Given all the weird messages in the machine, what about the error messages that get printed from time to time? - Unix's: "Not a TTY" - RSTS/E's: "?Program lost, sorry" - VMS gets a bit wordy with theirs, except for the gem in XDELTA (the kernel level debugger): "EH?" - Unix's: "Intruder Alert!" - Pick's: "This error cannot occur" - There's a language for the not-quite-yet-ancient Atari 8-bit called Action, which has a rather nice error message, error code 11, "Unknown error. You have somehow impared the Action! system error routines, so it can't tell you which error you have made.". Of course, virtually every error I got was error 11, except when it decided to give file not found errors for errors in other parts of the program. - Chrysalis window manager for the BBN Butterfly. It did something like this: "There seems to be an internal inconsistancy in wm. Things are only going to get worse from here. Sorry." - BSD 4.3 talk has given me a few times: "You don't exist. Go away." - One Data General 16-bit operating system (MP/AOS), which was cancelled something like 6 months after it was announced, had an error message: "You should live so long" which was the error message that was supposed to be generated when the time field overflowed. - On VMS, a program returns an integer, with the upper bits being an error code, and the bottom bits being flags on how to interpret the error code. A flag of zero means print an error message. Whomever implemented the C library exit, made it return that number directly, without realizing that the majority of extant C programs return 0 to mean no error. Thus when you ran the program, DCL printed something like: "ERROR: PROGRAM TERMINATED NORMALLY" Because of this brain-dead implementation, and the hard-headed DEC representative to the ANSI C committee at the time, is why the ANSI C standard had to invent 'EXIT_FAILURE' and 'EXIT_SUCCESS'. - Error message in the BIOS for a DTK 286 clone. The PC was rather flaky, and occasionally when booted would display: "Segment Doesn't Found" - While was working on a DTK 286, forgot to plug in the keyboard, the error: "Keyboard Error. Press to continue" - Another DTK 286 BIOS response: "FDC Error. User Check!" - A PRIMOS error message when the return stack gets corrupted: "FATAL ERROR ON CRAWLOUT" - Unix fsck's: "See a Guru." - ed's error message: "?" - RSTS/E: "?Protection violation", used whenever the system couldn't figure out what else to say. - At one site there were DEC RM03 disk drives with the easily-bumped front panel buttons. If one accidentally took the system disk off-line, the operating system detected this, and tried to put a message on the system console. Of course, the message file was on the off-line disk, so a hard-coded message "ERROR TEXT LOOKUP FAILURE" would appear; then it would try to tell you it couldn't look up the error, and try to read that file again. Instant loop. - The "athena" fortran compiler, bulit at JPL for the Univac 1108 running the JPLized version of Exec-8: "Help I'm confused at line 1023" In the above example, the module to be compiled was only 800 lines long. - The Philips P855 computer indeed did have an error message: "BUY MORE CORE" (It is a minicomputer from the early seventies, with 32K words of 16 bits ring core memory, built completely with TTL chips. Access time around 7 us) - RSTS: "Catastrophic error: program lost - sorry" It meant "I know your program is in memory, but I can't remember where I put it." - A russian MIR-1 (table-sized transistor machine with 4 Kb core memory, punch tape, typewriter and built-in interpreter of ALGOL-like language). This funny machine had ONLY ONE error message: "NO MEM". All other errors were indicated on lamps above tubler registers. - TRS-80 Model I Level I BASIC, which had only three error messages: What? (syntax error) How? (divide by zero, etc.) Sorry! (out of memory, etc.) - RSTS/E does something similar. It's response is: "What?" - CP/M is marginally more helpful. It repeats the entry that puzzled it, followed by a question mark: (example follows) PROMPT> A: ENTER> A:foo RESPONSE> foo? PROMPT> A: - The BBC B micro's BASIC had a (very good) renumber command. When you got the parameters wrong the response was: "Silly" - Occasionally here at work, we get an error cropping up on Novell network systems: "Error number 4 trying to report error number 4." - The RTE IVB Real-Time Executive operating system for HP 1000 minicomputers. Almost any keyboard input that was not one of the (arcane, cryptic, quirky, illogically-chosen) commands recognized by the OS got you: "FMGR 16" - Our VAX 11/750 here panics with: "Well, you ran into something and the game is