Thank you for being a valued part of the CNET community. As of December 1, 2020, the forums are in read-only format. In early 2021, CNET Forums will no longer be available. We are grateful for the participation and advice you have provided to one another over the years.

Thanks,

CNET Support

General discussion

Remove hardware bug?

Aug 28, 2010 11:13PM PDT

I'm running Win 7 Pro, 64 bit on a Dell XPS 8100 with 12 GB RAM. I've been playing with a new toy, a nook e-book reader, and I noticed an odd behavior that at least a few other users have also noted. I suspect the behavior reflects a bug in Windows 7 (and probably XP as well) but I don't know how to report the problem.

When I attach the nook to the PC using my micro-USB connector the nook automatically goes into 'USB Drive Mode'. The instructions for the nook state that after I eject the nook it should revert to regular mode but continue charging. That IS what happens if I use Windows Explorer to eject the nook (right click on the drive corresponding to the Nook and choose 'Eject'). However, that is NOT how I am accustomed to removing USB drives. Usually, instead of using Windows Explorer, I use the system tray service labeled 'Safely Remove Hardware and Eject Media' instead because it is more convenient. When I disconnect the nook using that technique, the nook does NOT return to regular functioning even though the drive corresponding to the nook disappears from Explorer. I have to disconnect the USB connector before I can use the reader.

Speaking from the perspective of a user, it is not clear that there should be any difference at all when I use one method or the other to disconnect a USB device.

I had not noted this behavior in the past, so I decided to experiment with some of my other USB devices. My rather ancient iPod Nano (2nd generation) appears to disconnect properly using either approach. My Garmin GPSmap 76CSx acts like the nook: When I 'Eject' using Explorer the device reboots and returns to normal operation, but when I disconnect using 'Safely Remove Hardware' the device stays in USB drive mode even though it is no longer accessible via Windows Explorer. Similarly for my Garmin Nuvi 255W, though that is a more complex situation because it automatically reboots and reconnects when it successfully disconnects so that device is never usable when connected via USB.

I did not do comprehensive testing, but I did note that the nook's behavior is similar using 32 bit Win XP Pro, SP3 on my old desktop PC. How I disconnect affects the device's behavior.

Since this behavior appears on multiple PCs using multiple USB devices I have to conclude that it is not a problem limited to my nook device, or even to nook e-readers in general. It appears to be something in Windows.

Is this a bug? Or just a poorly implemented 'feature'? How do I report it to Microsoft?

Discussion is locked

- Collapse -
If the book doesn't work after eject
Aug 29, 2010 5:22AM PDT

Then the bug looks to be in the book. You should contact them to ask for them to address their firmware in the book.
Bob

- Collapse -
It does work if I disconnect it ...
Aug 29, 2010 6:25AM PDT

The problem is that when I disconnect the device one way it behaves differently than it does when I eject it the other way. Since the behavior can be replicated on 3 different devices from 2 manufacturers under two versions of Windows I don't think it is clear that the problem is in the ebook reader's firmware.

- Collapse -
As an embedded author.
Aug 30, 2010 10:13AM PDT

The bug looks to be in the ebook as it is where the odd behavior is. That is, Windows is not on the ebook...

Bob

- Collapse -
As I said, that is not clear to me ...
Aug 30, 2010 10:50AM PDT

As a mere user I don't know that I appreciate whatever arcane distinctions there are supposed to be between the two methods for disconnecting attached devices. All I can say from certainty from my perspective is that one of the disconnect methods does not work properly. Since the problem exists for multiple devices from multiple vendors then it is obviously NOT a simple programming error in one device. Whether multiple manufacturers made similar errors (possible) or Microsoft made an error (not unheard of) is beyond my expertise but past experience suggests the latter is quite possible, and Occam's Razor dictates that the possibility be considered.

So, having made an effort to report the problem to Barnes and Noble, I also should make an effort to report to Microsoft. Which brings me back to the original question: How does a mere user report suspected bugs to Microsoft? I never have figured that out.

- Collapse -
Yes, it is arcane.
Aug 30, 2010 12:06PM PDT

It's a shame the ebook maker is not talking to you about this.

Let's keep this simple. The ebook is where the problem is happening.

Microsoft did not write or create the ebook so the bug is not theirs to fix. Hope this clears that up.
Bob

- Collapse -
Hmmm ...
Aug 30, 2010 12:52PM PDT

I won't continue to debate the issue with somebody who is better informed than I am about the details here, but I want to make sure you understand that absent unreported additional information not available to me (but possibly known to you) your position lacks a solid logical foundation. I could just as easily claim that the ebook is behaving the way it is because Windows did not send the appropriate signal during the disconnection process. Personally I have insufficient expertise to make a firm conclusion and I'm not sure why you are so dogmatic on the subject.

