h a l f b a k e r y
Where life imitates science.

idea: add, search, annotate, link, view, overview, recent, by name, random

meta:

account: browse anonymously, or get an account and write.

 user: pass:
register,

# Upper and lower case numbers

Just because I'd like them
 (+15, -3) [vote for, against]

It would also greatly increase the number of possible password combinations. Being upper case or lower case would have no affect on the numeric value.

It's not fair for numbers to have only one case. it limits their wardrobe.

 — duroncrush, Feb 13 2004

upper and lower case numbers http://developer.ap.../TE/TE3numbers.html
I was going to do this a year ago until I saw that they already exist [FarmerJohn, Oct 04 2004, last modified Oct 21 2004]

Using numbers in the proper case - David Bergsland http://www.graphic-...ing_typography.html
Lowercase digits do exist, but they don't look much different. They (some of them) just have descenders like some lowercase letters (g, j, p, q, y) do. [tuc, May 05 2007]

Text figures http://en.wikipedia.org/wiki/Text_figures
Wikipedia’s article on old-style, non-lining numerals. [Crissov, May 12 2007]

IV,999 + I = V,000
 — Worldgineer, Feb 13 2004

 So, you would have to use an uppercase digit at the beginning of every calculation, or at the beginning of every number?

<side> While writing this anno I finally figured out why some pedantic people see a difference between the words number and digit </side>
 — kbecker, Feb 13 2004

You'll be opening a whole can of numerical grammar next.
 — Fishrat, Feb 13 2004

Mathematically, lowercase '4' should be infintesimally less than 4.
 — hippo, Feb 13 2004

//Mathematically, lowercase '4' should be infintesimally less than 4.// You be useful for orders "Give me '500'g of that cheese" would imply "Certainly not more than 500g" (because I don't have the money with me). Plain "Give me 500g of that cheese" would imply "Somewhere close to 500g". Then of course the problem is how to pronounce upper case numbers. Currently we even pronounce upper case letters.
 — kbecker, Feb 13 2004

Given that monospaced fonts with upper- and lower-case letters exist, I don't think it would be too difficult to keep cased numbers monospaced.
 — nick_n_uit, Feb 13 2004

I'd just like to point out that throughout this anno I've used uppercase spaces between the words.
 — hippo, Feb 14 2004

They do look nice as superscript, tough....
 — MuddyBuddy, Feb 14 2004

Murder by numbers, ¹²³
 — thumbwax, Feb 14 2004

Ahhh No! This would give H4x0rz the ability to shout.
 — silverstormer, Feb 14 2004

IIRC, on punched cards, it was possible to overstrike a digit 0 through nine with a dash. Thus, a five-column numeric field could hold numbers from -99,999 to 99,999 (for negative numbers, type a dash on top of the first digit). Unfortunately, although a hand-typed card produced in such fashion would show the column as a "-0" through "-9", copying the card would change the printed symbol to a letter "J" through "R". Incidentally and interestingly, a COBOL program that was expecting a numeric field would still interpret the letters "J" through "R" as "-0" through "-9".
 — supercat, Feb 14 2004

Unabubba: Actually, I would suggest that typographical spacing issues would be a good reason to favor having two sets of numbers (whether they're called "upper/lowercase" or something else). Indeed, some fonts have two sets of digits (one proportionally-spaced, one not) but there's no easy way to type the set which isn't mapped to the keys 0-9.
 — supercat, Feb 14 2004

As UnaBubba pointed out, the range of numbers is limitless. Thus there's no point in making upper and lower case numbers.
 — yabba do yabba dabba, Feb 16 2004

Yeah, he's got your digit.
 — Fishrat, Feb 17 2004

I've always wanted lowercase numbers, but could never think of a good reason to have them.
 — st3f, Feb 17 2004

Use the shift key to make them uppercase. I do it !000 times a day.
 — shibolim, Jun 25 2004

Roman numerals.
 — DesertFox, Mar 21 2006

 I just thought about posting superscript and subscript numbers, so that you can use powers in plain text. It turns out that these exist in unicode but aren't in a continuous series so they're both difficult to find and often not designed consisitently so they don't fit well together. (sup:?¹²³??????, sub:?????????? -- may not render in all fonts or browsers).

