When the Web page says that a tool is not supported, it means that if you find a problem and contact technical support, they’re just going to point you back to the Web page


I file this under the category of People refuse to read what is right in front of them.

There used to be a number of utilities available for download which all go by the name PowerToys. And they all contain text like this:

Note: We take great care to ensure that PowerToys work as they should, but they are not part of XYZ and are not supported by Microsoft. For this reason, Microsoft Technical Support is unable to answer questions about PowerToys.

This sentence isn't exactly in the fine print either. It's right there at the top of the page.

This nevertheless does not stop a customer from contacting their support representative, who then turns around and sends email to the product team that goes something like this:

I am working on a customer case, 602214129.

The customer reports that the XYZ PowerToy does not work under conditions A, B, and C. My questions are

  1. Is a fix available for this?
  2. What are the support limitations around this tool?

The tool was downloaded from ⟨http://www.microsoft.com/…⟩

Somebody then has to point the customer liaison to the text right at the top of the Web page they linked to that quite clearly spells out the support limitations of the tool: It comes with no support. Apparently both the customer and the customer liaison failed to notice it.

Comments (35)
  1. Maybe a peer-to-peer support group would work.

  2. Anon says:

    Microsoft support reps on the groups are substantially worse than useless 90% of the time.

    They rarely actually bother to read questions; when they do read questions, they rarely understand even the most basic concepts involved, when they aren't skipping half of the question.

    When they provide answers, I don't believe they ever actually look at the answers. The MS support forums are littered with posts from Officlal MS Support Reps who respond to questions about, say, the Windows Mail client with 'solutions' that involve updating Exchange Server.

    What I'm saying here is that the vast majority of outward-facing support, ESPECIALLY the non-US/EU support, is drastically in need of education (or re-education).

    I assure you, the Customer Liason didn't 'fail to notice' the warning. They never even bothered to open the site, so they never encountered a warning to miss.

  3. Aaron says:

    When the support-rep writes, "What are the support limitations around this tool?", it suggests to me that they already KNOW it is not supported.

    But they don't want to be the bearer of bad news to their customer. They are hoping-against-hope that there is some sort of support or fix, even though they know there probably isn't.

    They're forwarding the question up the chain of command, so that they are not the "bad guy", they are only the messengers who deliver the bad news to the customer.

    "I'm sorry, Mr. Customer, but I checked with the Chief Architect/Developer/Manager, and he was unmoved by your plight. He referred to this website that states there is no support. I fully sympathize with you, but unfortunately, there is nothing more I can do on your behalf. I tried my best, you understand.". :)

    [I suspect you're right. -Raymond]
  4. Jack B Nimble says:

    I thought the first rule of the user experience is that the user will not read anything plainly put in front of them.

  5. Simon says:

    @Anon – I can only speak of my experience but the MS Customer Reps I've met in the UK have been pretty useful. That's a Rep who is responsible for one company, which is what I've always assumed Raymond is talking about in these examples. Agreed, some of the stuff on the forums is….not helpful.

  6. alegr1 says:

    @Anon: totally agree. The MS reps in the forums not only don't care to give a meaningful answer, they don't care to pass the problems to developers.

  7. Jim says:

    @Anon and alegr1: I donot think this problem is unique for MS. Any large company has the same problem. I donot think that the custom rep even knew where to send the customer request. Plus, it is not general his or her job to make sure the problem passed on, it's her/his job to record the problem!!

  8. Joshua says:

    Getting real solutions from MS seems harder than pulling teeth to me.

    Supported or unsupported doesn't seem to make any difference anymore.

    I have 100% failure rate getting solutions.

  9. Roger says:

    I'm a Microsoft employee – technical field roles for 17 years.  Snarky comments aside, I can tell you this: we are typically measured/reviewed based on customer satisfaction and service contract renewals.  I will do whatever I can to assist my customer even if I know what the outcome will be.  There is always a chance that the person who wrote the PowerToy will see the request and make a quick fix.  At worst, there will be no support and I have at least TRIED to assist the customer with the request.  Chances are this support representative actually DID read the website.

  10. Another approach (perhaps combined with peer support) is to ship the source for the toys.

    Then if the issue is in the toy, the customer has a chance to fix it (or more likely to find a peer dev to fix it for them.)

    And if the bug is below the toy, this is no longer a "the unsupported toy is broken" support case, but rather a "the OS is broken" support case, which the product group is likely to find much more interesting.

  11. Yuri says:

    I also second that MS support is one of the best in the industry, when you have substantial MSDN subscribers in your corporation.

  12. Joshua says:

    > "the OS is broken" support case

    Right up to when they won't. I reported this first in W8 beta, re-reported in W2012, re-reported in W8.1 beta and still not fixed. Number of affected users that care has got to be in the thousands. For those affected, this is an accessibility issue.

