab, or Open Link in New Window menu items and the url will display the complete path of http://blogs.msdn.com/blogfiles/imtesty/WindowsLiveWriter/VistaRant3Cansomeonepleasetesttheboundar_EC04/LV1_2.jpg. You can also select the Copy Shortcut menu item to copy the complete path to the clipboard, or you may select the Properties menu item and then double click the Address:(URL) and copy the complete path to the clipboard or (although the properties General tab doesn’t show the complete path unless you highlight the path and move the mouse down while highlighting; which I agree is not ideal). See, there really are different perspectives and solutions to a given problem. Of course, the more we understand the system the more we can understand the more we can explore those various perspectives and potential solutions.

I do not agree with the statement “it is fervent passion for the discipline that I admire most in testers.” I do admire people who demonstrate passion as expressed through their constant desire to learn and improve their knowledge and their abilities of a profession regardless of that profession. But, no I do not consider ‘fervent passion’ to be the single thing that I admire MOST in a person; but it is an admirable characteristic. There is probably not a single trait or characteristic that I admire most, it is a collection of traits, characteristics, skills and knowledge, and the ability to balance them appropriately within a given situation.

I am not really sure where you are driving with all the ‘emotion’ hyperbole. Humans evoke various emotions in response to various environmental stimuli. Since I am still mostly human (although I have some stainless pins and screws and a titanium plate holding parts of me together, I have not yet been fully transformed into an android) I was surprised by the anomoly in the listview. But, (this is important) I expressed that emotion AFTER I discovered the anomoly. As I said, my emotional state had no bearing on the discovery of this particular anomoly.

I suspect that you are trying to convince me that I rely on emotional responses to test, but I must say that I really don’t rely on my emotions when I am analyzing a problem or designing effective tests to evaluate a given hypothesis. I might have emotional reactions while testing, but often I find that a logical argument with facts is more effective in driving a point as compared to unbridled rants, whining or crying, sniping, or petty complaints (although they do sometimes work…especially crying).

Personally, I rely mostly upon my in-depth knowledge, my cognitive abilities to recognize acceptable and unacceptable states, my abilily to think critically about a problem space and design effective tests within the appropriate and possible contexts. Sometimes, when those tests result in an unexpected outcome I am surprised or disappointed or over-joyed, because I am, afterall, still human!

Monday, March 03, 2008 7:34 PM by I.M.Testy

So maybe the useful information from a tester would be that the graphic isn’t terribly useful information; or that it’s too wide and therefore bumps the interesting stuff off the screen; or that the bitmap doesn’t scale down to permit the file data to remain visible. This is the George Carlin heuristic: “Some say the glass is half full. Others say the glass is half-empty. I say the glass is too big.” The point is that there’s more than one way to look at the problem, and more than one way to solve it–and that reasonable people will have different answers about both the problem and the solution.

Another example where it might be acceptable to show properly truncated information (or ‘fragments’) could be the display of really long file names (say > 80 characters) on the desktop. Although I have plenty of desktop space, long file names are generally truncated to 2 line rows below the icon and displays the entire file name when the icon is selected.

2.) Yes. One can experience a wide range of emotions to a situation…again, depending on the context of the specific situation.

Fair enough. Now, you call this “Vista Rant #3″; you name the graphic above “Cansomeonepleasetesttheboundar”; you call yourself “I.M. Testy” here, and you say “Seeing this behavior the first time made me think I was losing my mind!“. That sounds like surprise and confusion to me. Are you sure that, as you say, “my pre or post emotional state had no bearing on the discovery of this defect”? And if it did, would that be a problem? Or might it be something that might give you important heuristic insight into the significance of the problems that you find?

“What I dislike are some of the simple anomolies that could be easily exposed…”–isn’t dislike an emotional reaction too? Could that emotional reaction tell you something about things that you’d like to change about the way Vista is tested?

Would you agree that passion is an emotional engagement with something? If someone said, for example, …it is the fervent passion for the discipline that I admire most in testers, would you agree with him?

Cheers,

—Michael B.

Monday, March 03, 2008 4:07 PM by Michael A Bolton

Two follow up answers:

1.) It depends on the context! In this particular case it is not possible to display more useulf information due to the size limitations of the window. Another example where it might be acceptable to show properly truncated information (or ‘fragments’) could be the display of really long file names (say > 80 characters) on the desktop. Although I have plenty of desktop space, long file names are generally truncated to 2 line rows below the icon and displays the entire file name when the icon is selected.

2.) Yes. One can experience a wide range of emotions to a situation…again, depending on the context of the specific situation, the person, and the person’s frame of mind (mental state) at that time in that specific situation.