over. panic: ... " - The ancient Bell Labs BESYS-7 Fortran II compiler (ca. 1966), for the even more ancient IBM 7094, would, if sufficiently provoked, tell you: "PLEASE RUN THIS PROGRAM ON SOME OTHER MACHINE" - Apparently in one of the early Tandy main frames, the engineers included an error message for when the machine was physically bumped. Since this condition is so rare while the machine is running, it went un-noticed until the machine was out on the market. The first irate and confused costumer received dis-belief from technical support when they reported the message: "Beam me up Scotty, I'm sucking mud!" - DECsystem-10, BLISS-10 compiler would give up with the error message: "punt!" - DECsystem-10, SAIL compiler would (in serious difficulty) respond: "Dry rot" - The very first PDP-11 diagnostics (circa 1970) that I have on DECtape had a single message: "ERROR SOMEWHERE" - DOS-11, the predecessor to RT-11, had error messages that consisted of a letter and three digits followed by a six-digit octal number. If there was some problem (ANY problem) opening a file, you got a message like "F024 000000" - Taken from the 'he' helpfile for sccs help: "argument too long Dost thou jest? Wilst thou mock HELP?? Please limit your blitherings in arguments to less than fifty (50) characters." - System V the message "Bad magic" - The Algol compiler used at Case Institute of Technology, after finding 25 errors in the source (e.g., like you spelled BEGIN as BEGNI), would print: "At this point, we suggest you try re-reading the manual." - I've also heard tell of a Fortran compiler that would print out: "Programming by Monte Carlo methods is frowned upon." if you had too many errors. - The PDP-6, the prototype for the PDP-10, circa 1964; in his talk for DECUS, Peter J Hurley describes the situation thus: Since there was no swapping, everything had to fit in memory. Programs were shuffled to close up the unused space. But because the BLT [BLock Transfer] instruction was used, shuffling could only move programs toward lower addresses (5K monitor, remember?). Thus, after one's compile had been running for twenty minutes, growing towards someone else's job, one might get an error message "Core available, but not to you" and could try to persuade the other person to kill his job or, more likely, would have to kill one's own job, type CORE 0 to cause shuffling, and start again. Actually, it wasn't quite that bad, since one could, if the program was reasonably well-behaved, CTRL/C it and SAVE the image before forcing a core shuffle, and then restart the saved image afterwards. - PIP for the PDP6, was invented by "Dit" Morse as a demonstration of device independence. Its original name was ATLATL, which stood for "Anything, Lord, to Anything, Lord". This was appropriate, as it took a certain amount of prayer to get anything to move between media. In those days, when TTY's had backarrows (instead of underscores) that key was used instead of the equals sign in PIP. This, it was felt, was sufficiently obvious that anyone who, for example, tried to read from the line printer got a message like: "You gnerd, device LPT: can't do input!" - Microsoft Word for Windows, error number 127: "Wrong number of dimensions." - In a diagnostic program on the Apollo DN series, circa 1985, typing "?" elicited a response of: "You must be from Prime. Type [control-something-or-other] for help." - On the first Atari 400 computer, there was an Atari BASIC error listed in the manual as: Error 0 - "Power not on" - The UCSD Pascal system as a developmet system on apple ]['s: "return from uncalled subroutine" - awk: "bailing out near line 1" - Inmos Occam compiler: "Expected RESULT, found RESULT" - A Xerox 1108 (Lisp machine). The front panel gave all the error codes, the messages where in a book. The lab favorite: "Burdock tried to access the EtherKludge" - On our system (SunOS 4.0.3) talk says: "Who are you? You have no entry in /etc/utmp! Aborting..." "This machine doesn't exist. Boy, am I confused!" and "Target machine is too confused to talk to us" - compiler error: (from MPW C): "Too many erors on one line (make fewer)" - Error from rn when trying to cancel an article that does not exist: "Cancelling null articles is your idea of fun? :-)" - From TRSDOS 2.2's BACKUP command: when you insert the wrong disk during a swap it responds "WRONG DISK, DUMMY!" - I remember my old TRS-80 Color Computer. It only had 2-letter abbrevisations for all errors. The one for "file not open" when you tried to read/write a file was: ?NO ERROR also for "Subscript out of range": ?BS ERROR and "Out of Data": ?OD ERROR - The version of DISSPLA that we used to have on the VAX