    All MS responses have been incredibly stupid ones about graphics drivers.

    answers.microsoft.com/…/f52a2002-d748-485c-8b9c-4862806ea94a

    [It's possible that nobody has figured out how to reproduce the problem, which makes it difficult to investigate and fix. -Raymond]
  13. Joshua Ganes says:

    Looking at this from the customer's perspective is kind of sad. The customer uses the power toy enough to encounter a bug and cares enough to ask for help. Sadly, the project has been abandoned by its creator, leaving no realistic channel for support. This customer is left grasping at straws for a solution to his problem that will probably never be addressed.

  14. marksolsoft says:

    @Joshua's issue/Raymond's response: I suspect the main issue is the design decision that ClearType is no longer available in a bunch of MS apps (including IE and the newest Office), regardless of OS setting.  So you end up with a mishmash of subpixel-smoothed and grayscale-smoothed text, and it's pretty horrible.  From what I gather, there were some bugs too, but you'll see a lot of howling is about ClearType.

  15. Joshua says:

    [It's possible that nobody has figured out how to reproduce the problem, which makes it difficult to investigate and fix. -Raymond]

    I cannot believe that what I can reproduce 100% of the time by taking a snapshot of the initial metro screen MS can't reproduce at all.

    [It turns out that nobody at Microsoft has your computer, with its exact mix of drivers, applications, custom settings, etc. You happen to have access to such a configuration 100% of the time. I find it surprising that you've never heard of a configuration-dependent issue. -Raymond]
  16. David Crowell says:

    I'm going to say that paid support from Microsoft is great.  I had an issue at a previous job with performance between a web server and database server that just shouldn't have existed.  The tech assigned to used a memory dump (I think – it's been a while) to diagnose the problem.

    The issue was that our weird network configuration hit a corner-case bug in the OLEDB driver.  We didn't even have to pay for that case, since it was an MS bug.  :)

    This was ten years ago, the details escape me.

  17. Christian says:

    I also found the MS support to be excellent a few years ago.

    From reading about the font rendering problem, the gist seems to be that Windows 8 is optimized for tablets, and tablets can rotate the screen by 90° and then cleartype does not work anymore so well, because R G B pixels are not next to each other, but below/above each other. Therefore allegedly Microsoft has disabled ClearType with colours and instead uses greyscale antialiasing for IE10 on Win8 and also in Metro. That's sad and should be VERY easy to reproduce.

  18. Azarien says:

    What I find annoying is that Microsoft employees' answers usually come duplicated.

  19. Azarien says:

    What I find annoying is that Microsoft employees' answers usually come duplicated.

  20. 640k says:

    1. The linked page does not work. Please fix asap.

    2. If the power toy would have been open source the customer would have had a chance to fix it themselves. Now it's just broken abandownware which will never be fixed.

    [(1) I'm surprised somebody as smart as you doesn't know how to use the Internet Archive Wayback Machine. (2) The source code often incorporates internal libraries which aren't open-source. So even if you had the source code that wouldn't help you. -Raymond]
  21. John Doe says:

    Actually, from Vista on and even from Windows CE, Windows Mobile 6 on I believe, ClearType has the ability to render properly in either RGB, BGR, horizontal or vertical.

    For IE10, I can't tell why it doesn't use ClearType and just detect/respond to the current orientation, since it'll have to reflow. No excuse.

    For Office, it could be due to buffering pages to save battery. But when you rotate the device, you've already paid for the rotation animation, my guess is that redrawing the current visible page(s) wouldn't be much more trouble, the CPU and probably the GPU already woke up to full speed. So again, no excuse.

    Or maybe, ClearType rendering is just a CPU/GPU hog that's mostly unnoticeable with higher DPI that someone just decided to ditch it, as devices will have increasingly higher DPIs. If they're high-end, Of Course™.

  22. Engywuck says:

    If I read "We take great care to ensure that PowerToys work as they should" then I'd assume that – although support by "Microsoft Technical Support" is officially not available – they'd either be interested in (grave) errors anyway or could at least point me to someone who cares/has written the software/…

    Asking for "support limitations" is another thing in this case :-)

  23. Azarien says:

    The sad part is that "THERE USED TO BE a number of utilities available for download". Microsoft really should keep an archive of all old downloads (with remarks that they are not supported anymore). I hate searching the web for some old file that is increasingly hard to find.

  24. Brian_EE says:

    @Azarien: It said several toys were removed due to security bugs/issues. If MS had left them avaiable, then when one of them was used as a malware attack vector, there would be an uproar "MS left known vulnerable programs available to unsuspecting users!!!!"

