"Keyboard not present ... Press any key"

... As Homer Simpson once said: "Where's the 'anykey'"...

The title of this post is an actual error message spotted on, I presume, a Windows computer. Another:

"You lied to me when you told me this was a program"

or how about:

That makes 100 errors; please try again.

Or this:

Maybe you should try asking a human?

Actually, I lied above when I blamed Windows. These are all examples of Linux or Unix errors, taken from this massive list on one of my favorite web sites. Have a look.

Tags

More like this

Have you read the Unix Haters Handbook yet?

http://en.wikipedia.org/wiki/UNIX-HATERS_Handbook

(at the bottom is a link to a legal pdf copy, since the book is now out of print - i suggest you save it and open it externally rather than just click and let adobe load in the browser; it hung my firefox and i had to kill it.)

By Joe Shelby (not verified) on 13 Jul 2008 #permalink

"Keyboard not present ... Press any key" is neither a Linux or Windows error, it's a PC Bios error. It used to show up on some old Bioses, before booting any OS. And yes, it was fscking stupid. Even stupider, while it could be disabled in the bios settings, it used to be enabled by default, even on server-type hardware that usually had neither keyboard or monitor.

I got this one once:

You are not supposed to be here. Initiating self destruct to preserve integrity

By Who Cares (not verified) on 13 Jul 2008 #permalink

Is this kind of error like the apocryphal (we hope) rat in the KFC bucket? The last word of a disgruntledg employee?

More like the late-night word of an coder who's been up too late too many nights in a row. I remember testing the interface for a new app with a Y2K drop-dead deadline and finding at least one error message that I showed to the testing coordinator rather than logging as an issue. It needed to change before anything went live, but if it went through official channels, we were likely to lose a programmer.

I do like all the errors that say, "You're never supposed to see this." Kind of an object lesson in always assuming someone else will see your most obscure code.

Well, error codes can be used, and often are used, as a kind of documentation during the bugging process. Such codes can be easily left behind when the debugging crew goes home at the end of the day.

Also, there are some Linux utilities where the best part is the error codes

About any time now, we'll start seeing error codes in lolcatspeak.

"u r zo fubar"
"all ur code r b long 2 us"
"Seeling Kitteh spank!"

PERSIMIN DENIED! Who u thnk u r?

404. Y DAT NUMBER? SEELING KAT EVN HE NOT NO.

UR DATA HE R GONE. NOM...NOM...NOM...g

Oh, I didn't say it wasn't a really good error message (although, sadly, not in LOLspeak). In fact, it expressed my sentiments with a fair degree of precision. It just couldn't stay.

Greg:

Persimmon Denied? Eh, I like blood oranges better anyway.

I think I actually managed to generate that error once on an old Compaq. Anyway, if you want really fun error messages, look up the error messages from the original MPW C compiler for Macintosh. The MPW developers were some loopy bastards.

"String literal too long (I let you have 512 characters, that's 3 more than ANSI said I should)"

"...And the lord said, `lo, there shall only be case or default labels inside a switch statement'"

"a typedef name was a complete surprise to me at this point in your program"

"`Volatile' and `Register' are not miscible"

"This struct already has a perfectly good definition"

"Only one parameter per register please "

"type in (cast) must be scalar; ANSI 3.3.4; page 39, lines 10-11 (I know you don't care, I'm just trying to annoy you)"

"This array has no size, and that's bad"

"Huh ?"

Most of the time all Windows has to say is a variation on

"An Error Has Occurred".

or

"File not found"

Members of the choir can stop reading here.

I know that in both cases there is much more information available, in particular what g**d!@#@#@#@Qd file was not found.

Particularly galling is that there is a perfectly good API available to a) retrieve an error code and b) print a human readable translation of it, c) know that it is a file system (or whatever) error, and d) include some indication of what the program was attempting to achieve (e.g. "Tried to load file c:\windows\systemn32\hpdrv.dll, failed with error -7 File Not Found" - fictitious numbers for example only).

Windows of course is not alone. Because it is not a Windows problem but a programmer and policy problem.

Of maximum gallation is that for all OS's, pick any one, this kind of non-information has been observed and complained about since its inception for at least 20 years. None of them give useful info most of the time, and it is not so much an error of the OS as a shortcoming (and probably impossible schedules) of the error handling coding by the many thousands of programmers whose collective work these systems are. They handle the errors fine, the OS does not crash, but telling the poor mortals in charge too often takes the back seat.

In the very early days, whoever was operating the computer was an expert working at a very low level, from the command line, and what was being attempted was clear in all its details, so the fact that something failed was typically all the operator needed to know. Today, well, we the operators are so far removed from the underlying layers of behavior, and probably far from familiar with what those do (unless one is an author at the low level) that the low level info is pretty much useless. Reminds me of the C++ STL compiler errors - source code and what the compiler actually processes are so far removed the errors while voluminous were virtually indecipherable. And what's with most support shared libraries still sticking with 8 or so character abbreviations for their function, when the OSs today support 32 or 256 character names? Enough to write a paragraph about what the file does and where in the hierarchy it sits?

I am sure I am not exempt, for all my complaining. But I try, these days. I try.

By GrayGaffer (not verified) on 13 Jul 2008 #permalink

I've programmed error messages like "This should never happen" and "You weren't meant to see this" on a number of occasions, generally when the particular circumstances the error would trap were impossible (or at least highly unlikely), as a debugging aid during development. Sopmetimes they don't get removed or replaced with something more sensible.

My best friend is fairly heavilyy into RPGs. One of the greatest error messages I've ever seen came from one of his games:

Ratatosk scurried up Yggdrasil, the giant World Tree, to insult Viddenir, the eagle at the top of the tree, and to tell him that Nidhoegger, the root-gnawing dragon, throwed an exception.

That even beats "An unknown file was not found." Or my previous favorite:

"User malfunction."

About any time now, we'll start seeing error codes in lolcatspeak.

deyz alredy doin it. srsly.

By themadlolscientist (not verified) on 13 Jul 2008 #permalink

My personal favorite (not the exact wording/numbering but essentially what it said) is "Windows encountered Error 101 while trying to report Error 101".

"file not found" style errors seem to have trained some users to assume that all error messages will be equally unhelpful, so they simply see the word 'error' and give up.

Example; I often sit on the #ubuntu IRC channel, and just about every day someone comes on asking what to do when they encounter an error like (for example) "The program 'links' is currently not installed. You can install it by typing: sudo apt-get install links"

Ubuntu's error messages are often surprisingly helpful. But only if you bother to read them.