I figured it was a bit of a 'me too' so posted the though as an annotation here. I hope you don't mind my annexing your idea, duroncrush.
 — st3f, Apr 18 2006

This seems like a complex idea. I hate to be negative, but it's irrational. It's a pi in the sky concept if you ask me - and it has no integrity. Get real.
 — MaxwellBuchanan, May 05 2007

I think it's an idea who's prime has come.
 — normzone, May 05 2007

 //Reader's Digest used to do that with page numbers, having some big and some small and some dangle below the line like the tail on a small y.//

 Traditionally, typeset numerals often had varying widths, and non-uniform height and baseline. The round loops of the zero, six, and nine shared roughly the same vertical center; the stem from the six would extend up from that, while the one on the nine would extend downward.

 Having numerals designed in that fashion makes it easier to recognize a group of digits as a block (since e.g. "4906" will have a very different shape from "4609") but harder to read and process individual digits. Printing digits with uniform baseline, ascent, and spacing will make it easier to read individual digits, but will make all sequences of digits look much more similar.

 The situation quite analagous to the upper and lowercase alphabets in many typefaces. Uppercase is more legible, but lowercase is more readable. Someone who had to copy a sequence of gibberish letters such as an activation code could do so more easily in uppercase (e.g. "MPRNMJBXDO") than lowercase ("mprnmjbxdo"), but in general lowercase text in human language can be read more easily than uppercase.

(nb: Trebuchet seems to be better than some other fonts at lowercase letter sequences, and not as good at uppercase; I think my point still stands, though).
 — supercat, May 07 2007

 Rectilinear digits (such as those seen on digital clocks) could be used as the uppercase, while the curvy kind could be lowercase.

1 would be anomalous, having only one case, but that's okay since we don't need new 1's conflicting with i's when the original already looks like an l.
 — phundug, May 07 2007

i) Lower case numbers already exist
ii) No really, they do.
iii) I just can't remember where I last saw them
iv) Hmmm.
v) I'm sure I last saw them somewhere around here.
vi) Maybe I imagined it.
vii) Carry on.
 — zen_tom, May 07 2007

Baked, like Supercat explained. Open Type even supports four styles of digits, onum being what you want: - Lining numerals (lnum) - Old-style numerals (onum) - Proportional numerals (pnum) - Tabular numerals (tnum)
 — Crissov, May 12 2007

 One of the things that has always bothered me about numbers is how countries can't agree on how to separate your thousands, millions, etc. Is it:

 1000 1,000 1.000 1 000

 so how about "capitalizing" every third digit to save space. It would look like this:

1oo0
 — Bootbuckles, Aug 19 2010

 No! do it intuitively: 10ooo 150ooo 2ooo000 30ooo000ooo

 But don't forget the idea is about ALL numbers:

 so 153_425_220ooo The ones between the underscore are lowercase (more like handwritten font) numbers.

 A capital = sign would be very useful for C(capital)+ and C(capital)sharp.(PS Never get too close to someone giving a lecture about C++. There's a lot of spit involved)

I call on all HBers to try this system at least ince, and to use it on HB fr0m n0w and 0n.
 — pashute, Aug 19 2010

 This could get out of hand. Clearly most people here are refering to base 10. If it's extended to hex and binary, joy would be felt throughout the universe. By which I mean, everything would be hedious.

However, it can easilly be seen as a doubling of the base factor. Binary, decimal and hex would be increased to base 4, 20 and 32 respectivly. Interestingly, (or not perhaps) base 1 would be increased to base 2, changing the whole number sequence from having a sungular value to an infinite range.
 — saedi, Aug 20 2010

 Here's a substitution code I just came up with, based on lower-case letters that look kind of like numerals:

 jzeasbrygo

 j=1 z=2 e=3 a=4 s=5 b=6 r=7 y=8 g=9 o=0

 *j and not i, because i is already in use in roman numerals. Not l because it can be confused with I.

 **r is 7 backwards, and I understand some people will be bothered by this. My theory of glyphology says that letters written backwards should be equivalent to each other, or else somehow distinguished with additional strokes. (In my ideal world: b = d . But I can't change this now.)

***y for 8 because nothing else fits, except certain versions of g, and I used g for 9. x might fit 8 best, being composed of the middle part of that numeral, but like i, it's already in use in roman numerals. In SI units, prefixes yocta and yotta both represent an exponent related to 8, for octa.
 — Bootbuckles, Aug 27 2010

Not widely seen but some small-caps typefaces have two sets of numbers to match both sizes of caps.
 — st3f, May 25 2016

Always wanted to use them, but could never remember which ones had descenders... or when to use lc numbers properly.
 — FlyingToaster, May 25 2016

How about upper and lower class numbers?
 — r_kreher, May 25 2016

If I had a lower-case three, I could use it in my user name. I always though that st3f had the wrong word-shape.
 — st3f, Jun 15 2016

 [annotate]

back: main index