https://hypothesis-open.slack.com/archives/C6Y375TT3/p1678893693075289
I suspect this would be a better place than issues. I've been using hypothesis for years, and the friction when annotating has always been a barrier for me. IMO, annotation should feel like a native part of the browser, which it just isn't when you need to click the badge. There are three approaches I've seen discussed
Active everywhere
This one is by far the most simple, but introduces many problems. Some pages without content bound to a permalink are undesirable to annotate on, like timelines and search engines, so seeing annotations there would be terrible. Additionally, the tooltip could conflict with tooltips on the original page, like editing tooltips or native annotation functionality (though it would be nice if everyone just used hypothesis 😁)
Active if there are prior annotations
With this one, it's not active everywhere, but not much better. The same timeline problem applies, and I frequently find myself being the first one to leave annotations. It enhances the social aspect, but purely as an annotator, I don't think anything is gained
My idea: heuristics
There are certain qualities of content that can effectively determine if
- the user is likely to want to launch hypothesis
- Anticipating that would not degrade the UX, even if the user didn't actually want to.
The most obvious heuristic is meta tags, which are a blessing. For articles to embed appropriately in sites like discord, twitter, and slack, (as well as SEO?) websites add meta tags which can give some insight into the content. Specifically, the og:type property with the content of article seems to be widespread among the content I annotate, and is standardized. In my opinion, this is the best of both worlds. Hypothesis, by design, is unlikely to be intrusive in articles, and in almost all cases it will already be active when the user goes to annotate.
https://hypothesis-open.slack.com/archives/C6Y375TT3/p1678893693075289
I suspect this would be a better place than issues. I've been using hypothesis for years, and the friction when annotating has always been a barrier for me. IMO, annotation should feel like a native part of the browser, which it just isn't when you need to click the badge. There are three approaches I've seen discussed
Active everywhere
This one is by far the most simple, but introduces many problems. Some pages without content bound to a permalink are undesirable to annotate on, like timelines and search engines, so seeing annotations there would be terrible. Additionally, the tooltip could conflict with tooltips on the original page, like editing tooltips or native annotation functionality (though it would be nice if everyone just used hypothesis 😁)
Active if there are prior annotations
With this one, it's not active everywhere, but not much better. The same timeline problem applies, and I frequently find myself being the first one to leave annotations. It enhances the social aspect, but purely as an annotator, I don't think anything is gained
My idea: heuristics
There are certain qualities of content that can effectively determine if
The most obvious heuristic is meta tags, which are a blessing. For articles to embed appropriately in sites like discord, twitter, and slack, (as well as SEO?) websites add meta tags which can give some insight into the content. Specifically, the og:type property with the content of article seems to be widespread among the content I annotate, and is standardized. In my opinion, this is the best of both worlds. Hypothesis, by design, is unlikely to be intrusive in articles, and in almost all cases it will already be active when the user goes to annotate.