Thanks for the example. I can see what you mean by the CSS, it'd be nicer to differentiate between the original text & commentary using size of font or something? Also assume there is a limit to what text you can drag in? collect.readwriterespond.com/aldrich-press-…

@mrkrndvs Ultimately you could have more text than the original, so there's really no limit to the text beyond what your system will allow. I've not tried annotating .pdfs, but I suspect you could do it similarly to how I've done books via Kindle which let's me export/import data (example: http://boffosocko.com/2012/06/17/big-history/#Highlights,+Quotes). Having all the data on my own site (even if it's private) allows me much easier search via categories, tags, etc. as well. Typically for books I'm using fair use guidelines, though I'm often respectful and keep lots of my data viewable only to me, so copyright isn't a huge concern generally. For book length pieces, the Amazon Kindle will typically limit you to a percentage of the book to prevent copyright issues from the start. I don't think I've ever run up against their limit. Part of the process is to get enough to excerpt the thought to be able to find it later via search and reference it. I can always refer back to the original if necessary.

For more substantive annotations I try to add them slightly off to the side in a different font/size, but you'll also notice that some of the blue highlights also have "hidden" annotations if you hover over them with your mouse. Things can be a bit crowded with the additional highlights added by Hypothesis (and other people), though I can choose to disable them if I want for my personal work. Hopefully it's relatively obvious based on what's on the page which parts are the original text and which are my additions. When excerpting articles like Watters', the canonical URL on the page points to the original, so all the SEO and links ultimately redound to the original.