I was wondering if anyone has come across anything like this in their travels? If so, please let me know.
To my knowledge, most tools allow you to check and replace on a one-by-one basis. They don't give you an easy way to see all spelling/grammar errors on a document-wide level.
What I Envision
Sample Input:
Quote:
This is the 1st erro. My names is tex. second, this is a diferent error. And this is a 2nd erro. i converted 2 diferent dollars to erros when I visited Europe.
|
Potential Output:
- (5) Spelling
- (3) erro -> error
- This is the 1st erro.
- And this is a 2nd erro.
- i converted 2 diferent dollars to erros when I visited Europe.
- (2) diferent -> different
- second, this is a diferent error.
- i converted 2 diferent dollars to erros when I visited Europe.
- (3) Grammar
- (1) is -> are
- (2) Capitalization
- (1) Second
- My names is tex. second, this is a diferent error 1.
- (1) I
- i converted 2 diferent dollars to erros when I visited Europe.
Seeing them broken down this way would allow you to easily tell,
at a glance, how many of which types there are + surrounding context.
You could also focus purely within a category to see if the recommendations are actually correct:
- (diferent -> different). You would be able to "Replace All" in one swoop.
- But (erro -> error), "dollars to erros" should be "dollars to euros".
Current Tools (That I'm aware of)
One-by-One Checking (Most tools are like this)
If using Word/LibreOffice, you have to go through one-by-one in the order they appear, and Correct/Ignore/Ignore All.
Pain Point: Word's grammarcheck frustrating, because you can only "Ignore" and CAN'T "Ignore All". On a huge book, this takes forever.
Pain Point #2: In large works, there are typically common issues repeated throughout the entire document (author misspells "erros" + consistently misses a comma before/after certain words). You can't easily tackle all comma errors in a given pass, or solve them consistently, because you're constantly flipflopping between all the different types of issues.
Pain Point #3: Another frustration in Word is when you get the dreaded "Too many spelling or grammatical errors" and it refuses to show you the red/blue squigglies within your document. This makes seeing context much more difficult.
List-based Grammar-checking
LanguageTool's standalone tool allows you to get an entire list of grammar errors in the order they appear:
This is pretty great! And it allows you to "Ignore All" entire rules.
But because you can't easily tell how many hits this rule actually caught in the entire book (is there just 1? Or 50?), you sometimes don't want to Ignore the entire rule.
Pain Point: Because you can't easily ignore, there are a ton of false positives clogging up the list.
Pain Point #2: You also aren't too sure what Ignoring a certain rule would effect. Take for example, the "Capitalization" grammar rule:
1. Ignore that
specific instance of that
specific word?
- That second -> Second would be ignored, future ones would be caught
2. Ignore all future instances of that
specific word?
--- second -> Second would be ignored, future "second"s also ignored.
3. Ignore
all cases of capitalization errors?
--- second -> Second would be ignored, but so would i -> I.
List-based Spell-checking
Sigil/Calibre allow you to use the Spellchecking Lists to see errors in mass.
This allows you to see all misspelled words in the entire book in an easy to view list, but you can only see individual words, not the surrounding context:
You have to manually search for these words, or doubleclick the word X times in the Spellcheck List to jump to each instance.
Pain Point: Again, when working on huge books, this can be a pain, especially if it's dealing with words that are very common, but potential misspellings in some contexts (or OCR errors, like "modem"<->"modern").
Reason Why I Thought of This
I recently was looking at what they added into the latest versions of Word, and saw they introduced an "Editor Pane":
https://answers.microsoft.com/en-us/...3-6ea27c8f31f1
It finally allows you to split the Spelling + Grammar into separate steps (one of my pet peeves with a lot of the tools, constantly flipflopping between "Spelling Mode" and "Grammar Mode").
Sigil's method is by far my favorite way to spellcheck, but I think the hybrid approach I mentioned above (with lists AND context), would be yet another way to correct errors efficiently.