|
|
|
|
|
|
by jacamera
2181 days ago
|
|
|
Not a great first couple of impressions! 1. Safari uses a completely non-standard extension model and right now I'm the only developer so we're Chrome and Firefox only for the time being. Definitely sucks. 2. That's great feedback on that modal. Seriously, it never even crossed my mind that we don't explain why the extension is required before displaying it. Also that the "Add to Chrome" button makes it seem like it will trigger an install instead of bouncing you over to the Chrome store where you can see screenshots and a description of how it works before choosing to install. We really need to fix that ASAP. Re tracking: When the extension is triggered (limited permissions, it only runs when you click the icon) it runs a script on the web page that tries to identify the primary text of the article. Once identified, we mark the individual words as having been read, starting at the top left of the viewport. Since we keep track of individual words you can scroll around and read out of order and the tracking will still work properly. This process runs on a timer that allows for a pretty fast reading speed (probably around 500-600 wpm) but doesn't allow a user to just sit on the page for a long time or scroll right to the bottom. As you point out you can cheat it by scrolling very slowly through the entire article. Still waiting on that mind-reading API! |
|
|
Re extension: I'm not clear on why an extension is needed for reading on readup.com (as opposed to on publishers' sites). Seems to me you can do all the tracking just fine with regular sandboxed JavaScript running on the web page? To be clear, when I'm not logged in or on an unsupported browser, I'm presented with two options: "Read it on Readup" (with a "Get Started" button) and "Continue to publisher's site", the former apparently being the recommended path; when I'm logged in there's only the "To read on Readup, add the Chrome extension" modal, but if I understood what you track correctly, reading on Readup shouldn't require any extra capabilities, whereas reading on publisher's site should. So in theory a user should be able to read everything on readup.com without using an extension, just like they would on instapaper.com or feedly.com (but with some added tracking). Is that not the case simply because you haven't had the time to develop native tracking, or did I miss some fundamental limitations?
Re timer: 500-600 wpm is pretty fast on average but I can certainly beat that when reading information-sparse content (e.g. most digital magazines) or when I consciously try to read fast, and I know people who read way faster than me... Also note that some people can maintain a high comprehension level even at high speed while others struggle to comprehend even when reading word by word. I guess your compromise may be okay, but it's certainly rather crude (not that I have a better idea).
Btw, Safari 14 is adding WebExtension API support.[1] (You're probably already aware of this but doesn't hurt to share.)
[1] https://developer.apple.com/documentation/safariservices/saf... (documentation is crap though, at a glance)