Monday, March 03, 2008 1:36 AM by I.M.Testy

Would you agree that showing fragments of useful information where it’s possible to show more useful information could be viewed as a bug?

Would you agree that surprise or confusion are emotional reactions to a situation?

—Michael B.

Sunday, March 02, 2008 3:38 PM by Michael A Bolton

Yes, you are quite right. Boundary conditions not only occur at the extreme ranges of linear values, but they also often exist at the edges of equivalence class subsets. This particular boundary defect could have been found via testing the appropriate invalid class subset for width < length of filename window, or specifically via boundary analysis applied correctly.

I think your comments in the second paragraph really speak to our differing perspectives, and as you say having multiple models or approaches is arguably favorable. That is why I appreciate the symbiotic relationship between exploratory testing, systematic analysis, and other approaches.

The value I see in identifying this as a boundary error is that it gives the developer a specific point of reference, or value to consider in fixing the defect.

The listview controls in Xp and Vista are different animals and Vista provides some cool additional extended properties. The reason why there are column headers in all views is because the LVS_EX_HEADERINALLVIEWS extended style is set to show the column headers in all the view modes to allow the user to sort the view in different view modes.

The file description is truncated because the image of the file is actually a rendition of the BMP image, while for the more generic file types it is an ico file that is displayed. Personally, I wouldn’t consider this a boundary defect because the truncation is visibly obvious as illustrated with the elipsis characters. Essentially, the size of the window that displays the file information resizes correctly and withn the boundaries of the Explorer window.

Personally, I wasn’t annoyed by the Vista bug. I really don’t get annoyed very often, and software rarely annoys me. So, I don’t want to burst your emotional bubble, but my pre or post emotional state had no bearing on the discovery of this defect. I discovered this defect using my skills of observation, my cognitive abilities to know this was a defect, and my in-depth knowledge to understand why it is a defect.

Automated tests at Microsoft are always designed and developed by engaged human testers. (We unfortunately don’t have a magic machine that spits out automated tests, and we rarely rely on simple record/playback or rote scripts.)

I agree, engaged testers who are understand the functionality in-depth, and approach the problem from multiple perspectives (exploratory, systematic analysis, automation, etc) provide an excellent defense against the pesticide paradox.

Friday, February 29, 2008 3:11 AM by I.M.Testy

This is an example of why I think we should pay careful attention to emotions like annoyance, or surprise, or confusion. These are contributions that humans can make that machines can’t detect. By all means, leverage machines for what they’re good at, but they’re just tools. They shouldn’t be mistaken for the consciousness that wields them. Now that you’ve been surprised and annoyed by the Vista bug, you can probably code a test for it, but until the point that you recognize that kind of a problem, automated tests will be blind to it.

An engaged human tester is an excellent defense against the pesticide paradox.

Friday, February 29, 2008 1:54 AM by Michael A Bolton

1) a difference between the highlight on /a folder, with a folder visible/ and the highlight on a /file with no folder visible/ (the transition between the second screenshot and the third); and

2) a difference between a file highlighted /when a folder is visible/ and a file highlighted /when a folder is not visible/ (the difference between the third screenshot and the fourth).

There appears to be different behaviour based on these distinctions. I could describe it in a number of ways, but a boundary-based explanation would be come retrospectively to me. That is, I would conceive of a boundary only after I had seen the behaviour. But I’m also not sure how helpful it is to describe it as a boundary anyway. I think I’d prefer to think in terms of welcoming multiple models, rather than trying to retrofit an explanation onto my favourite one. (Except that having lots of models IS, arguably, a favourite model.)

I was more curious as to why there’s a “date modified” column header visible, but no dates visible. (I’m travelling, and have no access to my kids’ machine, which is the only Vista machine I’ll let in the house.) So I tried to reproduce this under XP. It took me a while to recognize what was going on. I realized that your problem above depends on multiple columns. Then I realized that multiple columns are available only under List mode. Under XP’s Classic View, when you ask for List Mode, there are no details and no column headers. In Vista, there are column headers in list view, but (apparently) no details under those column headers. So, to me, I could report the bug, but really I want to report an issue: what’s with the column headers?

Friday, February 29, 2008 1:29 AM by Michael A Bolton

In XP, you can right-click anywhere on the row to select the row. E.g. try right-click, delete. Yes, it deletes the file you thought it would.

In Vista, when detail view is shown, you can only right-click on text. E.g. right-click, delete in space on a row. You’ve probably just lost a folder or something.

Nice, eh?

I’d hoped that this would have been fixed in SP1 as it is a dangerous UI change.

Wednesday, February 27, 2008 3:32 PM by DPilcher