EWW Is Not Eww (also, info books are great)

So I find it hard to read. Yes, some people are like that, but I have developed a technique to aid me on my readings: (Only for English, not for my Mother Tongue Hebrew, but that's another topic).

It is all based on a little Python script I wrote called Clipboard Speaker.

I mark a bunch of text, copy the text, and press a keybinding which starts the script. The script then reads the contents of the clipboad, then gives them to a text-to-speech engine that reads it. There are some additional minor complexities, but that is the main workflow.

Now let me expand on that.

The three keybindings I have set are, again, defined through the Gnome configuration UI:

• super-c - Takes the regular clipboard which is filled by usually pressing C-c.
• super-x - Takes the weird clipboard which is normally filled just by marking text.
• super-z - Stops the text-to-speech engine.

Normally you would like to mark a bunch of text and press super-x, without first pressing C-c and then super-c, but sometimes that does not work. I do not know why, as in the case with Firefox.

As an aside, Firefox has a Reader Mode which also has built-in text-to-speech functionality. It is admirable and I love that Mozilla chose to implement it, but I find it limited - one cannot control where each batch of text to-be-read starts and where it finishes. Usually whole paragraphs are chosen. Sometimes one wants to sit back while the text is zoomed in, but the paragraph slides off the viewport, so one cannot follow the text below a certain point.

Anyway, I want to read in Firefox, so I mark a bit of text, then press C-c (text copied to clipboard) s-c (script takes clipboard contents, starts the TTS engine and passes the text). I mark text, C-c s-c, mark text, C-c s-c.

Now, if I read something in eww, I have a cursor by default and I can comfortably move the point to the start of bit of text I want to read, press C-<SPC>, move point to the end of the bit of text, then press s-x. Easy.

By the way, a fancy thing I have added to the script's functionality, is that the text-to-speech engine keeps receiving new text in its stdin every time I press s-x or s-c, so you can mark text, press s-x or C-c s-c several times and each of those bits of text would be played in order.