Where does the taskbar get grouped button titles from?

Date:April 8, 2004 / year-entry #135
Orig Link:https://blogs.msdn.microsoft.com/oldnewthing/20040408-00/?p=39883
Comments:    52
Summary:If the "Group similar taskbar buttons" box is checked (default) and space starts to get tight on the taskbar, then then the taskbar will group together buttons represending windows from the same program and give them a common name. Where does this common name come from? The name for grouped taskbar buttons comes from the...

If the "Group similar taskbar buttons" box is checked (default) and space starts to get tight on the taskbar, then then the taskbar will group together buttons represending windows from the same program and give them a common name. Where does this common name come from?

The name for grouped taskbar buttons comes from the version resource of the underlying program. You can view this directly by viewing the properties of the executable program and looking on the Version tab.

To set this property for your own programs, attach a version resource and set the name you want to display as the FileDescription property.

Comments (52)
  1. Jeroen says:

    Back from holiday? Or did you forget to append the [notification] on this entry? :-)

  2. Jeremy Bloath says:

    On Win2003 server, it doesn’t appear to be that simple:

    I recently replaced notepad.exe with metapad.exe: Renamed it, replaced the copy in dllcache. It’s there to stay.

    Now, when I start up several instances of %SystemRoot%notepad.exe via the SendTo menu or "association", I get metapad.exe, but when the icons get grouped in the taskbar, it says "NOTEPAD" in the taskbar, accompanied by the Windows notepad icon, the blue one. That icon is not in the executable.

    If I start %SystemRoot%notepad.exe using the Run command in the start menu, or from cmd.exe, the system calls it metapad, uses that icon, and puts it in a different taskbar group from the instances that it calls notepad.

    If I rename metapad.exe to "foo.exe" somewhere else and run it, the taskbar group is named "foo", and again it’s a distinct group.

    Finally, metapad.exe has a "spawn new instance" command. When I use that, the new instances belong to whichever taskbar group their parents belonged to.

    In all of these cases, the icon for each instance in the popup taskbar group menu is the first one in the executable.

  3. Spong says:

    It seems intuitive to have the most recently used windows sorted nearest the taskbar, so that they’re selectable quickly, in much the same way that recently used windows alt-tab first.

    This would prevent having to "read" the contents of the taskbar item to switch to the next one.

    Please consider adjusting this for Longhorn (I’m assuming it’s too late for SP2 now), it’s a real productivity waster and the main reason I turn grouping off :(

  4. Raymond Chen says:

    Everybody seems to have a different opinion on the "correct and obvious" order the buttons should appear…

  5. Anonymous Coward says:

    Raymond, perhaps the fact that everybody has a different idea indicates that grouping is not as convenient as it should be? Yes, It does clean up the taskbar, but it feels like bandaid applied onto the original Windows 95 UI.

    Could we hope for something like the Mac’s Exposè on Longhorn? Or perhaps something even better?

  6. ed says:

    I would like the ability to change a window’s ordering in the taskbar by dragging it.

  7. David S says:

    I like how tasks are ordered by program (even when not grouped in a single button). If I have a browser window open, I know where subsequent browser windows will appear. It sorta divies up the space nicely and I know what’s where without actually looking at the taskbar.

    That said, I’d still like to see native multiple desktop support to help save space. Twould be nice.

  8. Wilhelm Svenselius says:

    Code them all, turn only the best one on by default, and make the others accessible through a reg hack (TweakUI). Of course, that would take way too much time.

    Wild tangent: Will the PowerToy calculator in XP be the default calculator in Longhorn, will there be something better, or are you sticking with the crippled default one? I don’t really see a reason why XP couldn’t ship with a fully featured scientific/graphing calculator (Ti-83+ equivalent, at least), seeing as most of the features required are real easy to code (or can be grabbed from a free algo library) when you’re not restricted to a calculator-size memory space.

    Or would that be "leveraging the monopoly"?

  9. Raymond Chen says:

    "Code them all": I hope this wasn’t a serious suggestion. Just the testing cost would be enormous.

  10. Raymond Chen says:

    We receive all sorts of suggestions via http://register.microsoft.com/mswish/suggestion.asp including the ones above. Feel free to add your vote there.

    Noting of course that it isn’t a popularity contest. Just because it’s the most popular option doesn’t mean it’s the *best* option. And of course all features need to be balanced against each other. Spend a month on grouped taskbar button customization and you lose a month on something else. ("A month?" you ask. Remember, there’s a lot more to a feature than writing the code.)

