Do you hit the same annoying Windows Vista crashing bug day after day?
Please, please, please click the 'Send information' button when you see this crash dialog.
Why?
If in the very unlikely event that you are the first person to encounter and report this bug, a new entry in our bug database is entered automatically.
If anybody else encounters the same bug, and reports it, our automated crash reporting system finds the correct bug in our database, and then updates a counter. (Basically, there is a field in the bug that indicates that X people on the internet have encountered this bug.)
If you don't report the crash, that counter is not updated.
Why is that important?
Our ship room (a bunch of guys who decide which bugs should get fixed and added to SP1, and which bugs are too minor to be fixed) rely a lot on this counter. If the counter reaches more than [redacted], we fix it.
So, every time you encounter any crash - hit that 'Send information' button. Please.
24 comments:
hat is funny about this is. It is common sense yet most people don't realize it. They are afraid microsoft is going to steal everything on their computer.
What about "this behavior is by design" issues?
Someone: This post wasn't about 'by design' issues - it was about crashing bugs; The shiproom approves all crashing bugs that 'occur frequently'
'Occur frequently' means that people upload their crash results, and we add them all up.
A crash is never 'by design' - that's the first thing they teach you in Windows dev orientation.
That "report bug" dialog box is so stupid. I hate when it comes up. Because so often, it only comes up when you're simply over-working the RAM and are forced to abort certain applications.
And no, I don't trust Microsoft to take data from my computer every time a program crashes.
Having everyone automatically report crashes won't fix bugs. It's up to Microsoft to do a good job making the program and making it stable.
And BTW: it would be much better if there would just be a "report bugs" button in the Windows Help window, so instead of bothering everyone even more than if there would just be a crash, and people would actually be able to give some real data about exactly what they were doing or clicking, and they would also be able to report when there wouldn't otherwise be a dialog coming up. And yes, there would have to be a bigger support team reading all the reports. Can't handle that?
I sent reports everytime an application crashes. Sometimes I get solution available for some of these errors... but when I try to open it, an error appears: Impossibile to Open the Solution.
So, there is an errors report tool to report "errors report" errors? LOL
It's useless, as Windows Vista.
I've wondered in the past whether it would be better from a development point of view to enable an opt-out crash report system, where the OS reports all crashes by default unless the users disables it. If the system is offline, the reports are stored until the next time a connection is detected.
This way users who have a genuine objection to the system may disable it, whereas those who simply hit no on anything that pops up are not given the opportunity to exercise their ignorance.
I think that the main problem is that most users can't be bothered - it's faster to not send the report, especially on some of the systems that shouldn't be running Vista in the first place.
Ideally, you should only be asked once, and then WIndows could remember the setting. If the user wants to change it later, they can use Control Panel. That would make it easier, and also would result in more reports due to an inherent opt-out design.
honestly, given the fact that the program I'm running might have had sensitive data loaded (for example, a sensitive document loaded in Word), and the fact that Microsoft has probably the worst record of any software company in history for deliberately acting against the interest of their customers (such as deliberate spying, WGA, secret automatic updates, backing TPM, supporting MS-controlled document lockdown (again with TPM), abusive DRM, etc.)
What makes you think I would trust that Microsoft won't do anything with my private data when they don't even promise not to.
I report those bugs that occur when nothing remotely sensitive is involved. If I'm not sure I want the whole world seeing whatever the data was... forget it.
(Also, if I've reported the same bug more than about 5 times, I tend not to report it again when it happens every time I boot. Should I? Or does your system limit the number of reports it accepts from the same person/computer/windows serial number/IP address
If MS tested software properly in the first place and didn't try to use it's user base as a cheap source of testing after the fact, we would all be better off.
Imagine a company developing a new car and not bothering at all with the reliabilty testing, instead assuming that customers buying a car would ring up every time something didn't work (and that they wouldn't be brassed off about having to do so), and that eventually enough of a pattern would emerge that say, as 3000 cars had gone on fire then maybe the petrol lines ought to be made of something stronger. The wheels only fell off 20 or so cars, so that wouldn't be dealt with.
Fixes are available to anyone prepared to leave all the doors unlocked on their car 24x7 on the off chance the company might want to take the car off the road with no notice and fix it, perhaps just as the owner is about to go to work, or hospital.
I am sure it's a winning marketing model.
Fortunately, I think people are at last savvy enough to realise what is going on, and that is why Vista has bombed. (apart from home users where it is preloaded and there is no choice - although many are choosing to pay for XP again and load that over Vista)
MS also often confuse security with inconvenience of use. If they make something more difficult to use, it must be more secure....
Do people really think that Microsoft doesn't test the software it sends out? I think it is idiotic to assume that any software company would send out software without conducting as many tests as possible/conceivable.
