Steve Ballmer Authored the Windows 3.1 Ctrl-Alt-Del Screen 169
Nerval's Lobster writes According to Microsoft developer Raymond Chen, Steve Ballmer didn't like the original text that accompanied the Ctrl-Alt-Del screen in Windows 3.1, so he wrote up a new version. If you used Windows at any point in the past two decades, you can thank him for that infuriatingly passive 'This Windows application has stopped responding to the system' message, accompanied by the offer to hit Ctrl+Alt+Delete again to restart the PC (and lose all your unsaved data).
Update: 09/09 15:30 GMT by S : Changed headline and summary to reflect that Ballmer authored the Ctrl-Alt-Del screen, not the BSoD, as originally stated.
"Stuff that matters" (Score:5, Funny)
This story belongs in idle.
Re:"Stuff that matters" (Score:4, Insightful)
Seriously. There's nothing to discuss.
Ballmer wrote the message. So what?
Re:"Stuff that matters" (Score:4, Informative)
Re: (Score:2, Insightful)
Since Ballmer wrote the message, and the message was quite good, Ballmer is a developer.........
From TFA
"Okay, Steve. If you think you can do a better job, then go for it." Unlike some other executive, Steve took up the challenge, and a few days later, he emailed what he thought the Ctrl+Alt+Del screen should say.
The text he came up with was actually quite good, and it went into the product pretty much word for word.
Re:"Stuff that matters" (Score:5, Funny)
Seriously. There's nothing to discuss.
Ballmer wrote the message. So what?
Bet you wouldn't say that if Bennet had posted this story. But the again it would have been a philosophical piece about how while he likes the color blue, its not his favorite color blue, and how he wished that all error display screens should be *his* favorite blue color, and how dare the manufacturers of all the different OS's not consult him and get *his* opinion on what makes for a really nice blue color, even though each of those OS manufactures have their own ideas as to how things should be done and they have most likely done their own research into colors, but anyway that should all be scrapped and re-implemented Bennett's way (at their own expense of course) and while their at it could they also make it so every program works exactly the same on every different combination of computer and OS as it's a major hassle having to learn how to do things differently whenever you site down at an unfamiliar computer, but then again why should computers be unfamiliar in the first place, maybe it would be better if they all had a dedicated "Bennet" login so that he would just be able to sit down at any computer and just use it the way he wanted to, in fact what would be even better if all that research into melding telepathy and machines was finally completed so that he wouldn't even have to sit down at a computer as it would simply recognize him from a distance and would then fire up its 3D holographic welcome display (which BTW is fully detailed 3D model of Bennett himself - on a pedestal) so that he can instantly get down to his .. Oh look! Squirrel!!
Re:"Stuff that matters" (Score:4, Insightful)
Bet you wouldn't say that if Bennet had posted this story. But the again it would have been a philosophical piece about how while he likes the color blue, its not his favorite color blue, and how he wished that all error display screens should be *his* favorite blue color...
Awesome. Thanks for that. It almost makes having to suffer through Bennet's use of slashdot as his personal blog worth it, just to see it satirized like this. :)
Re: (Score:3)
LOL ... that's it ... Bennet is essentially Gilderoy Lockhart!!
You, sir, are brilliant!
Re: (Score:2)
Excellent sir, although you still haven't beaten Jonathan Coe.
Re: (Score:2)
Re:"Stuff that matters" (Score:4, Insightful)
Oh, how I wish I have mod points right now. The article itself and this article are both worthless.
Re: (Score:2)
Even as a long term Linux user who hasn't done much windows related since XP SP2 was news, I do still read and actually enjoy Raymond Chen's blog, and even bought his book.
The dice article is pretty pointless though as it just mirrors Raymond Chen's post.
Re: (Score:3)
Re: (Score:3)
It's a minor piece of trivia that was certainly not worth posting on slashdot in lieu of another Raymond post, but it's something I would've enjoyed reading when going through his blog. Check it out, there's tons of interesting technical and historical information there.
Re: (Score:2)
Agree.
Even as a long time Linux user, I still enjoy his blog (and even bought a copy of his book). It's interesting, easily digestible, and well presented tidbits from someone who actually works with the stuff.
Probably doesn't belong on slashdot any more than his other posts though as you said, however I would have been ok with it if they'd linked directly to his post rather than linking to the dice article which is pretty much just a mirror of it.
Re: (Score:3)
So now we know who to send the counseling bills to...