here reportedly gave the error message: "Out of memory. Choke, gasp, wheeze" - The Algol compiler for the ICL 1900 used very occasionally collapse with the message: "The impossible has happened!" - From Version 7 UNIX: "Values of B will give rise to dom." I don't know what it meant but it had something to do with removing a directory with a name beginning with '.' - My favorite is the PL/1 compiler message: "COMPILER UNABLE TO ABORT" This is the message when the compiler has attempted to abort the compilation five times, and has failed each time. - This is what the Univac 2200's os1100 os produces, when I fill my program file beyond its capacity: I/O TYPE 01 CODE 22 CONT 12 REENT ADR: 015245 BDI: 403034 PACKET ADR 045301 AN ATTEMPT WAS MADE TO WRITE BEYOND THE MAXIMUM ASSIGNED SPACE FOR A MASS STORAGE FILE. AN ATTEMPT WAS MADE TO EXPAND A MASS STORAGE FILE BEYOND THE MAXIMUM ASSIGNED SPACE. A READ FUNCTION FOR A MASS STORAGE FILE SPECIFIED AN ADDRESS (WORD 5 OF THE I/O PACKET) THAT IS BEYOND THE MAXIMUM ASSIGNED SPACE. A READ OR WRITE FUNCTION FOR A WORD-ADDRESSABLE MASS STORAGE FILE SPECIFIED A MASS STORAGE ADDRESS (WORD 5 OF THE I/O PACKET) AND A TOTAL DATA COUNT. WHEN THE MASS STORAGE ADDRESS IS ADDED TO THE TOTAL DATA COUNT, THE RESULTING ENDING MASS STORAGE ADDRESS IS GREATER THAN 2*/35-1. A READ OR WRITE FUNCTION FOR A SECTOR-FORMATTED MASS STORAGE FILE SPECIFIED A MASS STORAGE ADDRESS (WORD 5 OF THE I/O PACKET) THAT IS GREATER THAN 2*/30-1. ADI ONLY: REFERENCE ATTEMPTED BEYOND THE ASSIGNED FILE WHEN THE FILE IS CONFIGURED AS A FH-432 OR FH-1782 DRUM. (Filename: FEK*ONT) END MAP. ERRORS: 1 TIME: 33.978 STORAGE: 054203/014304/035416/3/0220776 ERR$ TYPE 03 CODE 00 CONT 12 REENT ADR: 045147 BDI: 000015 USER EXECUTED ER ERR$. - If you run "strings" on the executable of gawk (the MS/DOS version, at least), you'll see a line with this message: "initstate: not enough state (%d bytes) with which to do jack; ignored." - When I was at Purdue, the IE department had a DG Nova system that would respond to attempts to run object programs formatted for a DG Eclipse system with the message: "YOU CAN'T DO THAT!" - Here are some that I found in reading the string-pool from Knuth's TeX: [Note that I included these from the actual file, so the one with 'can fix can fix' below is what's actually there!] (That makes 100 errors; please try again.) You can now delete more, or insert, or whatever. Sorry, I don't know how to help in this situation. Maybe you should try asking a human? Sorry, I already gave what help I could... An error might have occurred before I noticed any problems. ``If all else fails, read the instructions.'' This can't happen. I'm broken. Please show this to someone who can fix can fix I can't go on meeting you like this. One of your faux pas seems to have wounded me deeply... in fact, I'm barely conscious. Please fix it and try again. Interruption You rang? IMPOSSIBLE. BAD. A funny symbol that I can't read has just been input. Continue, and I'll forget that it ever happened. I suspect you've forgotten a `}', causing me to apply this control sequence to too much text. How can we recover? My plan is to forget the whole thing and hope for the best. Dimensions can be in units of em, ex, in, pt, pc, cm, mm, dd, cc, bp, or sp; but yours is a new one! I'll assume that you meant to say pt, for printer's points. - If things go amiss in Interactive Data Language, as they frequently do, you get the following just before the core dumps. "Something Rotten in Denmark, Interp Stack Not Aligned" - The most common error message we got from a modula II compiler that I used at an other company was: "Unexpected ';', expecting ';'" - The early versions of TeX had this classic, which I believe the people at Stanford even had printed on T-shirts: "You can't do that in horizontal mode." - My favorite was on the IBM 7094. Occasionally, the COBOL compiler would die with just: "COMPILER THWARTED". - Doing a strings on our version of lint yields this error message; I have no idea how to get it to spit out... stack overflow, maybe? "lint's little mind is blown." - When a graph plotting program on the Amiga runs out of memory it give a red box with: "Hot Damn! You need more ram!" - I noticed this one while hacking graphics routines into the Z80 portion of Radio Shack's (Microsoft's) TRS-80 Model 16 M68000 Xenix. The Z80 handled all the of the IO in the machine and somewhere imbedded in the code was the message: "Shut her down, Scotty, she's sucking mud again!" - On a modified version of the IBFTC Fortran compiler for the IBM 7040 the following error was displayed if inconsistent data were detected: "ERROR 1164 HOW IN THE HELL DID