Version 0.7.0 released. ▪
Version 0.7.0 of the FeedTree software for users and publishers is now available for download. This version collects a number of bug fixes and new features; a complete list of checkins is here. Some highlights:
- Proxy: New statistics and info in the web UI (screenshots)
- Proxy: Support for conditional HTTP GET for feeds being polled
- Proxy: All feeds are now formatted as RSS 2.0 on the client side for better compatibility with feed reading apps (this might be switched back to Atom 1.0 as adoption and toolkit support improve)
- Proxy: More robust handling of
contentfeed elements, so that RSS 2.0 fulltext feeds aren’t truncated for proxy users - Publisher: New one-time configuration script, as well as
/etc/init.d-style run control scripts for Linux installations - Publisher: New, more flexible format for
publisher.conf(based on JSON, see [141]) - Publisher: Debian package installation option (see downloads page)
(It is strongly recommended that all current users download the new version; please report any issues to the mailing list or the developers.)
It’s worth noting that the website has received some attention as well; the new FeedTree for users and for publishers pages give a high-level schematic overview of the problems FeedTree solves.
Above: Diagrams from the new introduction pages, inspired by the original FeedTree poster.
February 20th, 2006 at 2:31 pm
I use Firefox’s livebook marks to manage all of my feeds. Please make a Firefox extension that would have Live Bookmarks update via FeedTree.
February 20th, 2006 at 2:41 pm
Actually, no plugin is required. Install the FeedTree client proxy, and then configure Firefox to use your proxy for its browsing. All Live Bookmark traffic should now be FeedTree-enabled. (I don’t have setup instructions handy, but it should be very similar to Thunderbird, which we already have some documentation for.)
February 20th, 2006 at 5:07 pm
FeedTree Gets Slashdotted
Feedtree, the P2P feed updating service, just got Slashdotted. The service is still a bit too geeky to be a consumer play, but I’m interested in everything that’s collaborative and/or P2P.
…
February 20th, 2006 at 5:09 pm
I had expected FeedTree to be RSS client (like Bit Torrent and other P2P stuff). I am noticing a bug though. Whenever I try to directly view an RSS feed in Firefox it tries to download it rather than display it like it used to do.
By the way, is it possible to run FeedTree as a service on Windows XP? As nice as FeedTree is for RSS, I do not want to have to navigate to the .exe file and run it everytime I want to use Firefox, and it is using up precious task bar real estate running as a windows application.
February 20th, 2006 at 8:01 pm
Richard: How did Firefox display your feeds before? The FeedTree proxy tries to be good about setting the correct MIME type on feeds it serves up to the user. As for the service, I haven’t yet had time to invest the engineering effort in wrapping the proxy as a Windows service. I’m aware of tools like Java Service Launcher and Java Service Wrapper, just haven’t gotten around to choosing one and integrating it.
February 21st, 2006 at 8:00 pm
Dan, when I would go to say, http://www.somesite.com/rssfeed.rss without using FeedTree as a proxy, Firefox would display the RSS feed like any other XML file but when I did that using FeedTree as a proxy, Firefox opened up the download window and asked me whether I wanted to save the feed as a file or view it with a program. Live Bookmarks were unaffected and worked the same as before, but I could not view feeds to see the XML markup, and having a website of my own, that is a problem for me.
February 21st, 2006 at 8:12 pm
What’s happening is that FeedTree is generating feeds with the current “correct” MIME type for RSS feeds—that is, application/rss+xml—instead of whatever the server sent. I suspect you were able to get by before because your site was serving up RSS as text/xml or something else that Firefox isn’t afraid of.
In a perfect world, all feeds would be using a distinguishing MIME type; this would make the FeedTree proxy’s work a lot easier. Right now, since there’s no widespread adoption of the feed MIME types, it has to attempt to parse anything that even smells like XML. If the MIME were more widespread, FT could switch on the MIME alone and avoid wasting a lot of time parsing non-feed XML.
Despite all that, Dave Winer is saying to use text/xml anyway, because he likes to be able to see XML in his browser, too. So I guess you’re not alone.
In a pinch, couldn’t you just save the file to the desktop and take a look at it in your favorite text editor? The experience wouldn’t be dramatically different than reading the raw XML in the browser.