I'm not at all surprised that Barnes and Noble has been silent on this subject. I've become accustomed to lousy support in every portion of the tech industry from Microsoft on down. My own experience is that good quality support (eg: offering intelligible answers to simple questions) is the exception rather than the rule. That is not limited to the tech sector by the way. I recently had a bank representative tell me that he could not answer a VERY simple question unless I came to the bank branch in person and talked to one of the tellers. I've been a customer there for ~20 years, and they STILL tried to brush me off. Unbelievable.

- Collapse -
"Windows did not send the appropriate signal during the disc
Aug 30, 2010 1:07PM PDT

"Windows did not send the appropriate signal during the disconnection process"

You nailed the issue in that passage. But I am not debating with you but hoping to save you time in contacting Microsoft. Microsoft did not create or write the firmware or drivers for the ebook. This would be interesting to read how you fare there but I wanted to share with you that since it is not a Microsoft product, driver or issue you should find no help there.

If it is a disconnect issue, then the device is where the fix must lie.
Bob

- Collapse -
I think I understand ...
Aug 30, 2010 9:16PM PDT

Saving time is a good thing, and from an engineering perspective if Windows behaves in a certain way then the designer of the attached device has to make the device function in the existing environment. I can't argue with that.

OTOH, if you agree with the statement that "Windows did not send the appropriate signal during the disconnection process" then how is that NOT a bug in Windows?

I recognize that there will always be bugs in any software complex enough to have commercial application, but surely that does not mean we should fail to report them when they are discovered?

- Collapse -
Let's look at other changes from one Windows to another?
Aug 30, 2010 9:44PM PDT

As you compare Windows versions you will find other changes from version to version. This area is deeply discussed and you can make a case that changes of behavior are either bugs or simply changes. But the Microsoft juggernaut that it is, if Windows 7 changed how this works then it's the new "standard" and hardware makers get to fix it.

I could give you examples of other changes that resulted in hardware makers adjusting to Microsoft but all that would do is show that Microsoft is unyielding when it comes to this area.

For example Vista's method of obtaining IP addressed crashed many routers. Was that a bug in the OS or a fault in the router firmware? You could argue both sides. In the end the router makers updated their firmware and machines.
Bob

- Collapse -
Let's put it another way
Aug 30, 2010 11:33PM PDT

Let's put it another way, which may be a bit easier to understand.

Your expected behavior of the Safely Remove Hardware icon (which generally stopped being necessary with XP) is that it will go about unmounting the device and prepare it for removal from the system. This is based on several years and countless experiences which all end more or less exactly the same way.

Now you connect your Nook device, and it does not respond in the expected/desired way. So you have the experience of all the other devices which DO respond as you expect vs the one lone device that doesn't. THAT is how people like Bob know that the problem is with the Nook, not Windows.

And just a little FYI aside. Starting with XP, Microsoft altered the way USB devices are handled. By default they are set up so that you can just remove them at any time (they aren't being written to). The tradeoff is some loss of performance on file transfers. So unless you have gone out of your way to alter this behavior, you don't even need to go through the safely remove hardware routine anymore. As long as there aren't files being written to the device, you can just pull the cable any time.

And another FYI aside... In business there's something called the 80/20 rule. Basically, about 80% of your business is going to come from about 20% of your customers. You buying the odd book here and there over 20 years means you've probably actually COST the company money each time you've come in. The profit margin on whatever you buy simply won't even put a dent into the various overhead costs associated with running a business. So unless you're buying several hundred dollars worth of merchandise every visit, it doesn't matter if you were the first customer through the door of the first B&N store, and the first one through the checkout line. That's just business. We today, unfortunately seem to grossly exaggerate our own importance in the grand scheme of things. Sites like Facebook and Twitter tend to make people think that others really care about what they're doing. We didn't before these sites came along, and we still don't after. Do you really care that an old college buddy of yours got up to go get a soda? Unless he's going to grab you something while he's up, probably not. It's the same basic idea with companies like B&N, which likely won't be around much longer anyway. Seem to recall them putting themselves up for sale a few months ago, and it doesn't seem like they've found any takers yet. Sadly, they will probably go the way of Circuit City soon. It will be particularly sad, because not far from where I live is a small little strip mall that once had B&N, Bed Bath and Beyond, and Circuit City... One of those is already gone, another is circling the drain. Doesn't bode well for the one left standing.

- Collapse -
Actually, it isn't just one device ...
Aug 31, 2010 4:05AM PDT

In fact, it is not even a new behavior. And I don't think the problem is in my 'understanding'. It's just something I noticed because the idiosyncrasy became relevant with the new device. As I noted in the OP, my Garmin GPSs also show the same behavior as the nook, and the behavior is found in XP as well as Win 7. That is one reason I am skeptical of the claim that anybody can KNOW the problem is in the nook as opposed to Windows. The reason I did not notice the behavior in the past is that I rarely use my GPS in USB Mass Storage Mode, and the Garmin user manual does not specify any expected behavior on disconnect. As a result, I had no specific expectations. I noticed it with the nook because the user guide explicitly described what is SUPPOSED to happen when the nook is ejected.