Re: (Score:2)
Considering the way the old BSOD messages were handled in 3.x? The current implementation that he built was useful, exceptionally useful at that. Not only did you get a main dump, but it also partially dumped directly to the screen and let you troubleshoot cleanly from that.
Amiga (Score:5, Funny)
Nothing will ever top "Guru Meditation" :)
Re:Amiga (Score:5, Insightful)
If a system can display an error message, it is not messed up enough.
Re:Amiga (Score:5, Interesting)
If a system can display an error message, it is not messed up enough.
Yeah. A really good Amiga crash would randomly poke the graphics and sound chips and the machine would look and sound as if it were about to explode.
Re: (Score:3, Funny)
There is an in-between. I once saw the error message:
ERROR: There is not enough memory to display the error m
Re: (Score:2)
Re: (Score:2)
I'd like to have the *option* to continue to save my work even if there was a chance of data corruption. For example, take the common NT blue screen IRQL_NOT_LESS_OR_EQUAL. That fact that my buggy network driver tried to access paged memory in the wrong sequence is miles away from catastrophic. And it certainly doesn't take priority over something I've be
Re: (Score:2)
I'd like to have the *option* to continue to save my work even if there was a chance of data corruption. For example, take the common NT blue screen IRQL_NOT_LESS_OR_EQUAL. That fact that my buggy network driver tried to access paged memory in the wrong sequence is miles away from catastrophic.
Believe or not, but such situation is actually catastrophic. A network driver might feel like a small component of your system, but as it happens to be tightly integrated into kernel space and when it does something illegal, there's a real risk of data corruption or other system misbehavior.
Actually, with NT6, display drivers were moved into userspace, so if a display driver crashes, one just gets a blank screen for a couple of seconds, and after that a system tray popup informing about the driver restart.
Re: (Score:2)
You're comparing apples and oranges as far as the technical details. I'm saying Win 3.x let me continue when it saw problems, and NT could also do that.
Not really. The kind of situations where Windows 3.x let you try to continue, Windows NT just handles transparently. In Windows 3.x, with cooperative multitasking, a single application can refuse to relinquish the CPU. If this happens, you have three choices (outlined by the dialog box):
In a system with protec
Re: (Score:2)
This error message appears at /. as well. :)
Re: (Score:2)
That's good if you want to really weird people out. I prefer the pure panic-inducing power of lp0 on Fire!
Re: (Score:2)
Oh, I don't know ... Slashdot's "varnish cache server" is up there. :-P
Guru Meditation (Score:2, Interesting)
The term "Guru Meditation Error" originated as an in-house joke in Amiga's early days. The company had a product called the Joyboard, a game controller much like a joystick but operated by one's feet, similar to the modern-day Wii Balance Board. Early in the development of the Amiga computer operating system, the company's developers became so frustrated with the system's frequent crashes that, as a relaxation technique, a game developed where a person would sit cross-legged on the joyboard, resembling an I
old message (Score:3)
Re: (Score:2)
Re:old message (Score:5, Funny)
So this [techsupportforum.com] is something you'd prefer?
Re: (Score:2)
So this [techsupportforum.com] is something you'd prefer?
That looks like something the BOFH would do. Yes, No, close button - they would all be devastating.
Re: (Score:2)
More like the BP(rogrammer)FH. I'd put that dialogue at the closing window, leaving the user wondering whether the question was "do you want to save your work before closing?" or "all your work will be lost if you close without saving first. Proceed?"
Re: (Score:2)
I prefer the Windows 3.1 BSoD [deviantart.net]
Well, that's interesting, because TFA [msdn.com] precisely talks about Ballmer designing the Windows 3.1 BSOD. The BSOD in the article looks different than your screenshot.
Most likely there are multiple BSODs in Windows 3.1, and Ballmer designed one of them.
I miss the BSOD (Score:5, Insightful)
I'd rather get some cryptic information about stop codes or an error message than a condescending sad face accompanied by a reboot request. At least I can look up the code and get a ballpark idea what the issue is without firing up windbg.
Re: (Score:2)
yeah at least it's something.
something is better than nothing.
not much better, but ...
Re: (Score:2)
I'd rather get some cryptic information about stop codes or an error message than a condescending sad face accompanied by a reboot request. At least I can look up the code and get a ballpark idea what the issue is without firing up windbg.
I like 'An unexpected error occurred..."
We need more expected errors. These unexpected ones are clouding the issue...
Re: (Score:2)
Yes, old BSOD was great tool for the IT crowd. The Windows 8.1 gives now just a page with awfully large text saying something like "sorry, something was wrong, we're rebooting." Good luck with finding the reason for the crash and most importantly, fixing the problem. This dumbing down of software is just disgusting.