YOU GET HERE" - From our Harris VOS system. JOBCNTRL ER 2211 : IT'S NOT NICE TO FOOL POP! >he 2211 YOU JUST TRIED TO FAKE-OUT MOTHER NATURE, AND SHE CAUGHT YOU! SUPER- VULCAN NOW HAS YOUR NAME ON HIS ENEMY LIST, AND YOU CAN BE CERTAIN THAT FUTURE ATTEMPTS TO RESOURCE LFN 0,3,OR 6 WILL RESULT IN YOUR BEING ABORTED, SPINDLED, MANGLED, FOLDED, PUNCHED, DELETED, AND DEALLOCATED. - My OS (QNX) has a command called TSK (short for TASK) which allows you to view information to do with tasks (code size, id's, son, dad, etc) when I first saw it a friend of mine showed me the list of commands, of which one is tsk tsk, I tried it, and it came up with the following message: "Tsk tsk? Have I been a bad computer?" - In an earlier version of BSD (4.1?) if you did [I think it was a] "who" and you were the only one on the system, it would print something like: "Are you lonely?" - My favorite real error message, found in the User's Manual for the Atari 800 computer, which produced only numbers for errors, so you had to look up the translation in the manual: "ERROR 0: POWER NOT ON" - Shortly after I started work at the Stephen F. Austin State University computer center as a support person, we had a coed come in with a very strange problem. She had been trying to do her FORTRAN homework, and had run across a bug in the FORTRAN compiler (ANSF on Honeywell CP-V). On her printout was some diagnostic information, followed by the words: "Break Rob's knuckles." - My favorites were from the older Apollo OS's, two of the systems errors were (I believe I remember them correctly): "Can't find wicked faraway objects." - I worked at a timesharing shop many years ago. One machine ran an operating system called Micos. When something went wrong, an error number would be printed on the console. At the end of the manual for this system was an appendix that described the error numbers in more detail. My favorite (which was printed when the disk drive crapped out) was: Error 4: this is definitely a hardware or software error. - While performing some maintenance on the swap disk on a DecStation 3100, the following error message came up: ERROR: Not a Typewriter. Seems to me that Dec has a clear grasp of the obvious. - When I worked at Alphatype, we used a home-grown Z-80 assembler. If your assembly had any errors, you'd see a message at the end of the listing: 0004 ERRORS GOOD JOB, SPUD!! - From the Data General S200 Fortran error code list - "ERROR 155 - You can't do that." - When Multics switched over to its New File System, a check was added to ensure that only one disk volume "wanted" to be the root of the file system. It was considered pretty unlikely there'd ever be two, but why not check, just in case... ? I believe it was General Motors who called up the system support group in Phoenix, who then had to call the Cambridge (Massachusetts) development site, to find out why their system wouldn't boot one day. As I understand it, the GM site analyst called in, saying "We got this weird console message. It's in Latin." The translation offered by the author of NFS, Bernie S. Greenberg, solved the problem: "Unto the root is born a brother." - Geoworks Ensemble had an error message that said "I'm so confused." Apparently, an engineer really was that confused at that point as to what had happened with his code. He later forgot to remove the message. Regrettably, it was removed after the first bug release. - Before Turbo Pascal came out, I did some fussing with JRT Pascal. As it compiled, it would display each line of source on the console. On some programs, it would start filling the screen with garbage and then, at the end, say "no errors detected". - "SNA RO" Which is the Apple SYNTAX ERROR after you've hit RESET and switched back to 40 columns from 80. - There was an old application (PeachText) on the Z100 that would, if you fiddled around, jumping from program to program and doing things (it was a "works" like program), sometimes exit to the prompt suddenly. Just before it did so it would print, wherever the cursor happened to be at the time, "Not an error." - I don't think anyones mentioned the most insidious dumb error message of all: kernel: PANIC:DOUBLE PANIC Actually any of the "kernel:PANIC" messages tends to scare the willys out of most non-techie users requiring massive amounts of phone support to calm them down. - When I worked at CMU, we were recompiling local utilities after an OS upgrade on our Gould mini. For one particularly "whizzy" piece of software, the C compiler produced a message along the lines of: "This shouldn't happen; Call Bob" - The operating system in Burroughs Large Systems (currently Unisys A-Series) used to be