People should realize that certain things are just hard to test. Furthermore, one can almost never anticipate absolutely all of the usage scenarios/hardware combinations possible, especially with something as ubiquitous as Windows.
If MS DID test software properly, a patch would not be issued that blew up an earlier patch, or broke CORE behaviour - note *CORE* (read - we test for this before release) behaviour - not some obscure use of the software in an unforeseen way.
What core functionality blew up in your case?
I think a lot of people like Microsoft and/or Windows bashing. I have never heard any Linux aficionado expound any of the many advantages of Windows.
If I may be allowed to go a bit off topic here, open source is all well-and-good but it doesn't fit the bill for most enterprise software. Nobody in the enterprise world cares how well Linux multi-tasks and how it is rock-stable and never crashes (I don't accept these claims BTW). What they care about is how quickly applications can be knocked up and start being used. For most projects, an equivalent Windows solution would be up and running before open-source based solutions were past the planning phase.
Open-source is too cost-prohibitive. Not the licensing fees. Sure, that's free in most cases. I am talking about the human-resource cost. In software development, that's usually the greatest cost and under that head, Microsoft solutions are always the cheaper option.
If you are an open-source proponent, the next time you get the urge to expound it's virtues, imagine this: you are a father or a mother. Your child is burning with fever. Do you want to go to a pharmacy and buy medicine off the shelf (that some pharmaceutical company has developed behind closed doors) or do you want to conduct your own research into various active ingredients, run clinical trials and then formulate a drug for your sick child?
I am willing to bet at least 90% will choose the former. 5% would choose the latter. And 5% would choose a drug from a company that was perceived to be cool with snazzy looking tablets :) Also, most hospitals would choose the former option.
So based on what "me" said , you have jumper to the conclusion that he is a linux aficionado.
I think that he is making a reasonable point, that patches are release without enough testing. I personally have had experience of this several times, most notably with SNA Server patches.
Please keep an open mind about these arguments - it's not always Windows v Linux.
BTW Have you ever heard a Windows aficionado expound any of the many advantages of Linux?
Play nicely boys
SteveB.
I don't believe I mentioned Linux, I was referring to Microsoft and the lack of quality control on released patches.
I can think of a couple of instances off the top of my head (there are more). MS06-015 that patched explorer and caused us all kinds of mayhem. We had users ringing up who not only couldn't print, but could not save the work they had done, had Office hanging, couldn't open documents and so on. Support was tied up for days manually changing registry settings (the MS approved workaround) to rectify the mess.
I'd say not being able to use Microsoft Office to open and save files on a Microsoft Operating system was core behaviour wouldn't you?
MS07-069 broke Internet explorer which MS argue is a core part of the OS (I would beg to differ - it is an application, but that's another topic) and again the fix was registry changes - easy when it is a single user - not as easy in an enterprise with thousands of users. I'd expect a patch to be at least tested against all standard Microsoft software before release.
As to your other point, as you say, Windows has poor multi tasking when compared to other operating systems. If an office component hangs - such as when there is network problem and outlook hangs, all the other Office programs also hang - so that the user can't open a spreadsheet or Word while waiting for Outlook to wake up. That's still not fixed as of Office 11.
What if you have a bug that isn't a crash bug? I just discovered a nasty one today.
1. If you delete a file, it goes into the (real) recycle bin.
2. The recycle bin is really just a special file
3. Offline files creates a recycle bin for a network directory when you delete a file remotely
4. Back on the local computer, deleting the offline files recycle bin (3) puts a recycle bin file (2) in the recycle bin (1).
Result: infinite loop; explorer goes all to hell (consuming all CPU and RAM) when you search from the start menu (probably in other cases too, but I spent much of the day experiencing it from this). The cause is obvious: the Recycle bin file present in the recycle bin causes an infinite loop.
Your summaries are always top-notch. Thanks for keeping us apprised. I’m reading every word here.
Human Resource Software
As a Newbie, I am always searching online for articles that can help me. Thank you loans
You got a really useful blog I have been here reading for about an hour. payday loans
What a great post, I actually found it very thought provoking, thanks
Auto Accident Attorney Houston,Texas
i love this blog! keep up good works...its to be nice if we can know each other, do you think so?
Hypnosis Education
There are certainly a lot of details like that to take into consideration. That’s a great point to bring up. I offer the thoughts above as general inspiration but clearly there are questions like the one you bring up where the most important thing will be working in honest good faith.
hyundai parts
I would like to thank you for sharing your thoughts and time into the stuff you post!! charity & fundraising
I was searching information on pedicure and your blog proved to be the best on the same. Please keep updating beauty tips in your blog.
Giveaways
Post a Comment