I was making “The Coda version Grammarly” doc when I realized I needed some help. THERE ARE SO MANY POSSIBLE WRITING MISTAKES (and also it could use some improvement).
So I thought I would ask the community for some help.
My doc follows a simple rule:
Mistake ; Real word
With a “message” column and a “type” column. :table
The message column is the text that shows the display message. e.g. Did you mean cat?
The column “type” shows what kind of suggestion it is. e.g. Ct > Cat is a spelling type, But remove “I mean” is a remove type.
To see the doc, click here. (PLEASE don’t, the doc is crashing right now I’m working on fixing this).
It looks like you need to adjust the sharing permissions on that doc in order for it to be publicly viewable.
This is an interesting idea, but I agree you’d need a pretty massive data set of mistakes and corrections. Spelling, for example, presumably needs a full spellcheck dictionary? (And for spellcheck things, instead of writing the possible mistakes, you would just check if the word exists in your dictionary table. But spellcheck is also probably beyond coda’s capabilities from a row count and performance perspective). I expect Grammarly is also analyzing the parts of speech in a sentence to check for things like subject-verb agreement etc.
But yeah if it’s more around a subset of common misspellings and stylistic no-no’s, could work!
Yes, that data is what I was worried about. It would probably take two years to get all of the data in place. and for the permissions, I might have to invite you to the doc.
Also, It would be good to have a word bank but there are over 3000 words in english. I wonder if there is a hidden AJAX formula.
