Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

We built http://filr.io for the same purpose. You can follow feeds from various content sources. These feeds could be based on rss or any kind of content that changes over time.

Everything you read is tracked as read/clicked so that you can search using those filters. Also tagging and creation of own channel(invite only) is allowed.

Examples:

http://filr.io/channel/hackernews

http://filr.io/channel/Wikipedia-Picture-of-the-day

http://filr.io/channel/imdb-posters



With all these different ways of doing it, we're going to need some sort of syndication. It would need to be really simple though.


The problem with RSS is that the publisher needs to support it, and quite a few are actively moving away from it. For example, Wikipedia Picture of the Day RSS feeds don't work anymore: https://commons.wikimedia.org/wiki/Commons:Press_releases/Pi...

Also, considering the "fire-hose" nature of typical RSS feeds, there needs to be an intelligent layer that sits between it and the user for it to be useful.


> considering the "fire-hose" nature of typical RSS feeds, there needs to be an intelligent layer that sits between it and the user for it to be useful.

I have to disagree with that, I enjoy my RSS feeds uncensored, if a feed has too low of a signal to noise ratio then I unsubscribe. I have a few of my own filters set up in InoReader for stuff that I'm absolutely sick of hearing about, but none if it is filtered for me "intelligently".

There are plenty of services built on top of RSS that do filter and remix the content already, however.


newsblur has trainable filters.


Maybe we could use XML as a standard format.


Really simple syndication? Impossible.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: