reading-web-content.org
1 Reading Web Content
1.1 Background
I still find reading Web content in emacs to be way more efficient than in modern browsers like Chrome with ChromeVox enabled. Chrome+ChromeVox is what I use for rich Web Applications; but when it comes down to reading straight content ranging from technical documentation to news and current affairs, I find that I can read a lot more content in a fixed amount of time with Emacs+Emacspeak than I can with the Chrome+ChromeVox combination.
1.2 Welcome EWW: Emacs Web Wowser
The author of GNUS recently added package EWW (Emacs Web Wowser) to The Emacs repository — see his announcement. Emacspeak 37.0 included a small addition called shr-url that leveraged his earlier shr package; I've now added support for EWW in the Emacspeak source tree. Note that EWW is in the Emacs source tree, i.e. it will be part of Emacs 24.4, but you can use it now if you build your own version of Emacs, or obtain an Emacs that is build from the Emacs source repository.
EWW and SHR are interesting because they both leverage libxml2 to parse the incoming HTML. This is way faster than the native elisp browsing used in Emacs/W3. In my personal opinion, it also opens up more possibilities than Emacs/W3M with respect to manipulating and filtering content from elisp --- something that helps create a better reading experience when using Emacspeak.
1.3 EWW Tips And Tricks
I've mailed out a small patch to package eww that facilitates implementing such higher-level commands; for now, you can find that patch in emacspeak/lisp/patches in the Emacspeak svn repository. With that patch in place, you can do the following to efficiently filter popular news sites such as the New York Times (I use http://mobile.nytimes.com) or CNN (http://cnn.com).
When visiting these and other content-heavy sites, try:
- A
- Filter by attributes,
- E
- Filter by elements
- R
- Restore original contents
To flexibly obtain multiple views of a Web site.