  11. Robert Morris says:

    Ed, I just read about a program that lets you do that (in a recent Langalist edition). See: http://mlin.net/misc.shtml

    I haven’t tried it yet, but I just finished downloading it myself.

  12. Mr. Cynic says:

    Oh brother! If it weren’t for the occassionally insightful comments, I might think Norman Diamond is just a troll.

    You are assuming:

    a) the people who submit feedback are representative of all users who use the feature. Online polls and online feedback are highly self-selecting for certain types of customers.

    b) the people who submit the feedback are fully versed in all the appropriate issues and tradeoffs, as well as the effect of making the change on the rest of the system.

    c) the "most popular" is significantly the most popular. What if 10 different options each got 9% of the vote but the 11th option got 10%. Is it worth the work to make the change to annoy 90% of your customers by changing the behavior?

    d) the people who submit the feedback are submitting feedback on the right topic to begin with. How many times to people ask how to do X when afer further questioning they really want to do Y and just think that X is the best way to do it (when its not).

    Ford does not design its engines by taking a vote about various borestroke values. They evaluate the feedback ("I want a cylinder size X"), consolidate all the feedback and determine the users overall goals ("more horsepower") and evaluate all the various ways to achive that goal (more cylinders, different gear ratios, different engine parameters, variable cylinder timing, whatever). Then they choose the one that best meets the goal while taking into account all the various tradeoffs (Manufacturing cost, fuel efficiency, reliability, etc) and utilizing their car-design expertise to keep the car from exploding as its driving down the highway.

    Product design isn’t high school. Raw popularity doesn’t mean everything.

  13. Raymond Chen says:

    If popularity = quality, then McDonalds would be the best restaurant in the United States. A 0% income tax rate would probably be pretty popular too.

  14. Wilhelm Svenselius says:

    Raymond, don’t tell me Microsoft couldn’t afford it if they really wanted to. :) The hard part would be to justify affording it.

  15. Norman Diamond says:

    4/8/2004 8:26 AM Raymond Chen:

    > Just because it’s the most popular option

    > doesn’t mean it’s the *best* option.

    Normally when a company takes an attitude like that the company dies. The exceptions are monopolies.

  16. Spong says:

    Of course, I think having the most recently used item near the click-point is the best design!

    Raymond, I’d be interested to see a screenshot of your typical work desktop (sensitive info obfuscated, natch), so I can see which features you like enough to keep turned on yourself. I know you’re not representative of a typical user, but it’d be a fun exercise to see how Shell people live!

  17. Peter Evans says:

    Jeremy Bloath,

    Maybe your icon problem is because you’ve effectively spoofed the shelliconcache with your replacement. I haven’t worked with shell programming much, but I seem to remember having issues with this once when working with desktop shortcuts.

    PS I have seen a developer feeding frenzy. I couldn’t resist the painted corner Joel was in. :)

  18. Peter Evans says:

    "Code them all" How about the other extreme "Code none at all"? :) Just let the CUSTOMER create/install their own personal explore.exe :D Universal user preferences for all… err chaos. :(

    Smart shell features are really difficult to get right. There are just so many things dependent on the shell that aren’t even code related issues. What was that recent blog entry by Raymond about the default user picture??? HMM.

    Half the reason I enjoy Raymond’s entries so much. They really do give insight into the nonsense and noise involved in simple changes to the human interface design issues from a behind the scenes perspective.

  19. Raymond Chen says:

    Making the taskbar buttons reorder themseles by MRU was explicitly considered and rejected by the designers. Having the taskbar buttons move around as you clicked on them was just too confusing.

    My settings are absurdly retro because I imprinted on the old shell. So it’s classic Start menu for me with everything renamed so I can launch programs by just typing the first letter. E.g., to start a copy of IE, it’s WindowsKey PAII. (Programs, Accessories, Internet, Internet Explorer.)

    The thing is, this reordering is largely irrational, but it’s what I’m used to so I keep doing it.

    I do not consider myself a typical user.

  20. Duff McGee says:

    That’s interesting. I wasn’t aware that people actually used the WindowsKey. I’ve never seen anyone use it, nor used it myself.

    But I guess everyone starts programs the way that seems the most fast and simple to them.