written in a variant of Algol called Espol (Executive Systems Programming and Operations Language). This language was just for internal use by the company. My favorite syntax error from that compiler was: "If you know what this means...IMPLEMENT IT!" - On the Operating System I use, one of the wierdest errors is No requirements are obtained as to the exactness in size of a file in relation to the indicated size. - These are some of the error messages produced by Apple's MPW C compiler. These are all real. (If you must know I was bored one afternoon and decompiled the String resources for the compiler.) The compiler is 324k in size so these are just an excerpt I hope. I'm not sure where I stand on the copyright issue. "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" "You can't modify a constant, float upstream, win an argument with the IRS, or satisfy this compiler" "This struct already has a perfectly good definition" "This union already has a perfectly good definition" "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)" "Can't cast a void type to type void (because the ANSI spec. says so, that's why)" "Huh ?" "can't go mucking with a 'void *'" "we already did this function" "This label is the target of a goto from outside of the block containing this label AND this block has an automatic variable with an initializer AND your window wasn't wide enough to read this whole error message" "Call me paranoid but finding '/*' inside this comment makes me suspicious" "Too many errors on one line (make fewer)" "Symbol table full - fatal heap error; please go buy a RAM upgrade from your local Apple dealer" - The Macintosh System 7.* also has some weird error messages. This one occurs when a program 'exits' in a crash and the Mac tries to display why and how. If your Mac's boot HD is named 'Macintosh HD': 'The program has unexpectedly quit because Macintosh HD> Another one, this one appears when you ask, for instance, the Finder to copy a file to a folder, that is deleted by another program just before you copy a file to it: 'The command could not be completed, because it can not be found' - From the interactive help facility of xloadimage 3.0 or later: Type `?' for a list of options, or `.' or `quit' to leave the interactive help facility. help> `quit' The quotes around quit are unnecessary. You don't have to be so literal-minded! - I was at a demo of QuickTime when it came out a few years ago, and the demonstrators were bringing in some of the top-end Mac hardware of the day, but obviously hadn't had time to configure them completely. They were using a Barco projector, which happened to be on while they were messing around with the config, so the audience could see what was going on, much to the demonstrator's later dismay. One of the machines was lacking a necessary Ethernet driver, and so a lackey was sent running for a system disc caged from the computing department. When the driver was installed off the disc, the Mac was rebooted. The following error message had the hall in stitches: "Sorry, this version of system software is too new." An all-time classic, you'll agree :) - messages from Eudora, a POP mail program On asking for help on the priority of email messages: Highest:"This message is of highest priority." High:"This message is of high priority." Low:"This message is of low priority." Lowest:"This message is not worth the phosphors used to display it." On running low on memory on startup: "Memory is tight. You may need to close some windows, clean up your In, Out, and Trash mailboxes, or increase EudoraÕs memory size. Your current memory size is %dK; I suggest using at least %dK" The user is then presented with two options: "Quit" and "Live dangerously" On certain TCP errors: "The network failed (what this means is 'to be supplied in a future version of the MacTCP documentation', fat lot of good that does)." Another TCP error response: "Unexpected connection initiation segment read (no, I don't know what THAT means)." Yet another TCP error response: "That pesky MacTCP is acting up again." Error in displaying settings: "Couldn't display the settings window. Something is rotten in the state of Denmark." While decoding a binhexed file (I think): "Don't blame me; blame the CTB." On finding an MFS volume: "Your system folder is on an MFS volume, or something equally distasteful." And a few miscellaneous others: "I utterly abase myself, but I can't imagine what you mean by that date." "I'd love to help you, but that looks like it's in the past to me, and my time machine still needs a bit of work." "I looked really hard, but I couldn't find any addresses to make a nickname out of." - The most