Technique to avoid duplicate content

I know we are not in 2004 anymore and i never tried this, but what if we scrape posts the include lots of hidden words (hidden to the user not google) between legit words ?

lets say, the original post is: i like potatos
we would use javascript to hidden words like: i <span class=hidden>lol</span>like <span class=hidden>nah</span> potatos ?

Would still be duplicate as the rest of the text would be the same, rigtt??

You can use original visible content instead of this technique ....

Seems like a lot of effort, plus google is smarter than that. You can easily scan over that by omitting all divs/spans and other attributes for body text.

Why not just re-write the content and save yourself time, with the added benefit of fresh content?