4 ms·

Your list has some interesting ideas, but I think that (10)ArticleBody has been executed very well by the bookmarklet ViewText at http://viewtext.org/article?ur

by J3L2404 16y ago

Your list has some interesting ideas, but I think that (10)ArticleBody has been executed very well by the bookmarklet ViewText at http://viewtext.org/article?url=http%3A%2F%2Fviewtext.org%2F http://viewtext.org/article?url=http%3A%2F%2Fviewtext.org%2F I'll add the list: A FF and/or Safari extension to emulate the text focus that occurs when double tapping on a column of text on iOS devices. If you haven't seen it, it zooms the page to the width of the text column and is one of the best features of the iPad. It would probably be best implemented through a right click. I haven't played with extensions yet so I have no idea how difficult it would be, any thoughts?

ronnier 16y ago
Thanks for linking this. I've been working on ViewText as a side project. I'd love to hear feedback and ideas. Currently, you can request the content of an article in html, json, jsonp, xml, and pdf formats. It also works on RSS feeds.
- jacquesm 16y ago
  Ronnie, Do you want me to score your code relative to what I've been able to do with this so far on my test-set ?
  - ronnier 16y ago
    Sure, that's fine.
- mikeklaas 16y ago
  Might you be willing to share your code/approach? How does it differ from Readability's algorithm?
  - ronnier 16y ago
    I took readability's algorithm (in JavaScript) and ported it over to C#. It's not a direct port of course, as lots of things in C# are done differently than in JavaScript, and I didn't have the browsers DOM to work with. In addition, readability performs the processing while on the current page, leaving elements in place. For me, I just have the HTML string to work with. That's the biggest difference, otherwise I followed their idea on how to extract the main content.
jacquesm 16y ago
Thanks for that link, I'll run my test set through it. neat idea, that zoom!
- J3L2404 16y ago
  The bookmarklet wasn't reachable from the frontpage so I edited the link to one with the bookmarklet. I use it all the time, but the zoom would be better.

by J3L2404 16y ago

ronnier 16y ago

Thanks for linking this. I've been working on ViewText as a side project. I'd love to hear feedback and ideas. Currently, you can request the content of an article in html, json, jsonp, xml, and pdf formats. It also works on RSS feeds.

jacquesm 16y ago
Ronnie, Do you want me to score your code relative to what I've been able to do with this so far on my test-set ?
- ronnier 16y ago
  Sure, that's fine.
mikeklaas 16y ago
Might you be willing to share your code/approach? How does it differ from Readability's algorithm?
- ronnier 16y ago
  I took readability's algorithm (in JavaScript) and ported it over to C#. It's not a direct port of course, as lots of things in C# are done differently than in JavaScript, and I didn't have the browsers DOM to work with. In addition, readability performs the processing while on the current page, leaving elements in place. For me, I just have the HTML string to work with. That's the biggest difference, otherwise I followed their idea on how to extract the main content.

jacquesm 16y ago

Thanks for that link, I'll run my test set through it. neat idea, that zoom!

J3L2404 16y ago
The bookmarklet wasn't reachable from the frontpage so I edited the link to one with the bookmarklet. I use it all the time, but the zoom would be better.