Check out Minidump files. There's a good amount of information stored about the cause of a BSOD.
Even the Visual Studio 2013 is designed for the tablet users, I wonder why they still support programming at all, as it is is assumed to be too hard for their customers.
Visual Studio 2013 is a normal desktop application just like the previous versions of Visual Studio.
Re: (Score:3, Informative)
That only works if you're able get to the machine to boot again :-(
Windows 8 (Score:4, Insightful)
Re: (Score:2)
How the frog do you trigger a BSOD in Windows 8? Seriously, I've had driver failures but those triggered restart of the driver and no BSOD...
Re: (Score:3)
People are resourceful.
Re:Windows 8 (Score:4, Funny)
Re:Windows 8 (Score:5, Informative)
How the frog do you trigger a BSOD in Windows 8?
Apparently by installing updates [slashdot.org].
Re: (Score:2)
It's very stable now. I think I've had to reboot the system once in the last month.
Re: (Score:3)
What's the rest of your hardware like? GTX 660 doesn't need anywhere close to 1 kW, I'm doing fine with a 450 W PSU (also powering an i5-2500 CPU and one hard drive). Seems more likely that your previous PSU was just a very low quality one.
Greetings from the someone-might-be-wrong-on-the-internet dept. :-)
Re: (Score:3)
A friend of mine told me recently that he got one by putting his laptop on hibernation while the peripherals were still connected. He disconnected them while in hibernation, and then woke the system up again.
Re: (Score:2)
Ouch! Now that is a design failure...
You'd think - (Score:2)
You'd think he'd have someone to do that for him!
Anthill Inside (Score:5, Funny)
+++OUT OF CHEESE ERROR+++
Re: (Score:2)
+++OUT OF CHEESE ERROR+++
I'm just glad that Balmer didn't write:
This Windows application has stopped responding to the ayiiiiiiiiiiEEEEEEEEEE!
https://www.youtube.com/watch?... [youtube.com]
That explains a lot (Score:2)
'RESTART! RESTART! RESTART!' would have been a lot better. Clear instructions are useful. Screenfulls of BS just confuse people. All they can do is restart anyway.
Re: (Score:3)
Screenfulls of BS just confuse people.
It's not just BS to everyone though. And even without understanding what it was telling me by googling the stop codes I've been able to fix things based on good search results, especially for very common problems like driver errors. As another poster mentioned it's sometimes possible and a hundred times easier to search for a stop code and get a fix for a problem than it is to fire up WinDBG.
It is the frequency not message (Score:3)
Warning: Do Not Turn off this Jumbotron! (Score:3, Funny)
Applying critical patch 42 of 13,699,364...
I KNEW IT! (Score:2)
I knew it I knew it I knew it!
Hexidecimal (Score:4, Insightful)
Did he also decide to produce the Hex output that is entirely useless and without merit? I understand that's for debugging purposes, but who decided that was a good idea to leave in for a consumer-level OS? Seriously.
Re: (Score:2)
You know, these are the same people who have put the meaningless error messages like "something bad happened, if this problem persists contact your administrator".
Gee, thanks, it's my fscking machine, I'm the admin ... how about you tell me something meaningful about the issue so I can try to find it?
Microsoft seems to be eternally stuck between dumbing something down so far as to make debugging impossible, and spitting out gibberish messages that you need a wizard level guru to decipher.
And, more on topic
Re: (Score:2)
Actually, the message in that OS version is fairly acceptable for its purpose and context. It identifies the nature of the problem using understandable words, offers a course of action for recovering from it, and explains the potential outcomes of following it. That's pretty much what the user needs to know.
If you want to debug it you should use the logs anyway, so the message for the end user is better written in plain English.
Re: (Score:2)
Why is it not a good idea to leave it there?
Re: (Score:2)
Did he also decide to produce the Hex output that is entirely useless and without merit?
If you read the blog entry, this is talking about Windows 3.1's BSOD. A screen I honestly did not know existed, although Windows 3.1 is so old that I'd have been a kid, so maybe it popped up all the time if you used computers daily back then. I have no idea.
Windows only picked up preemptive multitasking in NT and later 95, so Windows 3.1 was cooperatively multitasked. Apparently if the running program didn't respond to incoming messages quickly enough (presumably a check in an interrupt handler?) a blue scr
Re: (Score:2)
LOL, then I can only conclude you never actually used Windows 3.1.
