The FeedZero Blog

The first update in a while! There’s a few minor backend changes but the biggest news is significant improvements to the system that reads our RSS feeds, hopefully permanently correcting the issue that we had when feeds would seemingly either not update, or update many days later.

This should mean that almost as soon as your favourite sites have updated their RSS feed, the new items will appear in your RSS feed.

We’re recommencing work on Feedzero after around 4 months of quietness (due to commitments on some other projects) and have a bunch of exciting plans, including improvements for new users, a new view mode, the ability to mark items as read (!!) and a few more bells and whistles. If you have any suggestions, please let us know.

A new feature from a recent update of Firefox that you might have missed is the ability to subscribe directly to feeds with FeedZero from the Firefox browser:

To add FeedZero as an option to your Firefox feed reader list, simply visit the manage feeds page on FeedZero - look in the left side column for the module titled “Add feeds directly from Firefox” and follow the instructions!

There’s been some issues with article timestamps being off recently; today we discovered that the server clock was off by several hours, which was causing a few problems. This has been corrected so article timestamps should now better reflect the actual space-time continuum.

Another fix has also gone into place for duplicate articles; if you notice any more please let us know so we can investigate.

There’s also been some back-end improvements for feed retrieval, resulting in better performance in gathering feed data.

For those that are new to terms like “RSS”, “feed”, and combinations of the two, we have a simple guide to RSS which explains some of the basics, including “What The Hell Is It?” and “How Will It Help Me?”. If you have any questions about RSS feel free to drop them in the comments and we’ll either answer them there or modify our guide as required!

We’ve also got a simple video that provides a visual overview for those that prefer their input with more of an audio/visual flavor.

There’s a known issue at the moment with some feeds not updating correctly - if you’re having problems getting new content don’t panic, it should be there soon. It is due to an issue with the service that retrieves the RSS feeds from the source dealing with the volume of feeds we have now; a fix is in the works.

Update 2008-04-10 - this has been fixed.

A small update today, just fixing a minor problem with the web server returning “page not found” errors as 500s, not 404s. This appears to have been affecting Google’s ability (or willingness) to index the feedzero.com domain, so we’re hoping this will fix that issue.

We have just uploaded our first update for FeedZero.com . This week we have concentrated on polishing up some of the more obvious bugs. Here’s a quick summary of the changes:

• Fixed ‘Contact us’ page
• Fixed the collapsing/expanding of articles on rating change (to stop the page jumping around when rating)
• Better handling for broken HTML in articles
• Better handling for timestamps on articles
• Don’t change state of manually collapsed/expanded articles when rating
• Improved support for RDF feeds
• Fixed a bug resulting in multiple feed subscriptions from a single site
• Separation of static content
• Fixed a bug preventing the paging system from working in Safari
• Performance improvements for the back-end feed downloader service.
• Use ‘pretty’ URLs like /articles instead of /ViewArticles/Index.rails
• Few tweaks for IE6 and IE7 behaviour.

We have also added linked this blog off the main FeedZero site. If you have any issues with any of the changes, please feel free to let us know in the comments!

Our next release should have a “read/unread” feature to keep track of which articles you have previously seen.

Welcome to the development blog for Feedzero, the Bayesian RSS filter!

Feedzero attempts to make your digital life a lot easier by helping to mitigate the increasingly common problem of information overload. As you try to keep up with more and more websites, you might find yourself subscribed to many different RSS feeds, with hundreds or even thousands of pieces of news coming at you on a daily basis.

Using Bayesian filtering - the same technology that is used to help combat spam emails - Feedzero learns what sort of information you like and what sort you don’t like. All it takes is a bit of training, and then shortly afterwards you’ll be receiving a customised page that contains information tailored specifically and carefully to your preferences.

Feedzero has just recently been unveiled to a small group of users and we’re currently making a number of changes based on their feedback - so please stay tuned for updates in the very near future.

We’ll be keeping this blog updated with the latest goings-on with the Feedzero website, so please stay tuned for more information as it becomes available.