Wikipedia:Reference desk/Archives/Language/2016 February 27

= February 27 =

CJK font size
Font sizes in Western typography for running text vary from 6 pt to 14 pt: 6-8 for small format books like pocket dictionaries, 8-10 for most books of normal formats as well as newspapers, 10-12 for office documents, 12-14 for children's books. The sizes larger than 14 are considered for titles. These numbers may be not accurate, as I could not find exact statistics or guidelines, and publishers may apply their own decisions and styles, but you get the idea.

But what are the most widespread and normal font sizes for CJK? I know that Chinese publishers do not employ the Western point size system, but rather use their own system of font sizes. I have no idea what systems the Japanese and Koreans use.

A side question: how many CJK characters per book or A4/A5 page, on average?

P.S.(1) I've managed to download a PDF version of Renmin Ribao, and it says the size of the running text is 8 pt (looks like "Chinese No. 6"). Seems quite small even for the Western standards. However, I do not know for sure how it looks like in the actual printed form.

P.S.(2) Someone claimed that the Chinese government standard is 22 lines, 28 characters per lines, font size No. 3, though I cannot confirm if there is indeed such a standard and how it is actually applied in reality. --Lüboslóv Yęzýkin (talk) 17:02, 27 February 2016‎ (UTC)
 * Looking through several Chinese books at GoogleBooks I empirically found out the answer for my second question: on average on a page of a Chinese book there are 25-30 lines and 30±2 characters in each line. What surprised me, that there are still a lot of books printed with the vertical layout (funny, that interspersed text in the Latin alphabet is printed turned by 90 degrees clockwise) with around 20 columns by 40 characters per each on a page. If I knew the actual physical size of those books I could calculate the font size in them.--Lüboslóv Yęzýkin (talk) 18:25, 27 February 2016 (UTC)

No any further ideas? By the way, I found out something about CJK point systems, though nothing is said what actual sizes are widespread.--Lüboslóv Yęzýkin (talk) 13:23, 3 March 2016 (UTC)

Hebrew gimel typography
There was a complaint that the gimel used at our article gimel function doesn't look so much like a gimel, but more like a nun.

The gimel-function article uses inline LaTeX, which renders thus: $$\gimel$$. Indeed that doesn't look much like the unicode HEBREW LETTER GIMEL, which is ג (on my screen that looks like the picture on the right).

I tried it in my local LaTeX, and got the second picture on the right:

So, question, what is the difference between these two renderings of gimel? I can't find the second one anywhere in our articles. But it's the one I think mathematicians expect for the gimel function. It would be nice to say, "oh, that's the calligraphic style from such-and-such a tradition". --Trovatore (talk) 22:00, 27 February 2016 (UTC)


 * That shape of gimel -- "nun with a dent in the lower bar" -- is not at all uncommon; e.g. File:Hohe_Synagoge_Prag_1.jpg uses this shape, too. It's certainly not specific to LaTeX or to gimel function;  shows a wide variety of gimel shapes in different LaTeX fonts. Some are more nun-shaped, some are less. Google Images brings up a gematria table, where gimel and nun are positioned next to one another, making their similarity even more prominent. --My another account (talk) 22:47, 27 February 2016 (UTC)
 * Thanks much. Is there a name for this style?  The gimel function article says that it's a "serif" version, but this is not borne out by the table at gimel, where the so-called "serif" version doesn't look much different from the other versions, at least on my machine.
 * Sorry that this part of the question is more "computing" than "language", but is there a typeface name that will guarantee the nun-with-dent version, or maybe a different unicode? It would be nice to get rid of the inline LaTeX at gimel function, which is contrary to math-project conventions, but all the unicode renderings that I've seen look like lambda-with-hook rather than nun-with-dent.  --Trovatore (talk) 23:30, 27 February 2016 (UTC)
 * Ah, it looks like the unicode is GIMEL SYMBOL, ℷ, rather than HEBREW LETTER GIMEL, ג. That's probably the solution for the gimel-function article.  But it would be nice to be able to explain the situation at the gimel article.  I remain a bit confused about the relationship between these two code points. --Trovatore (talk) 00:10, 28 February 2016 (UTC)
 * The distinction between GIMEL SYMBOL and HEBREW LETTER GIMEL doesn't seem to imply a visual difference: at least, on my screen they look identical. https://www.w3.org/Math/characters/iso8879/isoamso.html includes a reference glyph for GIMEL SYMBOL which is neither "nun-shaped" nor "lambda-shaped".
 * I don't think there is any semantic difference between the two shapes of gimel, any more than between the "double-story" and "single-story" shapes of Latin g, which didn't get a code point for each. At least the Latin letter has names for the variant shapes! --My another account (talk) 09:53, 28 February 2016 (UTC)
 * As for the meaningless difference between the double story Looptail g.svg and the single story g, you could also mention the meaningless difference between Latin a and script ɑ. HOTmag (talk) 10:15, 28 February 2016 (UTC)
 * In fact, there are two code points for two version of the Latin g, U+0067 and U+0261, the latter is used in phonetic transcription. The semantic distinction between two versions of gimel looks similar. We have as well distinctive U+2126 OHM SIGN (not U+03A9 GREEK CAPITAL LETTER OMEGA), U+212B ANGSTROM SIGN (not U+00C5 LATIN CAPITAL LETTER A WITH RING ABOVE), etc.--Lüboslóv Yęzýkin (talk) 13:49, 28 February 2016 (UTC)
 * LATIN SMALL LETTER G (g) and LATIN SMALL LETTER SCRIPT G (ɡ) appear identical on my screen, just as GIMEL SYMBOL and HEBREW LETTER GIMEL do. In both cases, the distinction is semantic but not visual. --My another account (talk) 14:01, 28 February 2016 (UTC)
 * I agree, the difference is primarily semantical, in many fonts such as Helvetica/Arial two gs look identical. But you have said you "don't think there is any semantic difference between the two shapes of gimel", while in fact the difference is semantical. Their different look is secondary, but may have been created on purpose by font disigners, though I believe there are enough fonts where two gimels look identical. (I've just checked and found that there are at least 10 fonts with two identical gimels).--Lüboslóv Yęzýkin (talk) 14:45, 28 February 2016 (UTC)