    You can't have it both ways.

  25. Actually, I have seen both the good and bad of Microsoft support. When it comes to the enterprises customers and support contracts, Microsoft support personnel go out of their way to make sure not only the problem is solved but also that it is solved admirably.

    When it comes to consumer software, however, Microsoft support can be described as Disgraceful with capital D. In one instance, I had a support case related to Windows Live Essentials and the support employee asked me to send a log. I agreed but only on the grounds that the log is sent over a secure connection. (I even obtained a digital signature and signed my email for encryption.) He refused cooperation and swore that it wouldn't be necessary because the log won't contain anything that needs security… until I sent him a redacted screenshot showing that my entire email address book was included in that log. He never replied and the case ended then and there.

  26. DWalker says:

    @David C:  The Power Toys were NOT "abandoned by their creator".  They were, as far as I know, internal tools that were useful but did not meet the bar for making into a product (which carries a support burden, internationalization requirements, etc.).  But, since they were useful, they were released as unsupported tools FROM THE START.

  27. DWalker says:

    Sorry, I directed my previous post to the wrong person — wrong @.  Apologies.

  28. @Joshua says:

    Wow that link contains absolutely no information about any of the configuration used to reproduce the problem, for all we know you could have a substandard monitor, or a bad cable! (heck that the OP needed glasses!)

    Do you really expect anyone (let alone a higher tiered Support Agent, or forbid a Developer) dig though each one of those links trying to find what should have been included in the original post?

    Instead of complaining that no one is investigating the issue why doesn't the OP go read the legendary "How To Ask Questions The Smart Way" http://www.catb.org/…/smart-questions.html should more people do that they'd get better support!

  29. Joshua says:

    @@Joshua: The issue has been fully investigated many times over. The problem reproduces for certain users for all configurations on all machines except mac laptops.

    Had you read, you would have seen that IE and Metro are affected but Chrome and Firefox aren't. This eliminates monitor and cable issues. Since it also reproduces over remote desktop and in VMWare this eliminates remaining hardware issues.

    [Um, you do realize that a bunch of UI features change when you run them over remote desktop or in a VM? -Raymond]
  30. Andreas Rejbrand says:

    Interesting. You use U+27E8: MATHEMATICAL LEFT ANGLE BRACKET and U+27E9: MATHEMATICAL RIGHT ANGLE BRACKET to mark up your URLs. That cannot be semantically appropriate, can it? Just curious, why did you make this choice over, say, the more semantically neutral U+003C: LESS-THAN SIGN and U+003E: GREATER-THAN SIGN (which are commonly used for other things, like in SGML/XML markup)? Personally, I'd guess U+2329: LEFT-POINTING ANGLE BRACKET and U+232A: RIGHT-POINTING ANGLE BRACKET is the best option (if you want to be fancy and careful about semantics).

    [They are intended to represent a placeholder, not as an URL marker. (It just so happens that the placeholder is for an URL.) I agree that the non-mathematical angle brackets are more appropriate here. I missed them when I hunted for the brackets originally. -Raymond]
  31. alegr1 says:

    >Therefore allegedly Microsoft has disabled ClearType with colours

    That's just *marginally* better than rendering ClearType and showing it on a display with mismatched resolution (see Windows 7 installer).

  32. Joshua says:

    > That's just *marginally* better than rendering ClearType and showing it on a display with mismatched resolution (see Windows 7 installer).

    Oh yes that bug. ClearType should default to greyscale when using failsafe graphics drivers (this is what you get on install for any graphics card newer than the version of Windows). Can't be reproduced during development for the obvious reason.

  33. ender says:

    > Actually, from Vista on and even from Windows CE, Windows Mobile 6 on I believe, ClearType has the ability to render properly in either RGB, BGR, horizontal or vertical.

    It had that ability on XP as well, but you had to either use a webpage on microsoft.com (which included an ActiveX control), or the ClearType Tuner PowerToy to set it (both also allowed you to tweak the rendering a bit more).

  34. xpclient says:

    The powertoy I really miss is the shortcut target context menu powertoy. Thankfully on 64-bit Windows, WOW64Menu works similar to it.

  35. Joshua says:

    Ref: answers.microsoft.com/…/f52a2002-d748-485c-8b9c-4862806ea94a (linked above)

    Posting here for completeness.

    Somebody (quite possibly reading from here) posted an answer about allowing the gamma of greyscale anti-aliasing to be tuned. I was able to test this using the system gamma tuner. This does "correct" the problem, but an optimal setting pushes the gamma so far into distortion range the system is effectively 16 color. A useful gamma correction is something like 20 (or is it .05). This means a specific gamma correction for greyscale will fix the problem if uncapped to allow absurd correction values.

Comments are closed.