  21. John Topley says:

    Weird, I use the WindowsKey all the time, usually WinKey + M to get back to the Desktop. Of course, many people aren’t aware of its existence…

    Hey Raymond, how about a blog entry on the story behind the introduction of this key and its even lesser-known neighbour?

  22. 4nd3r$ says:

    ah the win key is great, win+m,win+e,win+f :D

  23. 4nd3r$ says:

    btw raymond, is it possible to change the path when u press win+e (i want my documents and not my computer)?

  24. Raymond Chen says:

    It goes to My Computer for compatibility with Windows 95.

  25. Mat Hall says:

    One thing that annoys me about the taskbar is that if you move it to the top occasionally windows get wedged behind it, requiring Alt+Space M to move it to a more sensible location; also maximised apps sometimes get confused, leaving a blank space at the bottom of the screen where the task bar normally lives.

  26. Norman Diamond says:

    4/8/2004 8:40 PM Mr. Cynic:

    > You are assuming:

    followed by a list of things that someone was assuming. Mr. Cynic, who were you talking to? Who was making those assumptions? Or are you just trolling?

    4/8/2004 9:37 PM Raymond Chen:

    > If popularity = quality, then McDonalds

    > would be the best restaurant in the United

    > States.

    You almost have a point. But asking (including scientifically asking) which restaurant people think has the best food would very likely yield a different answer from asking (or observing) which restaurant people most frequently find convenient or affordable. You might also notice that McDonalds got where they are by doing what a lot of customers want but without forcing it down their throats.

    4/9/2004 5:54 AM Raymond Chen:

    > Having the taskbar buttons move around as

    > you clicked on them was just too confusing.

    Do you think it’s less confusing when the taskbar buttons move around randomly? (Only part of it is random. When a button moves it moves to last place, but it’s random when it will happen and it’s random which button it happens to. So far I’ve only seen this in XP, but intuition says 2003 is likely the same.)

  27. Why in the Start menu menu of most used programs sorted from top to bottom?

    IMHO mostly used programs must be more close to the Start button? than rarely one, i.e. head of this stack must be at bottom.

    sorry for bad language.

  28. Raymond Chen says:

    But people read from top to bottom.

  29. Raymond Chen says:

    Tell me it isn’t confusing to have something move *as you click on it*.

  30. Mr. Cynic says:

    Very true, my bad. I apologize. I assumed you had made those assumptions.

    Based on your objections to my assumption, are you saying that you did consider all of those issues and rejected them?

  31. Raymond Chen says:

    That’s not what I said. I was just pointing out that there’s a cost to every feature.

    (Some of the suggestions above were considered and explicitly rejected, like making the buttons reorder themselves based on frequency of use.)

  32. Alexey Volchkov says:

    4/12/2004 8:02 AM Raymond Chen

    >But people read from top to bottom.

    Not in this case. The sight goes from the point on which it was concentrated before. When i press the Start button (by mouse or by Win key) is only Start button visible, and when the start menu is appearing i am read from bottom (from the Start button) to top.

    And also the most used must be the most accessible(i.e. less mouse moving and keyboard use).

    And i think, that the autorearranged buttons in the taskbar and draggingdropping it features are not good ideas.

  33. Norman Diamond says:

    4/12/2004 4:17 AM Raymond Chen:

    > Tell me it isn’t confusing to have something

    > move *as you click on it*.

    You’re right. You are, and I agree. Nonetheless this reminds me again of one of the taskbar bugs I recently mentioned, where the taskbar autoscrolls as you click on a button. This results either in a no-op (except for the autoscroll) which requires scrolling back to the desired position and clicking on the button again (which works this time), or else actually restoring the minimized button in the clicked location in the taskbar row that got autoscrolled to. Yes it is confusing for a while, and it’s irritating permanently.

    But drag and drop might be equally sensible as it is in the menu entries of the Start menu, and an option to enable/disable drag and drop might be equally sensible. (Talking about the Classic Start menu here; I’ve never tried it with the XP version.)

  34. Raymond Chen says:

    Yes it’d be nice to be able to drag/drop the buttons round, but you have to do the cost analysis. How many people will benefit from this feature? How much is the benefit? Does the benefit outweight the opportunity cost? If you decide to do this, then you have to decide what other features *not* to do, since you have finite resources. (See "The Mythical Man-Month".)

