What is Digg?Sponsored by Best Buy
He sings, he strums, and he works at Best Buy. view!
www.youtube.com/bestbuy - Musician and Best Buy employee, Keith Parsons, rocks his Best Buy holiday campaign audition.
184 Comments
- inactive, on 10/12/2007, -5/+67sadly, typing "there is no cow level" doesn't open a copy of diablo 1. [/obscure reference]
- nevries, on 10/12/2007, -1/+42The problem is caused by the way Notepad (or Notepad2 which has the same problem) determines if a document is ASCII or UNICODE. Since there is no real way to determine the difference with a text document (since there is no meta data) a heuristic has to be used. In most cases this works fine, but sometimes it goes wrong. For long(er) documents the chance of a misfire is very very small.
- zumpiez, on 10/12/2007, -6/+44That's extremely relevant and helpful, thanks
- sigalrm, on 10/12/2007, -3/+33The string "this app can break" is just one of any combination of a sequence of 4, 3, 3, and 5 lower-case alphabetic chars [a-z] separated by a single space, that cause this bug/egg to show up. And you'll get different combinations of unicode each time. So attempts to translate this to "Chinese", etc. are really funny...
"xxxx xxx xxx xxxxx"... works too.
Probably a bitshifting thing. - Thuktun, on 10/12/2007, -0/+29This isn't an easter egg, this is Notepad trying to guess the file encoding and guessing wrong. Some aspect of that string's ASCII encoding makes Notepad think it's actually a 16-bit Unicode file. If you open the file, see the squares (which is usually what Windows shows if it doesn't have any glyph in the current font for that character), and save the file, open the result in a hex editor. You'll see it that it wrote the Unicode byte-order mark (BOM, 0xFEFF) before your string.
- inactive, on 10/12/2007, -1/+29You guys are missing the point--it's not an easter egg, there really is a bug in Notepad where it doesn't like certain sequences of characters!
Try this string instead: "aaaa bbb ccc ddddd" - Jerf, on 10/12/2007, -0/+24It's not a bug and it's not an Easter Egg; it's "broken as designed".
A file is just a series of bytes, not characters. It's *always* been true that "bytes" and "charecters" aren't the same thing, but now with even us native-English speakers routinely being exposed to Unicode we can't pretend anymore.
Notepad opens the file you create, and all it knows is that it is a sequence of bytes. It could be encoded in any of dozens of encodings that have been used over the years. One of the improvements to Notepad almost nobody who speaks English natively noticed is that Notepad has become encoding-aware, so it's not just dumping bytes to the screen and assuming that they are one of the standard ISO-* western encodings. Since the plain-text file has no markings on it as to which encodings are in play, it has to guess.
Sometimes it guesses incorrectly, and you get what happens here.
The shorter the text, the more likely it is to screw up.
I seem to recall that one of the checks it does is to analyse the distribution of letters to see if it's English or one of the other European letters, and, again, the shorter the text the more likely it is to screw up.
Apparently there's a Chinese encoding that uses the same bytes as English text. Bad form, but that's life.
XML solves this by asking all files that aren't UTF-8 to so identify themselves right up front with the encoding declaration in the processing declaration up front. Many other protocols have similar solutions where the character set must be explicitly declared, like HTML. But Notepad gets no such help.
It's actually impressive it doesn't screw up more often. - jobo5432, on 10/12/2007, -0/+21Try the following:
enter: ".LOG" (without quotes)
save the file, and close. Reopen as necessary. Enjoy.
-jim - BIGGY350z, on 10/12/2007, -2/+22Wow...what a friggen revelation. There's something better out there than Notepad. Crap no wonder I failed all my college papers.
- aNoble, on 10/12/2007, -2/+17Very strange. It's not an easter egg, I think it's an actual bug. Swapping the letters around to an extent still causes the same problem.
Try this:
siht ppa nac kaerb
(same word order but reversed letters) - Indrek, on 10/12/2007, -2/+17You knew about this a year ago and just kept quiet about it? Not cool, man.
- Indrek, on 10/12/2007, -0/+14The three spaces between the words?
- merreborn, on 10/12/2007, -1/+14Strangely enough this also works if you create the file in another text editor. I created a new text file, opened it in text pad, typed the string, and saved. When loaded with textpad, it contains "this app can break" as expected. When loaded in notepad, it displays the boxes.
Notepad's unicode detector probably sucks. - Indrek, on 10/12/2007, -0/+12Simply hitting F5 is a bit faster, though.
Still, nice find. - garble7, on 10/12/2007, -2/+13" After 獩 indignant grey harassment 湡 personal sounded Fun " is what Google Translates it to.
Actual characters 桴獩愠灰挠湡戠敲歡 - recursive, on 10/12/2007, -0/+11@masamunecyrus:
break has 5 letters, not 4
'aaaa aaa aaa aaaaa' kills it.
@bjtylerx:
That's because you are saving the subsequent files as unicode. - Momoru, on 10/12/2007, -0/+10Raymond Chen explains why:
http://blogs.msdn.com/oldnewthing/archive/2004/03/24/95235.aspx - huntr2, on 10/12/2007, -0/+10It's not an easter egg, it's a bug because it does it with other strings. Any string using the combination of words with a length of 4-3-3-5 seems to break it.
"four one two three" will generate the same effect, likewise
"aaaa aaa aaa aaaaa" will do the same.
Only seems to work with a-z or A-Z if you mix cases it displays normally. - redzombi2k, on 10/12/2007, -1/+11Actually as eli867 pointed out it also works on "aaaa bbb ccc ddddd".
However it does not change the content, just displays it differently.
Try opening it up in a different text editor, (I use not tab light)
and you will find that it will display it correctly. - Courthold, on 10/12/2007, -3/+13there is no cow level was a Starcraft cheat wasn't it? I'm actually in the cow level on Diablo 2 at the moment, still a great game!