Which is exactly what he said:
A screen I honestly did not know existed, although Windows 3.1 is so old that I'd have been a kid, so maybe it popped up all the time if you used computers daily back then. I have no idea.
Re: (Score:2)
Did he also decide to produce the Hex output that is entirely useless and without merit? I understand that's for debugging purposes, but who decided that was a good idea to leave in for a consumer-level OS? Seriously.
Ah yes. Everyone should have to set up a second machine, connect it to the other via a serial cable (having remembered to enable serial port debugging on the host prior to the crash), and then fire up their kernel debugger just to get the bugcheck code.
Putting a numeric error code (which usually comes with the symbolic name as well) on a consumer-facing fatal error is absolutely the correct thing to do. Once you've reached the kernel panic failure point there's not much most consumers can do anyway, so pr
Re: (Score:2)
???? Well, I guess you are proud to be an uneducated redneck. Just because it is useless to *you*, doesn't mean it is useless to everybody. To some of us, it is essential that the exception code be easily available. If it doesn't appear on the last screen the machine can put up before coming to a complete halt, where would you suggest it go? To a log file, when the file system might not be working? *sheesh*. Really, I'd like to hear where else you think it could be recorded in a manner that is both 1
Re: (Score:2)
Did he also decide to produce the Hex output that is entirely useless and without merit? I understand that's for debugging purposes, but who decided that was a good idea to leave in for a consumer-level OS? Seriously.
How is it a bad idea to present the information in a consumer-level OS? What would be better, not showing information?
Re: (Score:2)
> the Hex output that is entirely useless
useless??
If you don't know how to search for the first error code [google.com] you probably shouldn't be using Windows ...
Re: (Score:3)
Did he also decide to produce the Hex output that is entirely useless and without merit? I understand that's for debugging purposes, but who decided that was a good idea to leave in for a consumer-level OS? Seriously.
WTF? of all the idiotic things they have done, leaving the debug information available in the consumer-level OS was one of the BEST ideas. It gives even the most clueless user a chance to google it, or read the screen to someone that understands it, or in the case of my mother send a photo of it so someone can look up what went wrong.
Re: (Score:2)
He was probably in envy of the Amiga's Guru Meditation [wikipedia.org].
Never seen it (Score:2)
I must be unique then in that I have used Windows for 15 years and I've never seen that particular blue screen before. Had to google it after reading the article, and still can't find any other mention of it. In what version of Windows was it used?
Re: (Score:2)
Re: (Score:2)
No, I think just missed that one. Good for me, it would seem.
Re: (Score:3)
The screen allows the user to kill the offending application, allowing any other applications to continue to run (that is as long as the hung application hadn't corrupted the contents of RAM in so
Re: (Score:2)
Thank you. The article makes it sound as if it was used in versions after Windows 3.1, and specifically not in 3.1. I was lucky enough to not have a Windows PC in the old 3.1 days, so that would explain it.
Re: (Score:3)
Windows 3.1 had protected memory, apps attempting to access memory the didn't own accounted for 90% of blue screens. Of those 90% were trying to access 0000:0000.
Also Windows 3.1 multitasking was more complicated then that. It had preemptive multitasking between DOS shells and Windows. But windows itself used MacOS style (cooperative) multitasking.
Also note: Windows 3.1 did way too much in kernel mode. So any driver could corrupt memory, but not apps in general.
Prevent damage to your computer (Score:3)
Personally, I like the message that says "Windows has been shut down to prevent damage to your computer." I wonder who came up with that one.
Re: (Score:2)
Re: (Score:2)
Re: (Score:3)
We're sorry, Someone threw a chair and smashed your Windows. Please press CTRL+ALT+DEL to reboot.
Infuriatingly active (Score:2)
The BSOD message can't be more infuriating than what Macs say when they reboot after a kernel panic: "You shut down your computer because of a problem." It always makes me want to shout "YOU shut YOURSELF down due to a problem YOU caused!"
"Keyboard Not Found" (Score:2)
"Keyboard Not Found"
"Please press F1 or all your work will be lost."
Re: (Score:2)
"Keyboard Not Found"
"Please press F1 or all your work will be lost."
"Keyboard not found"
"Press F1 to continue"
Was my favourite error message. I had it constantly on booting my old 286 whether the keyboard was there or not and you would just press F1 to go on.
Flying chairs (Score:3)
I think Windows 8, that perverse boot sector virus, ought to have updated the BSoD to show a video of Steve Ballmer throwing a chair across a room. No doubt he's done that a few times in his office as the BSoD popped up.