  35. Chris R says:

    This is true Alexey, but if you’re a wierdo like me and have your task bar up the side of the screen, then the start button is at the top left, and level with the menu when it appears :-)

    I personally would like to be able to move things around on the task bar, and I might even like to be able to group things together myself somehow, although I think it’d be a UI nightmare to figure out how it should work.

  36. Mr. Cynic says:

    Sorry Raymond, I wasn’t clear – I was referring to Normand Diamond’s complaint that I had unfairly claimed he had not considered a lot of issues in his "only monopolies can choose to not use the literally most popular solution"

    I certainly agree every feature has a cost.

    Unless you weren’t responding to my comment at all. Unthreaded conversations. Arrgh!

  37. Norman Diamond says:

    4/12/2004 10:05 PM Mr. Cynic:

    > Very true, my bad. I apologize. I assumed

    > you had made those assumptions.

    Apology accepted.

    > Based on your objections to my assumption,

    > are you saying that you did consider all of

    > those issues and rejected them?

    ‘Fraid I don’t understand your question. I didn’t make any assumptions. I merely pointed out the observation that, normally, companies which act on their own ideas of what’s best and ignore what the marketplace says is best (in whatever way for whatever reason) tend to die. And the exceptions are monopolies.

    The American saying "The customer is always right" is a slight exaggeration, but it’s full of meaning. If the customer knows what they want then the customer knows what they want. If you want to sell to a customer then the customer is usually more concerned with the customer’s needs than with your needs, and you’d better direct yourself to be concerned with the customer’s needs. Except when you have a monopoly.

  38. Raymond Chen says:

    The most commonly requested feature is not necessarily the one most people want. People who make requests are self-selected.

    Microsoft does a lot of end-user research and incorporates large quantities of feedback from "normal" people, since "normal" people are the overwhelming majority of computer users.

    Let’s face it: We’re not normal.

    If you took all the people who read this blog (optimistically: 2000 people; realistically much less), they’re completely insignificant compared to the 150 million computer users just in the United States.

    So be careful extrapolating the most popular request among the computer elite to what would be the most popular request among all computer users.

  39. Norman Diamond says:

    4/13/2004 10:46 PM Raymond Chen:

    > Let’s face it: We’re not normal.

    Agreed. For example normal people don’t ask for bug fixes because they think they’re doing something wrong themselves. But if you did scientific research and let end users compare the operation of a machine with a bug fix against a machine without the fix, you might find that normal people need them just as much as us abnormal people do. But I still can’t figure out that animated dog. Even though normal people need less powerful search tools than us abnormal people do, how many of those normal people really said that the animated dog made it possible for them to get their results?

    By the way yesterday I was trying to search a directory for all files containing the string "HAVE_CONFIG_H" (without the quotes). The animated dog said there were none, not even the file that I was looking at (not modifying) in an editor. So I tried throwing other bones at the dog. Searching just for "CONFIG" (without the quotes), it found a ton of files. Well, the dog’s handler says I can search for either a word or a phrase. Maybe "HAVE_CONFIG_H" in a file is a phrase, containing the word "CONFIG", but my first search string "HAVE_CONFIG_H" wasn’t a phrase when it was in my input to the doggy instead of in a file. So I tried telling the doggy to search for phrase "HAVE CONFIG H" (without the quotes, but i.e. as a phrase with spaces instead of underscores). No luck, the dog didn’t find them either. Anyone remember when grep was invented? Anyone know how to teach a new dog old tricks? Do normal customers really demand this garbage?

  40. Raymond Chen says:

    Normal people don’t create files of unknown type with an ascii editor. They search word processor documents or spreadsheets, which have filter handlers.

    If you want you can attach the ‘ascii text file’ filter to file types that you know are ascii files.

  41. Centaur says:

    Sometimes I wonder why I am using an OS designed for “normal people” :)

    (Of course, I use it in a highly un-normal way, and normal people would have a hard time trying to use a system tuned for me…)

  42. Norman Diamond says:

    4/14/2004 5:53 PM Raymond Chen:

    > Normal people don’t create files of unknown

    > type with an ascii editor.

    OK. Then Microsoft should direct OEMs to agree to refund the corresponding portion of the purchase price when abnormal users want to return the normal OSes that came on our machines. Even though Linux is as far from perfection as Windows is, at least grep works.

    > If you want you can attach the ‘ascii text

    > file’ filter to file types that you know are

    > ascii files.