- Lorian, on 10/12/2007, -13/+22Yeah, this isn't an error with Notepad, it's an Easter egg.
- silic0n, on 10/12/2007, -2/+11And I thought they had to take all Easter Eggs out of Windows versions past 2000 because the military use it and didn't want 'unspecified features'.
- Lagged2Death, on 10/12/2007, -1/+10I tried creating a "this app can break" file with a programmer's editor. The file I created is verifiably a plain-old ASCII file. That breaks Notepad, and also breaks a Notepad replacement called Metapad that I frequently use. Metapad actually shows a warning: "Detected non-ANSI characters in this Unicode file." Something about that file causes some programs to try and interpret the text as Unicode, which doesn't work, because it's not. I doubt it's an Easter egg, since it effects multiple apps.
"xhis app can break" and other small variations also work the same way. - lane.montgomery, on 10/12/2007, -1/+10how about this?
"k a r t h i c k si a tnaig ssabmud" - Indrek, on 10/12/2007, -0/+8Fasten your seatbelts, ladies and gentlemen, and put on your flameproof helmets. The text editor wars are coming to Windows.
- Indrek, on 10/12/2007, -1/+9Nice try, but no cigar. Try "minu ema oli müüja".
- agentgonzo, on 10/12/2007, -0/+8Because it's not working as intended. It should display the text. It isn't. That makes it a bug
- Indrek, on 10/12/2007, -0/+7A lot of strings of 4+3+3+5 lowercase letters seem to work.
"test foo bar jello" works, but "test foo bar jelly" doesn't. Odd...
It doesn't even mind if you use umlauted characters. "õõõõ äää ööö üüüüü" works as well.
Has anyone figured out the pattern yet? - agentmith, on 10/12/2007, -3/+10...other than myself and kepsux, anyone else name the notepad file test.txt?
- toveling, on 10/12/2007, -0/+7Doesn't work under wine, running the real windows notepad.exe from another partition.
- crizo, on 10/12/2007, -0/+7You beat a few of us to it... But here's a link that pinpoints the exact cause to the function "IsTextUnicode"
http://blogs.msdn.com/oldnewthing/archive/2004/03/24/95235.aspx
Not an easter egg, and not a bug that's likely to get fixed. - Turkwoyz, on 10/12/2007, -1/+8it breaks on both my hardrives (one's FAT32, the other NTFS) so that's not it
- dWhisper, on 10/12/2007, -2/+9That has to be intentional. I wonder if the person who coded that in still works for Microsoft, and if he's just been waiting for a chance to laugh?
- muffinking, on 10/12/2007, -1/+8I get "慡慡戠扢挠捣搠摤摤" with that one. And what's worse, digg spellcheck finds no errors there.
- ErrandboyOfDoom, on 10/12/2007, -10/+16桴獩愠灰挠湡戠敲歡
- kepsux, on 10/12/2007, -0/+6http://i7.photobucket.com/albums/y258/KarnEvil60/Graphic_5_22_20065_04_18PM.jpg
- Namco, on 10/12/2007, -4/+10break.txt
- garble7, on 10/12/2007, -3/+9In Vista the boxes turned to Chinese characters. They would probably turn to chinese if you had the font installed or the language. Now, what does it say in chinese?
- bjtylerx, on 10/12/2007, -0/+6"aaaa bbb ccc ddddd" broke it for me too - if anyone else is trying this make sure to open a new file...if you delete the boxes from "this app will break" and type something new, nothing will break.
- recursive, on 10/12/2007, -0/+6Other combinations seem to work. The shortest I found so far is 4-3-1. In other words, 'aaaa aaa a' breaks notepad.
- kitsonk, on 10/12/2007, -1/+7Well notice that it switches from single byte encoded to double byte encoded. "th" "is" " a" "pp" " c" "an" " b" "re" "ak"... so I think for some reason, notepad is getting confused on how the file is encoded, it is just the right length that it misreads the file and just the right charecters to make it think it is valid.
- ironcamel, on 10/12/2007, -3/+9xxxx xxx xxx xxxxx didn't work for me. Does anyone know any other strings that would cause this behavior?
- agentgonzo, on 10/12/2007, -3/+9... but if you save it to a samba share on another windows machine, it still breaks.
- Otto, on 10/12/2007, -1/+7"abcd abc abc abcde" also causes it.
Seems like almost any set of lower case letters in 4 3 3 5 sets will trigger it. - ErrandboyOfDoom, on 10/12/2007, -1/+6"this app can break" only worked for me the first time, and I can't figure out what I'm doing wrong the second time...
- dasch, on 10/12/2007, -0/+5There's also an option in the Danish version of MS Word that reads "Beskyt mod horeunger" -- which translates to "Protect against bastards"...
- tomobc, on 10/12/2007, -2/+7You mean this "桴獩愠灰挠湡戠敲歡" ? I think it's meaningless. Seems like some randomly produced characters.
- Indrek, on 10/12/2007, -1/+6Bingo. Create a text file with one of those "forbidden strings". Then open Notepad, hit Ctrl O and observe how the encoding changes from the default - ANSI on my computer - to Unicode when you select the file.
Also, you can make any text file "break" the same way by saving as it ANSI and opening as Unicode. - aNoble, on 10/12/2007, -0/+5And yet using a capital anywhere seems to "fix" it.
Try:
This app can break - IcedZ, on 10/12/2007, -0/+5HAH.. yeah.. that was stupid. I was up late... yeah, thats it. I'm tired =)
-
Show 51 - 100 of 184 discussions
Browsing Digg on your phone just got easier with our enhancements to the 
© Digg Inc. 2009
— Content created and posted by Digg users is 