Now that I know what is SUPPOSED to happen my expectations have changed. From an engineering perspective I've already spent WAY too much time on the topic, because there is a simple fix that reliably works. If I disconnect the 'wrong' way I can reconnect the nook and then disconnect it the 'right' way. Also, I really do not think I will want to read the nook often while it is tethered to the computer. From those perspectives the problem is fairly trivial.

OTOH, the consumer in me is REALLY annoyed. When I go to the Win 7 'Safely Remove Hardware and Eject Media' applet in System tray and right click the choice I see is 'Eject nook'. That sounds an AWFULLY LOT like the choice Windows gives if I right click on the entry for 'nook' in Windows Explorer and choose 'eject'. And yet the two choices yield different results. What that tells me is that EITHER one of the choices is acting inappropriately (ie, there is a bug in Windows) OR that they were designed to act this way (It's not a bug, it's a feature). In the first case, it should be fixed. I understand it won't be a high priority fix, and MS may never get around to it, but the bug has been there since XP days so it SHOULD be fixed. In the second case, it's a REALLY STUPID design decision that involves a convoluted interface with arcane distinctions that at least 2 different companies have not been able to implement correctly. That still sounds like a Windows problem to me. And techies wonder why consumers find computers confusing?

Still, I know where I stand in the greater scheme of things. MS is not the same company it was when I could call Redmond to ask questions about MBASIC and actually get a real person who had real knowledge of the product. It's not that I want to go back to MBASIC and CP/M. I just hate the depersonalization that seems to be an inevitable result of massive growth. Nor do I really expect B&N will be terribly responsive. How could they? At least one of their top leaders has publicly disparaged e-readers in general and admits he does not plan to use one. That tells me a great deal about their commitment to the nook as a product. I hope brick and mortar book stores won't go away because I like to browse, and I don't think I'll want to give up real books until the e-readers are a LOT better than the current versions. But I do understand what you are saying about large businesses not caring a whole lot about individual consumers.

That said, I do think the companies should at least PRETEND to solicit consumer input about bugs and 'features'. Wasn't that one of the themes in a recent MS ad campaign?

- Collapse -
You're right.
Aug 31, 2010 4:36AM PDT

It's why Apple tends to get it right. They control the entire eco-system and you don't see this very often there.

If you want to see another odd USB behavior try these things at the end of some USB HUB.

Remember you are dealing with product, firmware and software from over a dozen companies which often outsource to the lowest bidder.

Please share how you get along with your Microsoft encounter on this.
Bob

- Collapse -
Now you're laughing at me ...
Aug 31, 2010 11:51AM PDT
Please share how you get along with your Microsoft encounter on this.
You already know that I won't make any progress; I don't think it is possible to reach real human beings at MS anymore without paying a fee to open a support ticket. Actually, that was the primary motivation for the OP in this thread. How DOES one go about reporting a bug without paying for the privilege? I have not found a way yet. Though if memory serves there used to be some MS employees who jumped in on the occasional thread in these forums?

It was not so many years ago that it was possible for ordinary users to access humans at MS at least from time to time. They were not always helpful but at least they had some product knowledge. I do remember when I complained about some behavior of Outlook Express (maybe back in the IE 5 days) the response was something along the line of "you didn't really think a free product would function properly did you?" though I don't think I would have described OE as "free" since using it required purchasing Windows. But I digress.

You are being practical: Windows does what it does and everybody else had best remember that and press on. I'm being idealistic: Windows should have a rational interface and work properly. We know who wins that discussion, but there is a risk to the practical approach. MS has a spotty record regarding compliance with standards, and right now there is little reason for them to improve. After all, the 'real' standard is whatever the major MS product does.
- Collapse -
Remember you are right.
Aug 31, 2010 2:44PM PDT

Windows from version to version has changed how things work. The makers of cards, usb things get to dance with the changes.

You could call it a bug, Microsoft calls it an enhancement.

Microsoft's phone numbers are not hard to find but you pay to talk to their staff.

-> You ask how to get their attention which is pretty simple. Make it something that is a security issue. Cosmetic or usability does not get any attention.

This is why the device maker is your only hope here. They write, issue drivers and in the drivers could change how things behave. We also covered how in this case the device itself is part of the behavior so it's firmware is in play as well.

-->> What's the bottom line here? I found that one device would not charge and be able to use unless I ejected it. Fix? It now has its own USB charger.
Bob