section 4.1

A Heading in Upright Shape

A paragraph in italics. A paragraph in italics. A paragraph in italics.

A sentence in upright shape.

A Heading in Italics

A sentence in upright shape. A sentence in italics, which occurs once in Benedetti (1585). A sentence in upright shape.

section 5.1

elements with plaintext model

The Roman number vii and the vulgar fraction ½. The line AB. The regularized form acutiſſimisque.

section 5.2

elements with inline model

This sentence contains a word in italics, a bold word, the combination of bold and italics, a word in Small Caps, a word in subscript, a word in superscript, a word in smaller font size, an underlined word, an overlined word, a struck-through word, a word in red, and a word in Sperrung. (The German word Sperrung is marked by <foreign> in the xml, but not in the xhtml.) The elements in the inline model do not need their own xml:space attribute because they inherit it from <s xml:space="preserve">.

This paragraph is in Fraktur if “Lucida Blackletter” is inſtalled. Note that this font has problems with the long s and does not contain the obligatory ligatures. (Engliſh is rarely written in Fraktur, though.)

section 5.3

quotes

This sentence contains an “inline quote”. This sentence contains a set-off

“inline quote within a sentence”,

A new paragraph starts here. This sentence does not contain a set-off.

“ Quote within a paragraph, but not within a sentence. ”

A new paragraph starts here. A normal sentence. “ This quote within a paragraph is one sentence long. ” A normal sentence. “ This quote within a paragraph is more than one sentence long. It ends here. ” This paragraph also contains the following set-off blockquote:

This set-off quote within a paragraph is three sentences long. This sentence contains a bold word. This set-off quote within a paragraph is three sentences long.

This is a short blockquote. It it is not within a paragraph. Short means that it consists of a single paragraph, or more precisely some <s> without an enclosing <p>, i.e. it is on the same level as a paragraph.

A paragraph with a single sentence between two blockquotes.

This is a long blockquote, which consists of two paragraphs. This is the second sentence of the first paragraph.

This is the first sentence of the second paragraph. This is the second sentence of the second paragraph.

section 11.1

linebreaks at different positions

Blah blah blah. A linebreak in the middle
of the sentence. Blah blah blah. A linebreak in the mid-
dle of a word. Blah blah blah.
A linebreak at the beginning of the <s> (bad). Blah blah blah. A linebreak at the end of the <s>, after the period (also bad).
Blah blah blah. Instead, linebreaks at the beginning or end of a sentence should be moved out, as between this and the next sentence.
However, the schema accepts all three versions, and in your browser you will see no difference.

Since it's only a convention which doesn't change the semantics of the text, one could ask whether it would make sense to move linebreaks into the sentence rather than moving them out. When post-processing the transcription, the period and "return" can be replaced either by

... .</s><lb/>
<s>...

... .<lb/></s>
<s>...

... .</s>
<s><lb/>...

section 11.2

pagebreaks at different positions

Blah blah blah. A pagebreak in the middle

(page 1)

of the sentence. Blah blah blah. A pagebreak in the mid-

(page 2)

dle of a word. Blah blah blah.

(page 3)

A pagebreak at the beginning of the <s> (bad). Blah blah blah. A pagebreak at the end of the <s>, after the period (also bad).

(page 4)

Blah blah blah. Instead, pagebreaks at the beginning or end of a sentence should be moved out, as between this and the next sentence.

(page 5)

However, the schema accepts all three versions, and in your browser you will see no difference.

