In the Trenches: a Guide to FFnet's War on Punctuation

If you're like me, you're probably getting tired of fanfiction.net's nasty tendency to strip punctuation from stories, summaries, and reviews.

Below is a (growing) guide to what punctuation FFnet strips from summaries and text of stories, and workarounds. Any corrections, updates, additional permitted/prohibited characters, new work-arounds, etc., appreciated.

Note: by reading further, you are agreeing to use the knowledge only for Good. In this case, that means you'll use your punctuation responsibly. You also agree not to be an evil spy/double-agent and tell FFnet about these workarounds, as they might then disable them and deprive the rest of us of our makeshift hyphens.

Summaries

The following standard keyboard characters are stripped from summaries:
` ~ ! @ # $ % ^ * ( ) _ - + = { } \ | ; " < > /

The following standard keyboard characters are not stripped from summaries:
& [ ] : ' , .

Additionally, all standard alphanumerics, most foreign language characters (Arabic, Cyrillic, Greek, Hebrew) and most accented Latin characters are left alone. Likewise, Japanese input via Windows's Language Toolbar also comes through unscathed.

Top Sekkrit workarounds for Summaries
(These all seem to work on Firefox 1.5 run on WinXP. Information as to their success on other browsers/systems appreciated, as are workarounds for other systems.)
See also this page for keyboard input of Unicode characters.

Using Character Map or Alt+numpad:
; Semicolon: The Greek question mark (U+037E)
‘ Left single quotes: Left Single Quotation Mark (U+2018) Alt+0145
’ Right single quotes: Right Single Quotation Mark (U+2019) Alt+0146
“ Left double quotes: Left Double Quotation Mark (U+201C) Alt+0147
” Right double quotes: Right Double Quotation Mark (U+201D) Alt+0148
– Hyphen: En Dash (U+2013) Alt+0150
— Longer Hyphen: Em Dash (U+2014) Alt+0151
﴾ Left Paren: Ornate Left Parenthesis (U+FD3E)
﴿ Right Paren: Ornate Right Parenthesis (U+FD3F)

Note that the ordinary open/close quote on Windows is nonstandard and not supported by other systems; the above forms of open/close quote are.
Using the WindowsXP Language Toolbar:
The following characters are not stripped when inputted in Japanese input mode, full or half-width katakana and hiragana:
‘～！＠＃＄％＾＆＊（）＿－＋＝｛｝「」￥｜；：’”、＜＞。・？
The following characters are not stripped when inputted in Japanese input mode, full-width alphanumeric:
‘～！＠＃＄％＾＆＊（）－＿＝＋［］｛｝￥｜；：’”，＜＞．／？

I imagine other non-Roman-character Language input will similarly provide forms of various special characters that are not stripped.

Story Text

The following standard keyboard characters are stripped from story text:
` ~ @ % ^ * _ = | \ < / [ ] { }

The following standard keyboard characters are not stripped from story text:
! # $ & ( ) - + ; : ' " , > ?

Additionally, all standard alphanumerics, most foreign language characters (Arabic, Cyrillic, Greek, Hebrew) and most accented Latin characters are left alone. Likewise, Japanese character input via Windows's Language Toolbar also comes through unscathed.

Top Sekkrit workarounds for Story text
This is the section you're really here for, isn't it? ^_^
Same disclaimers apply as above.

HTML Workarounds
HTML workarounds of the form &(numbers); will work if you're careful, but QuickEdit has some strange behaviors if you switch between Simple Mode and HTML Mode. I've found the following method works for all characters I've tried when using the HTML number:

1. Open the document you wish to correct in QuickEdit.
2. Switch to HTML mode.
3. Insert all desired ASCII characters using the HTML Number (see here for a chart). Do NOT switch to Simple mode.
4. Do NOT save until you have inserted all the HTML for the file, and then save exactly once.
5. Leave QuickEdit, and create story/add chapter as you would ordinarily.

Note that if you export this same file for editing sometime later, you will have to replace these characters with their codes again.

It's important to use the HTML number as some HTML Names (such as > and <) do not work. Also note that saving and switching modes converts the HTML code to the actual character; saving again or switching modes before saving will cause your workaround efforts to be eaten by the punctiaton-stripping monster.

Using Character Map or Alt+numpad:
See also this page for Alt-numpad input of Unicode characters.

٭ Not-Quite-Asterisk: Arabic Five-pointed Star (U+066D)
¹ Superscript One(for endnotes): Superscript One (U+00B9) Alt+0185
² Superscript Two(for endnotes):Superscript Two (U+00B2) Alt+0178
³ Superscript Three(for endnotes): Superscript Three (U+00B3) Alt+0179

Using the WindowsXP Language Toolbar:
The following characters are not stripped when inputted in Japanese input mode, half-width katakana and hiragana:
! # $ & ( ) - + ｣｢ ; : ' " ｡ > ､ ? ･
The following characters are not stripped when inputted in Japanese input mode, half-width alphanumeric:
! # $ & ( ) - + ; : ' " , > . ?
The following characters are not stripped when inputted in Japanese input mode, full-width alphanumeric:
‘～！＠＃＄％＾＆＊（）＿－＋＝｜￥；：’”，＜＞．／？［｛｝］
The following characters are not stripped when inputted in Japanese input mode,　full-width katakana and hiragana:
‘～！＠＃＄％＾＆＊（）＿－＋＝￥｜」｝｛「：；’”、。＞＜・？

Next update: Reviews

Questions, comments, suggestions, corrections? e-mail: (dontbugpeople AT gmail DOT com) with "FFnet" in the subject line.
Last updated 15 May 2006.