4 ms·

Tested this on pg essays, the social comments is a pretty cool feature, I initially thought you would also surface the footnotes, similar to gwern site (one sit

Tested this on pg essays, the social comments is a pretty cool feature, I initially thought you would also surface the footnotes, similar to gwern site (one site I doubt I'd ever use a reader mode on). Lovely tool, in terms of things I would pay for, the actual reader mode is not really the thing I would pay for. What I would pay for would be, highest to lowest: 1. An awesome search engine that can search text I read, think being able to search text on the sites in your browser history, there was a service that did this called Memex, and Rewind seems like it's trying to do something similar. 2. Linked texts, e.g PG might say X then Naval says something in the same area. 3. Reminders about articles I read x period of time and might want to re read. I'll think of other features I would be willing to pay for.

phgn 4y ago
Thank you for the feedback, really interesting ideas. > being able to search text on the sites in your browser history I'm curious, have you used this in Memex when it was still supported? > Linked texts, e.g PG might say X then Naval says something in the same area. Do you mean direct hyperlinks and backlinks between articles? So if the Naval post references a PG essay, you'd see that while reading the PG essay? > Reminders about articles I [...] might want to re read How should it decide which article you want to re-read?
- Whiteshadow12 4y ago
  Ignore the Linked text idea, in theory it is solved by the Memex like feature. It might lead to an over engineered trap. Regarding the Memex like feature I did use it a lot, it worked like full text search for the HTML, they had a sync service that worked by importing your search history, I'm guessing they downloaded the HTML and stored it, so they could search against it. My usage of the service is a couple of years old so things might have changed, at the time it would crash when I did imports. You already have search on highlights and full text on them, so maybe the Memex like feature isn't needed. > How should it decide which article you want to re-read? It could decide based on other users who also saved the article, this might require storing the data on your servers, maybe number of highlights made or just ask the users when they add it to their queue. Nice extension, the more I use it the more I like it. I've even disabled Firefox reader mode in favor of this. Hope you find a way to monetize, play around with other monetization strategies other than good-better-best or one size fits all price point. Maybe usage based on the queue? There might be heavy crowds of users who would have a higher willingness to pay for example the "Second Brain" crowd. Links Memex Site: https://memex.garden/ https://memex.garden/ Recent Pricing Guide I've been working through: https://news.ycombinator.com/item?id=33502567 https://news.ycombinator.com/item?id=33502567 Priceintelligently posts on hn: https://news.ycombinator.com/from?site=priceintelligently.com https://news.ycombinator.com/from?site=priceintelligently.co... Some Lago Blog posts on pricing: https://www.getlago.com/category/product-billing https://www.getlago.com/category/product-billing
  - phgn 4y ago
    Yeah, I'm wondering why Memex disabled the search feature. It obviously didn't work out for them. > It could decide based on other users who also saved the article That's a cool idea, I'll think about it. Maybe it doesn't even need to be other Unclutter users. I already ingest almost every Hacker News post for the social annotations feature and there are APIs for web traffic stats. Thank you for the links on pricing, that's very helpful. At least this Show HN has convinced me to keep trying for a bit longer :) If you have any other feedback later, just email me, open an issue, etc. Thanks!
kidlj 4y ago
[dead]
shermix011 4y ago
1. There's one called Operand.ai which does exactly that.
- phgn 4y ago
  Interesting. Do you use Operand regularly?
- Whiteshadow12 4y ago
  I love the internet this is perfect thank you. Only issue is no Firefox for now.

phgn 4y ago

Thank you for the feedback, really interesting ideas. > being able to search text on the sites in your browser history I'm curious, have you used this in Memex when it was still supported? > Linked texts, e.g PG might say X then Naval says something in the same area. Do you mean direct hyperlinks and backlinks between articles? So if the Naval post references a PG essay, you'd see that while reading the PG essay? > Reminders about articles I [...] might want to re read How should it decide which article you want to re-read?

Whiteshadow12 4y ago
Ignore the Linked text idea, in theory it is solved by the Memex like feature. It might lead to an over engineered trap. Regarding the Memex like feature I did use it a lot, it worked like full text search for the HTML, they had a sync service that worked by importing your search history, I'm guessing they downloaded the HTML and stored it, so they could search against it. My usage of the service is a couple of years old so things might have changed, at the time it would crash when I did imports. You already have search on highlights and full text on them, so maybe the Memex like feature isn't needed. > How should it decide which article you want to re-read? It could decide based on other users who also saved the article, this might require storing the data on your servers, maybe number of highlights made or just ask the users when they add it to their queue. Nice extension, the more I use it the more I like it. I've even disabled Firefox reader mode in favor of this. Hope you find a way to monetize, play around with other monetization strategies other than good-better-best or one size fits all price point. Maybe usage based on the queue? There might be heavy crowds of users who would have a higher willingness to pay for example the "Second Brain" crowd. Links Memex Site: https://memex.garden/ https://memex.garden/ Recent Pricing Guide I've been working through: https://news.ycombinator.com/item?id=33502567 https://news.ycombinator.com/item?id=33502567 Priceintelligently posts on hn: https://news.ycombinator.com/from?site=priceintelligently.com https://news.ycombinator.com/from?site=priceintelligently.co... Some Lago Blog posts on pricing: https://www.getlago.com/category/product-billing https://www.getlago.com/category/product-billing
- phgn 4y ago
  Yeah, I'm wondering why Memex disabled the search feature. It obviously didn't work out for them. > It could decide based on other users who also saved the article That's a cool idea, I'll think about it. Maybe it doesn't even need to be other Unclutter users. I already ingest almost every Hacker News post for the social annotations feature and there are APIs for web traffic stats. Thank you for the links on pricing, that's very helpful. At least this Show HN has convinced me to keep trying for a bit longer :) If you have any other feedback later, just email me, open an issue, etc. Thanks!

kidlj 4y ago

[dead]

shermix011 4y ago

1. There's one called Operand.ai which does exactly that.

phgn 4y ago
Interesting. Do you use Operand regularly?
Whiteshadow12 4y ago
I love the internet this is perfect thank you. Only issue is no Firefox for now.