Post AO6QeDMcoJb9TisWLw by adamsdesk@mastodon.technology
 (DIR) More posts by adamsdesk@mastodon.technology
 (DIR) Post #AO4z51jVAe3DLjcET2 by mike@fosstodon.org
       2022-09-30T04:56:21Z
       
       0 likes, 1 repeats
       
       Hey look! I word vomited out another blog post for something that's probably a stupid idea.https://mikestone.me/tell-me-a-story#100DaysToOffload
       
 (DIR) Post #AO56xhY4cHS0M5Hv4S by sayanarijit@fosstodon.org
       2022-09-30T06:24:42Z
       
       0 likes, 0 repeats
       
       @mike Why not plugins like Read Aloud?
       
 (DIR) Post #AO58si1FdgfWH7gOsC by fedops@fosstodon.org
       2022-09-30T06:45:51Z
       
       0 likes, 0 repeats
       
       @mike I've used 'links -dump' to save formatted text non-interactively to archive web pages to my notes. Graphics vanish that way, and JavaScript is never executed. Might be worth a try.
       
 (DIR) Post #AO5vI59rXf0X4vsKrw by mike@fosstodon.org
       2022-09-30T15:48:36Z
       
       0 likes, 0 repeats
       
       @sayanarijit That's fine if I want to stay at my desk, but I didn't see any of those extensions (in my admittedly brief "research" on the subject) that allowed the audio to be saved to MP3. I listen to these files when I'm away from my desk most of the time. The other bit that's a little bit icky is even though they're open source, they're using TTS engines like Wavenet and Polly. The fact that Mimic3 is local and private is a definite benefit in my opinion.
       
 (DIR) Post #AO5xzfQQmDohv5XHBg by mike@fosstodon.org
       2022-09-30T16:18:45Z
       
       0 likes, 0 repeats
       
       @fedops Interesting idea, but won't work out of the gate. It leaves in SO MUCH text that's not part of the article itself. One of the articles I tested left every option in a toggle menu in the text that was output. It could (in theory) be used to automate the text extraction copy/paste part, but I'd still have to go through and manually strip out so much irrelevant material from the file(s) before I could do the TTS conversion. That seems to be the biggest hurdle to this being easy.
       
 (DIR) Post #AO62Oflb2nse2DcB60 by fedops@fosstodon.org
       2022-09-30T17:08:12Z
       
       0 likes, 0 repeats
       
       @mike yeah, and I think a good part of it is that the big sites do everything to keep people from scraping their content.I've just given up on those sites. Most of the information can be gotten from other outlets. They can eff right off...
       
 (DIR) Post #AO63BCz6qNaIxU4xPc by clay@quanta.wiki
       2022-09-30T17:16:34.284142Z
       
       0 likes, 0 repeats
       
       @mike You're pointing out one of the biggest problems with the web today. Chaos over HTML. There's no way to pull just the raw "content" from an arbitrary webpage. Advertisers love this because it keeps them from being bypassed. So there's little chance the free (for profit) market will solve this. It's also one reason big business (including Google especially) doesn't want to peruse a "Semantic Web"...because if information on websites is categorized, and easily parsed by computers (not to render, but to comprehend), then it enables competitors to Google and Advertising interests to gain power.The world badly needs some kind of XML (or broader adoption of XHTML/HTML5 tags) that websites can publish to which are designed specifically to allow the main (non AD) content to be extracted. Finding the way to motivate the world to cooperate in this endeavor would be like herding cats.
       
 (DIR) Post #AO6QeDMcoJb9TisWLw by adamsdesk@mastodon.technology
       2022-09-30T21:39:02Z
       
       0 likes, 0 repeats
       
       @mike Good post. It seems to me you are overwhelmed. Try to reduce if you at all can.