• (nodebb)

    I suppose we won't discuss the small point that the code presented is probably not endian-safe, and on most modern architectures, it requires that the file's two first bytes are not "0xb0/0xb1/0xb3 followed by 0xab".

    Maybe that's what's wanted, but I have some doubts.

  • (nodebb)

    Maybe the lowercase numbers can be the odrinals: frist, secnod, thrid, and so on.

    I had thought there had been a comment in https://www.explainxkcd.com/wiki/index.php/2206:_Mavis_Beacon about uppercase and lowercase numerals in Chinese, but can't see it there.

  • Anon (unregistered)

    but how do we write a lowercase 3?

    Like this: ₃

  • Industrial Automation Engineer (unregistered)

    Well, 3 is lowercase. Uppercase would be # (as in Chanel #5, or the email adress [email protected], which sports an uppercase 2.)

  • ebbek (unregistered)

    or the email adress [email protected], which sports an uppercase 2. ...if you are using an English keyboard. On a Danish keyboard uppercase2 is a " character.

  • (nodebb) in reply to Industrial Automation Engineer

    I'm sorry, but this isn't true. 1 is clearly an upper-case &. It's plain as day on my keyboard.

  • NoOne (unregistered)

    What about ³ ?

  • Smithers (unregistered) in reply to ebbek

    ...if you are using an American English keyboard.

    FTFY. On an English English keyboard, uppercase 2 is ", just like in Denmark, while uppercase 3 is £.

  • (nodebb) in reply to ebbek

    if you are using an English keyboard. On a Danish keyboard uppercase2 is a " character

    Tsk. There are multiple keyboard layouts that one could call English (== designed for the English language), and they aren't the same in the "upper case" versions of the top row. Indeed, in context, we have to note that the one that's actually English (== used in England) has a " on the upper-case 2 position, and not an @.

  • Naomi (unregistered)

    There actually is such a thing as lower-case numbers, although they aren't very interesting. I'll provide a link in a follow-up comment so this one doesn't get held for moderation.

  • Naomi (unregistered) in reply to Naomi

    https://graphicdesign.stackexchange.com/questions/54423/why-dont-upper-case-numbers-exist

  • NoLand (unregistered)

    In ASCII numbers are clearly lower-case forms of the range SPACE, !, ", #, $, %, &,', (, ) — as may be discerned by having bit 4 set.

    (By simple analogy, what we know as numbers are simply lower-case punctuations with bit 4 set, while lower-case letters have bit 5 set. ;-) )

  • Tinkle (unregistered)

    Oxb1 says 'these are not the header bytes you are looking for'

  • Kenn Humborg (unregistered)

    And they completely missed 0xAbb0, 0xABb0, 0xAbB0, etc!

  • (nodebb) in reply to Industrial Automation Engineer

    Uppercase would be #

    Well on my keyboard, uppercase 3 is £...

  • Sauron (unregistered)

    Dev thinks hexadecimal numbers should have a difference between lowercase and uppercase.

    Expectation: Dev gets a lower pay.

    Reality: Dev becomes upper management.

  • Registered (unregistered)

    Lower case '3' is 'e': Think L337 - 1ee7

    But what is lower case 7?
  • DiscoGuy (unregistered)

    It's unfortunate they didn't account for $ABBA - clearly not a European app.

  • Officer Johnny Holzkopf (unregistered) in reply to PJH

    On a typical German German keyboard, uppercase 3 is § (paragraph sign). On a typewriter, lowercase one is i, uppercase one is I, because there is no 1; and as there also is no 0, lowercase zero is o, uppercase zero is O. We wrote whole programs consisting of Is and Os! But is XXIII < or > or == xxiii? Please ask a Roman Latin keyboard...

  • löchlein deluxe (unregistered)

    Clearly, "lining" numerals are uppercase and "old-style" numerals are lowercase.

  • (nodebb)

    I have my doubts whether the programmer in question was actually working for any sort of professional concern in any sort of professional capacity. This seems like student work ... as in very early in the stage of learning to program... ag shame man.

  • LZ79LRU (unregistered) in reply to DonaldK

    Oh you sweet summer child.

  • lzsiga (unregistered)

    Right, the code is not endianness-safe, but if the file-creation was written by the same guy, then it sort of works (though the-file format is incompatible betwen LE and BE platforms).

  • sten (unregistered)

    Might be that I don't yet have enough coffee in my system, but I had to read this one multiple times to understand the issue that they have to check for a string value at the beginning of the file, and not a hex number.

  • (nodebb)

    Everybody seems to be confusing the shift function of a keyboard with different cases in a typeface. In fact, there is no such thing as an upper or a lower case number. The categories do not exist.

    http://www.unicode.org/reports/tr44/#General_Category_Values

  • Strahd Ivarius (unregistered) in reply to Steve_The_Cynic

    The UK keyboard is English English. The other variants are Colonial English.

  • Strahd Ivarius (unregistered)

    For the Tagged Image File Format, the magical bytes at position 3 and 4 represent number 42 (the previous magical bytes tell us if the bytes are stored as big-endian or little-endian). In version 5 of the document describing the format, it was said that 42 was chosen for "philosophical reasons that would be tool long to explain here"... After version 6 (under Adobe tutelage), this explanation was removed :-(

  • Anon (unregistered)

    What's all this crap about Uppercase 2 being '" or sometimes £ ?

    If I press Caps-Lock so whatever I type is always Uppercase then I type a top-row number, I get the number not a symbol. Therefore that concept is just wrong.

Leave a comment on “Magical Bytes”

Log In or post as a guest

Replying to comment #:

Log Out

« Return to Article