For pagebreaks the situation is slightly different than for linebreaks. The arguments of "natural" versus "better for the tree representation" are the same; the "natural" argument is perhaps even stronger than in the case of linebreaks because the typists are instructed to type the entire contents of one page before going on to the next page. In addition, when an XML fragment of a page is created, a pagebreak within a sentence or paragraph would create an artifact sentence or paragraph at the beginning of the next page, unless the script takes this situation into account. However, a strong argument for one pagebreak at the beginning of <div> is that the <pb> tag contains the running head. (This applies when the div starts at the top of the page; if the div starts in the middle of the page and the running head alerady gives the name of the new div, there is nothing one can do.)

section 17.1

Test with a note by Georg Christoph Lichtenberg:

Sind wohl alle Wurzeln aus dem Unendlichen als zum Beispiel $\sqrt{[}m]{\infty }^n$ rational oder sind es nur die $\sqrt{[}m]{\infty }^{mn}$ wo $m$ und $n$ ganze Zahlen bedeuten? Ich sollte denken, denn da $\sqrt{[}m]0^n$ gewiß $=0$ und folglich rational ist, so wird auch $\sqrt{[}m]\frac{1}{{\infty }^n}=0$ rational und folglich $\sqrt{[}m]{\infty }^{-n}$ und da $n$ eine verneinte Zahl ausdrücken kann auch $\sqrt{[}m]{\infty }^n$. (L 824)

The result is somewhat disappointing. In the second version, all formulas representing single variables have been replaced by <var>, and the “= 0” has been replaced by a normal = and a <num> tag:

Sind wohl alle Wurzeln aus dem Unendlichen als zum Beispiel $\sqrt{[}m]{\infty }^n$ rational oder sind es nur die $\sqrt{[}m]{\infty }^{mn}$ wo m und n ganze Zahlen bedeuten? Ich sollte denken, denn da $\sqrt{[}m]0^n$ gewiß = 0 und folglich rational ist, so wird auch $\sqrt{[}m]\frac{1}{{\infty }^n}=0$ rational und folglich $\sqrt{[}m]{\infty }^{-n}$ und da n eine verneinte Zahl ausdrücken kann auch $\sqrt{[}m]{\infty }^n$. (L 824)

In the third version, the m-th roots have been removed:

Sind wohl alle Wurzeln aus dem Unendlichen als zum Beispiel $\sqrt{{\infty }^n}$ rational oder sind es nur die $\sqrt{{\infty }^{mn}}$ wo m und n ganze Zahlen bedeuten? Ich sollte denken, denn da $\sqrt{0^n}$ gewiß = 0 und folglich rational ist, so wird auch $\sqrt{\frac{1}{{\infty }^n}}=0$ rational und folglich $\sqrt{{\infty }^{-n}}$ und da n eine verneinte Zahl ausdrücken kann auch $\sqrt{{\infty }^n}$. (L 824)

In the last version, the <msqrt> has been replaced by <mroot> with additional <mi>m</mi>:

Sind wohl alle Wurzeln aus dem Unendlichen als zum Beispiel $\sqrt[m]{{\infty }^n}$ rational oder sind es nur die $\sqrt[m]{{\infty }^{mn}}$ wo m und n ganze Zahlen bedeuten? Ich sollte denken, denn da $\sqrt[m]{0^n}$ gewiß = 0 und folglich rational ist, so wird auch $\sqrt[m]{\frac{1}{{\infty }^n}}=0$ rational und folglich $\sqrt[m]{{\infty }^{-n}}$ und da n eine verneinte Zahl ausdrücken kann auch $\sqrt[m]{{\infty }^n}$. (L 824)

section 17.2

This is the same text with TexX4ht. One still has to re-format the result and fill in the LaTeX formula in the comment, but apart from that it seems to work well. However, it sometimes produces slightly illegal MathML; I had to insert <mstyle> in the two formulas with an = sign.

Sind wohl alle Wurzel[n] aus dem Unendlichen als zum Beispiel $\sqrt[m]{{\infty }^n}$ rational oder sind es nur die $\sqrt[m]{{\infty }^{mn}}$ wo $m$ und $n$ ganze Zahlen bedeuten? Ich sollte denken, denn da $\sqrt[m]{0^n}$ gewiss $=0$ und folglich rational ist, so wird auch $\sqrt[m]{\frac{1}{{\infty }^n}}=0$ rational und folglich $\sqrt[m]{{\infty }^{-n}}$ und da $n$ eine verneinte Zahl ausdruecken kann auch $\sqrt[m]{{\infty }^n}$. (L 824)