3 ms·

From the article > no one ever has to know what feeds you subscribe to. Whoever runs the server knows you're reading that feed from their server logs.

fredsted 12y ago
Note "feeds" is plural. There's no way for a server admin to get a list of other sites you subscribe to, unlike a centralized app like Google Reader.
- ciderpunx 12y ago
  Yep, that's a fair point.
onli 12y ago
True, but it is quite possible that is you yourself. If you selfhost your reader, you control the logs. If you use a reader, the server admin might now what feeds you have. But the server who hosts the feed (the blog) has no idea, tracking doesn't work easily - that is still an improvement.
- verbin217 12y ago
  I think he means the server hosting the RSS feed and not the server hosting your RSS tool (like the late Google Reader).
  - onli 12y ago
    In that case, the server hosting the feed does now nothing about you as soon as you are not the sole user on a server hosting a feed (I sure wasn't sure how to interpret that, so I tried to cover both cases)
    - ciderpunx 12y ago
      The server would know your IP address, and most likely your user agent and referrer by looking at their weberver log though, right? cf. eg. : https://panopticlick.eff.org/index.php?action=log&js=yes https://panopticlick.eff.org/index.php?action=log&js=yes
      - onli 12y ago
        Which server exactly? If you use a feedreader, it is not you or your browser that fetches the page, it is the feedreader itself. So a site only knows that their feed was fetched once from a feedreader, not who or how manye people read that feed.
icebraining 12y ago
That would be a poor assumption. How can it distinguish a single user's reader from a shared service like NewsBlur, which will request a feed on behalf of hundreds of users? The server knows that some aggregator is pulling its feed, but it has no way to tie that to a user. There are other ways, though: images in the feed that call back to the server, for example.

fredsted 12y ago

Note "feeds" is plural. There's no way for a server admin to get a list of other sites you subscribe to, unlike a centralized app like Google Reader.

ciderpunx 12y ago
Yep, that's a fair point.

onli 12y ago

True, but it is quite possible that is you yourself. If you selfhost your reader, you control the logs. If you use a reader, the server admin might now what feeds you have. But the server who hosts the feed (the blog) has no idea, tracking doesn't work easily - that is still an improvement.

verbin217 12y ago
I think he means the server hosting the RSS feed and not the server hosting your RSS tool (like the late Google Reader).
- onli 12y ago
  In that case, the server hosting the feed does now nothing about you as soon as you are not the sole user on a server hosting a feed (I sure wasn't sure how to interpret that, so I tried to cover both cases)
  - ciderpunx 12y ago
    The server would know your IP address, and most likely your user agent and referrer by looking at their weberver log though, right? cf. eg. : https://panopticlick.eff.org/index.php?action=log&js=yes https://panopticlick.eff.org/index.php?action=log&js=yes
    - onli 12y ago
      Which server exactly? If you use a feedreader, it is not you or your browser that fetches the page, it is the feedreader itself. So a site only knows that their feed was fetched once from a feedreader, not who or how manye people read that feed.

icebraining 12y ago

That would be a poor assumption. How can it distinguish a single user's reader from a shared service like NewsBlur, which will request a feed on behalf of hundreds of users? The server knows that some aggregator is pulling its feed, but it has no way to tie that to a user. There are other ways, though: images in the feed that call back to the server, for example.