Martijn/Anthony:

The thing that prompted me to write my last comment is my worry that someone who does not know the technical limitations due to RSS/ATOM, when reading the summary/description of the idea would get the wrong impression and have unreasonable expectations.

Hopefully I am wrong, maybe those people would also not use uservoice and never get here.

Pavel:

I could clarify that when I said ‘The Old Reader should store every item I have ever read’ I meant ‘ever read in The Old Reader’. But I thought this was clear enough. I never requested to be able to read what The Old Reader doesn’t know about.

Google Reader never went and found feeds on its own so it would never index items before someone subscribed to a feed. It could only read back to the day you subscribed and that exact behaviour can be copied by The Old Reader.

Also note that you already went wrong in your first sentence about the technology. RSS/ATOM feeds do not only expose ‘the last few posts’. In fact, I am subscribed to several feeds that go back to the beginning of the site. This is completely up to the website owner.

Pavel: we do realize that it's impossible to retrieve items posted long time ago due to the RSS protocol limitation.. However, for those items that we read in TheOldReader right now, we'd like to be able to find these items and read them again here on the TheOldReader after a year or several years from now. I.e. if it is possible to "start" storing all the posts from now on, we definitely would like it to be implemented. Thank you.

This is somewhat impossible to implement. The Old Reader can definitely store all the posts starting today, but it is not possible to go back.

Some technical details:
1) RSS/ATOM feeds only post the last few (10-20) posts. The clients (Google Reader, The Old Reader, etc.) check these pages every so often and see new posts. They then store these in their own storage. Google Reader has been around for a long time, and thus it was able to see all the posts throughout history and kept them all. The Old Reader is just beginning to look at the feeds, and thus cannot go back in history and find posts, especially since some of the websites no longer exists.

2) Google Takeout does not expose all the posts you read in the past, only the ones you "liked" and "starred" (based on what I see in the zip file). The Old Reader could import these, and reconstruct some history (might not be trivial), but there will be holes in history.

+1. Definitely very useful to search in all RSS feed history.

+1 I think that you should be able to able to import a list of read things from google reader and be able to search it.

Having a searchable RSS history was one of my favorite features of google reader. I read a lot of electronics / DIY websites and sometimes I become interested in something that I have glanced over a couple of weeks / months ago.

Thanks for the input tasos, that’s exactly what I found too. Most of the services touted as a GReader replacement do not support search (like Feedbin and NewsBlur) or do not keep feed items for ever (like The Old Reader).

A couple of self-hosted systems do support search (Fever, selfoss, Tiny Tiny RSS) but this means you have to be able to have a server running 24/7. Something a lot of GReader refugees will not be able to do. Maybe The Old Reader can look at some of those systems for inspiration?

ReplaceReader (http://replacereader.com/) currently shows The Old Reader as the fourth best replacement, fighting Bloglines for this place. Being the first of the many replacements to figure out backwards search would certainly boost user confidence!

Yes! This is exactly the thing I need for my "new google reader". I was testing other web-based rss readers (that are free) and none of them have this feature like google reader (... I know it's google). But I will definitely donate for this!!