mysterious error message I've ever seen came out of the PL/I compiler for System/370, and happened to a friend of mine who was sitting next to me in the lab at the time. When the following came up on her screen, she pointed it out to me: Uncorrectable bug in line XXXXX - from the mail client 'elm': something godawful is happening to me!!! Emergency Exit taken! All temp files intact! - Running FTP on our VAX, I wanted help so I typed "?". The response was - Hello Sailor! To get help, type HELP at the FTP> prompt. - On the subject of weird error messages, an IBM RS/6000 cheered me up a bit with this little gem the other day: ".prog.tmp.f", 1500-008 (S) COMPILER LIMIT EXCEEDED in prog -: Program too complicated to be compiled. Compilation ended. Reduce the -complexity of the program and recompile, or lower the level of optimization -and recompile. 1501-511 Compilation failed for file .prog.tmp.f - On a more historical note there's the old Acorn BBC Micro retort when given invalid arguments to RENUMBER... > RENUMBER 10,0 Silly - These are some of the error messages produced by Apple's MPW C compiler. These are all real. (If you must know I was bored one afternoon and decompiled the String resources for the compiler.) The compiler is 324k in size so these are just an excerpt I hope. I'm not sure where I stand on the copyright issue. "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" "You can't modify a constant, float upstream, win an argument with the IRS, or satisfy this compiler" "This struct already has a perfectly good definition" "This union already has a perfectly good definition" "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)" "Can't cast a void type to type void (because the ANSI spec. says so, that's why)" "Huh ?" "can't go mucking with a 'void *'" "we already did this function" "This label is the target of a goto from outside of the block containing this label AND this block has an automatic variable with an initializer AND your window wasn't wide enough to read this whole error message" "Call me paranoid but finding '/*' inside this comment makes me suspicious" "Too many errors on one line (make fewer)" "Symbol table full - fatal heap error; please go buy a RAM upgrade from your local Apple dealer" - I had MicroSoft's MSD tell me "No help available. Leave me alone." when I hit F1 with one of the status windows on the screen. - Then there's the Linux lp code, which does some minimal error checking on the lp device before printing. If it encounters a printer condition that's not ready, off-line, or out of paper, you'll see lp: /dev/lp0 on fire! - Here's some error messages from an Apollo workstation. There's a explanation of the second one in a book I read called "Programming as if people mattered", or some title like that - I don't have the book here, or I'd give the author/title/etc. The other ones I don't know where they came from - probably the same source. % stcode 1D01001E Vendor "Apollo" can not be deleted (network license server/server) % stcode 220009 unit will not fit thru 25" hatch (OS/magtape manager) % stcode 13010008 trait not supported for wicked far-away objects (object based systems/ trait manager) - In the B6700's predecessor, the B-5500, on-line error messages were very terse. But, if you entered a question mark you would get a full description of the error. (Context sensitive help in 1970! Another Burroughs innovation like virtual memory.) The problem was what to say when there was no pending error. So we came up with: I AM THE GENIE OF THE DISK, WHAT IS YOUR COMMAND? [Upper case because the B-5500 had no concept of lower case] - from the IBM Fortran (I can't remember whether it was the G or the H compiler), circa 1970: `Too many comment lines (max=19). Excess comments have been ignored.' - from a programm written in Borland Turbo-Pascal. It crashed with a "runtime-error 208". I consulted the online-help and got: "runtime-error 208: runtime error." - from VME on ICL mainframes: 'A message, probably the one above, has not been output to all destinations' Another classic is: 'Warning: an unexpected error has occurred' - Many years ago, I hacked an override switch for the write-protect notch on my C-64's disk-drive. I wanted to test it out, but rather than write a program which could detect the write-protect status, I loaded up some fancy disk-copy program I had. It could detect when you had swapped disks by monitoring the write-protect status, and it would flicker the screen whenever the status changed. This was perfect for the job, and I flicked away happily at the switch, checking that the status was overridden correctly by watching the background of the screen. After