    I don’t understand. After reading that, I found the doggy’s detailed setting option for file variety, and found text document as a settable variety. Then the doggy found no matches at all. I think this filter limits the doggy to look at filenames ending in .txt even though I didn’t put that restriction in the filename box. Where can I find the filter you’re talking about, that will make the doggy read all files as if they were plain text?

    (By the way, "plain text files" still doesn’t mean ASCII. Whether or not it includes Unicode, it still includes all code pages.)

  43. Raymond Chen says:

    The doggy doesn’t let you create new file types. Look: clearly the doggy is not for you. Stop trying to use the doggy for geek things. He’s not for geeks. He’s for my grandmother.

  44. Norman Diamond says:

    OK, so where the doggy gives an option to search for files containing a specified word or phrase, and I type in a word or phrase and the doggy doesn’t find files containing it, the reason is because the doggy is for your grandmother. You and I agree that your grandmother doesn’t expect the doggy to work.

    Nonetheless, what kind of filter were you talking about? And/or where is Windows’ equivalent of the "grep" command?

    Here’s some feedback from a Microsoft customer about the need to make the search utility easier to use: I need an animated cat, because cats know how to find things.

  45. Raymond Chen says:

    Grandma does not create files with unknown extensions and expect the doggy to find them. Grandma creates files with Lotus 1-2-3 or WordPerfect. Grandma wants to look for that letter from her newphew where he talks about his soccer team, and she finds it even though WordPerfect files are not plain text files (so a grep would have failed).

    findstr is the Windows version of grep. Go nuts. Just don’t expect it to work very well on things that aren’t plain ASCII text files.

  46. Norman Diamond says:

    4/15/2004 6:10 PM Raymond Chen:

    > findstr is the Windows version of grep.

    It worked, thank you.

    > Just don’t expect it to work very well on

    > things that aren’t plain ASCII text files.

    You’re right, and you’re more right than you have any right to be. In the Japanese version of Windows XP, "findstr /?" (without the quotes) displays help information in Japanese, and it says the search string should be a search string, and it doesn’t say ASCII. But then, exactly as you said, when I searched for a perfectly good plain SJIS text string in a bunch of perfectly good plain SJIS text files, findstr said the parameter was bad. I had to put the search string in quotation marks. A perfectly good plain SJIS text string, one word consisting of two Kanji characters whose meaning in English is "location", no spaces or punctuation or anything, and the findstr command complained unless the word was in quotes.

    Anyway thank you, it beats the doggy all to pieces.

  47. Norman Diamond says:

    Sorry for the second post in a row ^_^

    The doggie has grown weirder. In Windows Explorer I right-clicked the icon for my C drive, selected Search, and typed in the filename.txt of the file that I was looking for. It found 5 instances, of which 2 really exist, 1 is a shortcut in the recent links folder, and 2 seem to result from smoking illegal doggie drugs. Annotated screen shot at:


  48. It comes from the EXE itself.

  49. 4nd3r$, I know it is not as easy as using WIN+E, but I make a shortcut to Windows Explorer to specify the starting directory with a 2 pane view (i.e. showing the directory tree) like so:

    %SystemRoot%explorer.exe /e, C:

    Here’s a URL I found that shows a bunch of different ways to start Windows Explorer:


    Here’s a URL that explains the different switches:


  50. It comes from the EXE itself.

Comments are closed.

*DISCLAIMER: I DO NOT OWN THIS CONTENT. If you are the owner and would like it removed, please contact me. The content herein is an archived reproduction of entries from Raymond Chen's "Old New Thing" Blog (most recent link is here). It may have slight formatting modifications for consistency and to improve readability.

WHY DID I DUPLICATE THIS CONTENT HERE? Let me first say this site has never had anything to sell and has never shown ads of any kind. I have nothing monetarily to gain by duplicating content here. Because I had made my own local copy of this content throughout the years, for ease of using tools like grep, I decided to put it online after I discovered some of the original content previously and publicly available, had disappeared approximately early to mid 2019. At the same time, I present the content in an easily accessible theme-agnostic way.

The information provided by Raymond's blog is, for all practical purposes, more authoritative on Windows Development than Microsoft's own MSDN documentation and should be considered supplemental reading to that documentation. The wealth of missing details provided by this blog that Microsoft could not or did not document about Windows over the years is vital enough, many would agree an online "backup" of these details is a necessary endeavor. Specifics include:

<-- Back to Old New Thing Archive Index