Run, Ballmer, run (Score:2)
Re: (Score:2)
Ballmer always struck me as a Gump-like character who accumulated wealth and thus influence through no talent of his own. He stumbled into Microsoft with no more to offer the world than a guy off the street who pulls a slot machine arm and wins a billion dollar jackpot. At least Forrest was likable.
Yeah, thats exactly what he looks like... and it's more favourable than the truth.
Balmer is a salesman, its the same impression some used car salesmen like to give. Truth is, if he wasn't a capable salesman he wouldn't have lasted long at Microsoft (which he didn't when things went south).
What he should have written is ... (Score:2)
Re: (Score:2)
No, this is Steve Ballmer, not a Dalek. It would have been, "FUCKING KILL! FUCKING KILL! FUCKING KILL! "(TM)
There once was a CEO named Steve,
Who threw chairs at employees for reprieve,
He shot lasers from his eyes
To bury other guys,
And he'll Fucking Kill you if you don't believe.
The BSOD message didn't go far enough (Score:2)
This would be noteworthy if a chair came flying out of the monitor upon BSOD.
Anybody else remember UAE's vs. GPF's? (Score:2)
UAE's (Unrecoverable Application Errors) were the bane of Windows 3.1. When Windows 3.11 was released, MS proudly announced that UAE's were no more!
How did they pull off this programming miracle?
By renaming the error to "General Protection Fault".
And they vanquished THOSE in Windows 95 by calling it an "Illegal Operation"
After that, it was just [Program] Has an Error (using various wording, depending on version.
Developers, Developers, Developers, (Score:2)
It doesn't make sense with the last "Developers" missing.
Re: (Score:2)
"Infuriatingly passive"? (Score:2)
The data is gone...let it go (Score:2)
offer to hit Ctrl+Alt+Delete to restart the PC (and lose all your unsaved data).
If you have a BSOD, your unsaved data is already gone. How you move on from there (Ctrl-Alt-Del, or the power switch, pulling the plug, sledgehammer, etc.) is simply a matter of preference.
Reminds me of... (Score:2)
There was a guy who dreamt about being a great poet who could truly touch people's feelings. Unfortunately he lacked talent for coming up with rhyme, analogy, insight and so on.
But he found employment at Microsoft, where he finally made his dream come true as an error message writer, with classics such as "BSoD", "Press 'OK' to continue. [hongkiat.com]", "Catastrophic Failure.", "Abort, Retry, Fail? [wikipedia.org]" and many others that have touched a nerve on each of us over the years.
how about "Stack Crap, We Goofed Again..." (Score:2)
that screenshot was the Microsoft company song.
Re: (Score:3)
I'd rather know who wrote that stupid message that implied the user was responsible for Windows not being shut down properly when it was the festering pile of crap itself that fell over so I could drive all the way across the country and give them a slap.
I am sure many here will have a brief seething relapse when they see these words:
Because Windows was not properly shut down, one or more of your disk drives may have errors on it. To avoid seeing this message again, always shut down your computer by selecting Shut Down from the Start menu.
Re: (Score:2)
... always shut down your computer by selecting Shut Down from the Start menu.
You've got to admit, the advice is sound. "I see you tried to run a program. Sheesh. Remember ALWAYS shut down your computer...".
Re:Never liked the 'D' part of BSoD (Score:5, Informative)
BSOD happens when the kernel detects memory corruption. With a hybrid monolithic kernel like Windows that means all bets are off and continuing could very well case damage more damage.
Even if the memory corruption happens in an USB driver, it can overwrite critical kernel memory.
Incidentally, you *do* get more information. The kernel will initiate a kernel dump which can be investigated later.
Re: (Score:2)
One of the big advantages of a microkernel is that it lets you know what fucking driver just tried to write all over memory it didn't own.
Re: (Score:2)
What's worse than a forced reboot? A reboot that should have happened but was ignored, wiping out not only the last hours work, but your entire disk.
>99% of users would have absolutely no idea what choice to take. And no matter which they chose, they probably will wind up losing something. So why make it appear that it is the users fault if they lost data (by making the 'wrong' choice)?
Yeah, just suspend that thread that is in the USB stack. What could possibly go wrong? It's not like it could inter
Re: (Score:2)
I would genuinely like to know how you, as an 'experienced kernel developer', know, using only the information you said (backtrace, etc), that the reason the USB stack is trying to write to NULL is NOT because something stomped all over storage, possibly including the storage stack.