I had flicked it about 8 or so times, so the program had detected about 4 disk changes without a single file being copied, up on the screen appeared a message for a few seconds : What the hell do you think you are doing? - Another module of the same program crashed one day, with a side effect that the background colour was changed from black to white. This accidentally revealed the programmer's little secret: Up in the top corner had been written in black on black, "Aaron loves Zara". (Only the names have been changed to protect my faulty memory.) - A more serious problem was some in-house test support code I had to use on a project. The original contractors who had written the code were long gone. Admittedly the task it was designed to do was a difficult one, but the product was painful to configure, slow to start-up and agony to use. One day I was using it, and started to rub my eyes in disbelief. I could have sworn it gave an error, ever so briefly before clearing the screen, which said "Viking Line". Viking Line is the name of a passenger ferry service in Scandinavia, where the code had been written. Grepping the code (which looked like it had been written by a computer - no human could write code like that, surely?), I found I was right. That was the error message. It was some time later before I worked out that it meant "Out of memory". Since then I have often wondered what was going through the programmer's head that day. - The error for floating point overflow on the FastBasic interpretter on the Atari ST: "Too big to be real" - The Miranda interpreter: when faced with an un-ending procedure, it displays "Black Hole" - Certain panels of ISPF/PDF (developmenet tool for IBM mainframes) issue the "Member not found" message if you mistype a member name. - Banyan VINES has perhaps my favorite error message: "VNS-xxx: There is nothing you can do." - Microsoft Word for Windows (*not* on my machine :-)): "You cannot quit Microsoft Word." - Infozip unzip.exe: "Can't find foo.zip or foo.zip.zip. So there." - Lora-BBS: "Sensor can't detect intelligent life." "ERR 999: Sysop confused. Press any key to return to reality." - There is a wonderful error number in ICL's VME which, you do a help on it, says 'There is either a software or hardware error'. - My favorite came from a Periphonics Audio Response Unit. If you entered any incorrect syntax from the console, the unit responded with: EH? - On a ZX Spectrum, I once had: nonsense in basic - error - This reminds me of the funny message VMS once issued when you type the command "STOP /CPU" on a single processor machine. It was something like SYS-E-?????, Device not in configuration or not available. - MOPAC eigenvector following routring: "Hereditary positive definiteness endangered." - The following is a GAMESS error message. I get the quotes a bit wrong but you get the idea. They all use upper case because they're twenty plus year old Fortran codes which used to run on big IBM's and the like and now run dead fast on my Linux box. INVALID BASIS FUNCTION. LUCY: DON'T WORRY CHARLIE BROWN, WE LEARN MORE FROM LOSING THAN WE DO FROM WINNING. CHARLIE: I MUST BE THE SMARTEST GUY IN THE WORLD THEN. - Since most computer programming is done at 2:00 AM, I wrote: switch (commandtype) { case ... case ... case ... case ... lots of other cases default: elog("\aERROR - PROGRAMMER NEEDS MORE SLEEP\n"); } Also, in same program (it was a PDP-8 simulator): ERROR - Bit 11. Invoking automated guess. [This occurred when the struct I was using (char word[12]) would get a value in an individual element (each element was a bit) greater than one (a bit, obviously should consist only of 0 or 1, but since this array held values of 0..255 in each element...). Sadly, this was removed when a bit field was used instead of a char array. Of course the program continued to run after this error was encountered. :) ] - From the BBC micro. (It was made by Acorn, a 32k 8-bit jobbie with a Teletext screen mode. (BBC was the same BBC as BBC TV)) When the BASIC ROM became dislodged at our school, the computer started up with: Language? So we whacked it a couple of times by picking it up and dropping it a foot to the desk, to which it responded STEPtax error. Another beauty was: >LOAD " Missing " [if you _know_ it's missing, why don't you put it _in_ for us?!] - From a GUI called Maestro by Softlab, you get an error window come up something like this: +---------------------------+ | Error | +---------------------------+ | No error found for | | "" | +---------------------------+ - Here's on I received today at work: rcp: protocol screwup: unexpected That was from a DEC OSF/1 server (aka Digital UNIX)