Adding the Trademark symbol to a document

You don't need to put the trademark symbol on a reference page. Just type

^q* wherever you want to insert it in your text.

You know, I didn't even know there was a Character Macros feature on a

reference page, so I looked it up. It appears that it is there only for

converting characters to an HTML equivalent. So the macros set up there make

the conversions only when the file is saved as HTML. For example, if you use

^q* to insert a trademark symbol in your document, when you save to HTML,

the trademark character will convert to whatever you set, such as "(tm)."

^q*  = ctrl key.then "q", then shift key, then "8"

It sounds like you are typing the characters as I wrote them (^q*), but ^q

is FrameMaker's shorthand for controlq. Ctrlq * is the sequence. So you

should press and hold the control key, then press the q key simultaneously.

Then release both. Then hit the asterisk key.  FrameMaker will automatically

insert the TM symbol for you, assuming your paragraph format uses a font

that contains the TM character.

If you're using FM 9, the TM symbol should be available from the Character Palette, File > Utilities > Character Palette.

Hello Judy:

To further confuse you with options see picture below:

To my knowledge, virtually all text fonts (not symbol and wingding type fonts) have the trademark symbol. So inserting it from the Symbols dropdown should insert it from the font current at that point.

Van

Almost all fonts contain the TM symbol. The fonts that may not have it are

fonts not normally used for body text,  such as Symbol and Dingbats. Another

easy way to enter the TM symbol (on Windows) is to type Alt-0153. (You must

use the numbers on the number keypad, not the numbers above the qwertyuiop

keys.)

^TM

Seriously? I hope you don't do it that way! A super-scripted two characters "TM" is not the same as the Trademark character that is what a good document should contain.

Z

> A super-scripted two characters "TM" is not the same as the Trademark character ...

It's also going to break hypertext links if the string it's in is used for any list generation (TOC, IX, etc.).

In FM8 and later, use the Unicode character U+2122, entered as \u2122 in dialogs (such as to create a variable to represent it). It's a single character code point, inherently superscripted. I presume that on Windows ALT+0153 is mapped to it, as is the legacy Framemaker dialog notation \TM

If ™ doesn't appear, then either the font is a legacy codepage implementation, or if a Unicode font, the font failed to populate a glyph for codepoint 2122.

Poking around a bit more on it, there's some peculiar stuff going on with ™, and probably with €‚ƒ„†‡ˆ‰Š‹ŒŽ‘’“”•–—˜š›œŸ as well. These are what might be called the C1 Controls orphans. They are present in Microsoft Code Page 1252 fonts, in the 80h to 9Fh range, but were dropped for ISO 8859-1, as they occupy space reserved for control codes. Unicode adopted 8859-1, and not cp1252, so also kicking these orphans out of that space, known as the C1 Controls.

TM is code point Alt+153 (99h) in cp1252.
Code point U+0099 in Unicode is now an unspecified control character, rendered (if at all) as "XXX" in a dashed box.

If we look at a legacy cp1252 font, such as Type 1 Courier, in the Windows Character Map ap, we see TM at code point 153d, aka 99h.

If we look at Type 1 Courier in the FM9 or later Character Palette, code point range 0080h-009F is surprisingly shown as vacant, and TM is surprisingly found at 2122h (which is, of course, impossible, because T1 Courier is an 8-bit font that has has no code points defined above 00FFh). The other C1 orphans are similarly moved to their coordinates in Unicode space (all above FFh).

Is FM synthesizing a Unicode version of Courier, with the C1 Control orphans moved to new code points?
Hard to say.

No matter how I enter a TM in FM9 (keyboard shortcut, \TM, ALT+153, \u2122), FM changes the representation to ™. I haven't so far been able to figure out exactly what FM is rendering to PDF for legacy and Unicode TMs. Even if I could find a generic decode-random-text app, Distiller, Acrobat Reader, and the OS clipboard could all be messing with the text code points before pasting it into the app.

This sort of magic goes on everywhere, it seems. Pasting raw TMs (or other special characters) into this forum's dialog routinely results in them being converted to HTML Entity markup, ™ in the case of TM. So, dear readers, be thankful that you only have to write text that includes special characters, and you don't have to write code that processes them from random source encodings to arbitrary receiving/rendering aps and displays.

> Is FM synthesizing a Unicode version of Courier, with the C1 Control orphans moved to new code points?

To answer my own question, it sure seems like it. For T1 Courier, FM's Character Palette (CP) shows a considerable number of glyphs above U+00FF. Due to annoying behavior by both FM CP and the Windows Character Map, there's no quick way to build a comprehensive list.

But it's clear that only some of the glyphs above FFh are relocated C1 Control orphans. Many appear to be composed Unicode characters made up of glyphs from various legacy Courier characters.

For example, code point
U+0133 LATIN SMALL LIGATURE IJ
is shown. T1 Courier has no ligature IJ, so this had to be done by compositing I and J.
Another is
U+01E6 LATIN CAPITAL LETTER G WITH CARON

This one is more of a puzzle, because T1 Courier has no stand-alone caron. Perhaps an inverted circumflex, or a lower case v, is being used as a hack.

If an application has nominal Unicode support, performing arbitrary compositing is relatively straight-forward. Unicode does not provide every possible "accented" characters for all scripts. It relies on Combining Diacritics (what in older localization schemes were non-advancing diacriticals, but are more like retreating diacriticals in Unicode). Any app pretending to support Unicode has to know how to merge (scale, place and align) multiple glyph fragments into what appears as a single character.