While browsing a discord, someone posted a link to bionic reading. I posted it to
https://forum.level1techs.com/t/when-im-bored-i-tend-to-post-funny-cool-stuff/179005/721?u=bedhedd
I thought it would be interesting to spin it off to it’s own thread
Damn.
Our predictive machine brains sure are good at picking up patterns
Why does that help me? I’m not even dyslexic.
I really need this.
Yeah unfortunate they patented it
In europe
Bah humbug
They have a free API call tier if you guys really need it. I have no idea how fast you can consume it if it counts a processed word as 1 count of API call.
Before the disinformation age, this is undoubted useful in consuming vast amounts of information. These days, you could also increase the rate of disinformation consumption so its just another tool, rather than a new radical force in learning.
Does using it actually affect the US if they only trademarked it?
They trademarked it in CH and while it doesnt directly affect the US, to my understanding is that the Swiss government can complain to the US government via the State Dept., that a patent is being abused and the owners seek to litigate. Because the US like the Swiss, they will find the defendant and make them accountable somehow.
But don’t trademarks only prevent you from using their branding without permission? The core idea of applying bolding to text is not patented in the US right?
Is it really that easily applied? I thought it would be based on words they had worked out how to shorten. Like at a certain point you stop recognising them.
But if its just an amount of the word…
I know the world is clown town but generally pattets are supposed to appply to the implementation not the idea. So you should be able to repimplement it by either reverse engineering through clean room process or if it really is as simple just write novel code to do that.
No the rules I linked seem to be on someone’s guess. Tried looking up linguistics, but can’t seem to find how the technique works because their documentation is non-existant
I tried looking up the french patent
FR1755215
The number is on this page, but I don’t read/speak french
https://data.inpi.fr/brevets/FR3052587
Seems the api is public?
https://api.bionic-reading.com/convert/
I tried using on a epub, but it skipped chapters
Holy Hell - this is perfect! This would be so useful to me as someone who reads a lot of long-form articles online.
So I studied speed reading. I never practiced it to perfection but I can read at a pretty high rate and it works somewhat similarly to this. Children learn to read by spelling things out letter by letter. As an adult we learn to recognize entire words but a lot of people don’t do it at a very advanced level. You can actually learn to go to the next level which is to look at groups of words at once. I mean, the eye can physically see multiple words at the same time of course. But we have trained ourselves to do what a child does with single letters and just look at words sequentially. Speed reading is about looking at groups of words at once. The brain can handle it, trust me. There are techniques for looking at multiple lines at once as well (if that sounds crazy you’re not looking at multiple whole lines, you look at a vertical cross-section of lines like a striped RAID array. I can’t do that but apparently you can train yourself to be able to.
Anyway, bit of a tangent but what this tool is doing is acting as a great faciliatator to the first technique of speed reading. I just put a long article in there and it did indeed make it easier to read quickly and I don’t think I lost any information.
I want this and I want it now. But I don’t really want this to be another Grammerly (the world’s most successful key logger!) and send everything I’m reading to some third party service which will file all the information about me away. I just tried out their Chrome extension and that’s what it does. It identifies what it considers the important text of a page (not sure how) and sends it to their site API
Edit: I hit send to early.
I found the GIF above a few years ago. Since I found that quite interesting, I put some work into making a “txt speed reader” that would present a text file at 200 to 1000 WPM with fairly high contrast (blue on yellow for maximum visibility, off-tone orange on blue-ish black, etc.)
While it is “fun” to read at 650-ish WPM, it is VERY tiering to me. Have to see where I have that program (cobbled together in Processing)