Monday, April 20, 2009
FeedBlitz's RSS service has added a "FeedBurner Compatibility Mode" option to simplify the migration process away from FeedBurner. When enabled, third party blogging extensions and redirects that are hard coded for FeedBurner should also work for FeedBlitz, minimizing configuration headaches for publishers and eliminating potential confusion for subscribers.
Background
Many feed-based extensions to blogging software and services assume that, if a third party service is being used, it is FeedBurner. As such, they are often highly specific to FeedBurner, which can cause problems for publishers when they want to switch to a different service. Changing the settings can be a technical challenge (heck, even finding where those settings are managed can be something of an ordeal).
As the FeedBlitz RSS publisher base has grown over the least few weeks we have seen a number of issues related to this type of problem, and while it can be solved by changing configurations this can be quite a burdensome process for the 99% of bloggers not versed in web server arcana, HTTP protocols and regular expression syntax. Publishers and bloggers feel, rightly, that when switching RSS services "it should just work." FeedBlitz's FeedBurner Compatibility Mode brings that ideal another step closer to reality.
Dizzying Recursion
The primary issue is with plugins such as FeedBurner's own FeedSmith plugin for WordPress, and for hand built Apache-style mod-redirect statements (again, often with self-hosted WordPress sites). This class of web-server modification causes requests from an RSS reader to be redirected automatically to the publisher's FeedBurner feed, so all requests are served from the same location. This ensures that the publisher sees all their feed metrics in one place, and no matter which feed URL the end user tries to subscribe to, they always end up at FeedBurner, giving a consistent user experience.
When changing to FeedBlitz (or any other service) this becomes a problem. When FeedBlitz tries to access the blog's RSS feed it too is redirected by the blog's server back to FeedBurner. If FeedBurner is redirecting back to FeedBlitz a recursive loop is set up, and feeds don't update correctly as the RSS readers get sent running around in circles. Even if FeedBurner doesn't redirect back to FeedBlitz, FeedBlitz never gets access to the blog's original feed, and so is not only limited to showing what's available through FeedBurner, but also suffers from any delays that FeedBurner has in reflecting changes made to the underlying blog. Yuck!
FeedBurner itself obviously doesn't have these problems, even when such a redirect is in place. It clearly isn't being redirected. How does that work? It's simple: The changes made by FeedSmith and similar redirecting strategies specifically exclude FeedBurner from the redirection. As a result, FeedBurner can access the blog's source RSS feed without getting itself tied up in knots.
While it is possible to edit filters, remove plugins to enable FeedBlitz to work the same way, as observed at the top of this post the process can be difficult, burdensome and adds to the risk of human error breaking feeds for everyone. In short, it's complex work for most to do. It adds a lot of friction and risk to the process of evaluating FeedBlitz as a FeedBurner alternative. And that's just not right.
Making It Work
Problem solved. FeedBlitz's "FeedBurner Compatibility Mode" ensures that the exclusions these tools use for FeedBurner also apply to FeedBlitz - all automatically. No need to change filters, modify your blog's settings or learn regular expressions. It just works. For this reason alone, FeedBurner Compatibility Mode is enabled by default for all new feeds. It can be switched off (or back on) using the RSS / Settings screen.
Better yet, FeedBlitz's FeedBurner Compatibility Mode does not affect FeedBurner in any way, and so it is possible to view and test the two services side by side, allowing publishers full control over the timing and details of their eventual transition.
This new setting should work for any plugin or service checking user agent strings to decide how and when to apply exclusions. Even without "FeedBurner Compatibility Mode" being enabled, FeedBlitz will always break out of any recursive call it detects and return any cached data it has (if any).
The Exception That Proves the Rule
Once recursive setup that this can't help with is if a publisher sets up set up FeedBlitz as the source URL for FeedBurner, and FeedBurner as the source for FeedBlitz (this has happened a couple of times). While clearly circular in definition, FeedBurner and FeedBlitz cache feeds to serve to visitors. The recursion isn't in real-time; there are no HTTP redirects for FeedBlitz to detect, trap and break out of.
Instead, each service updates its cached copy periodically, each using the other as a reference. Your feeds will be served fine, but over time (several hours) they'll come with an increasing amount of baggage (e.g. an increasing number of flares in the same article) as they slowly update themselves. If you see this happening, it's not a bug, just a very slow infinite loop that has been set up.
Fortunately, assuming you are using both services to test / migrate to FeedBlitz while still temporarily serving your existing subscriber base at FeedBurner, it's also very easy to fix. Here's how:
FeedBurner Compatibility Mode - It Just Works
With the introduction of FeedBurner Compatibility Mode, FeedBlitz has eliminated some of the more technically challenging steps a publisher may need to take to switch services.
FeedBlitz - It just works! Start a 30-day trial today.
Background
Many feed-based extensions to blogging software and services assume that, if a third party service is being used, it is FeedBurner. As such, they are often highly specific to FeedBurner, which can cause problems for publishers when they want to switch to a different service. Changing the settings can be a technical challenge (heck, even finding where those settings are managed can be something of an ordeal).
As the FeedBlitz RSS publisher base has grown over the least few weeks we have seen a number of issues related to this type of problem, and while it can be solved by changing configurations this can be quite a burdensome process for the 99% of bloggers not versed in web server arcana, HTTP protocols and regular expression syntax. Publishers and bloggers feel, rightly, that when switching RSS services "it should just work." FeedBlitz's FeedBurner Compatibility Mode brings that ideal another step closer to reality.
Dizzying Recursion
The primary issue is with plugins such as FeedBurner's own FeedSmith plugin for WordPress, and for hand built Apache-style mod-redirect statements (again, often with self-hosted WordPress sites). This class of web-server modification causes requests from an RSS reader to be redirected automatically to the publisher's FeedBurner feed, so all requests are served from the same location. This ensures that the publisher sees all their feed metrics in one place, and no matter which feed URL the end user tries to subscribe to, they always end up at FeedBurner, giving a consistent user experience.
When changing to FeedBlitz (or any other service) this becomes a problem. When FeedBlitz tries to access the blog's RSS feed it too is redirected by the blog's server back to FeedBurner. If FeedBurner is redirecting back to FeedBlitz a recursive loop is set up, and feeds don't update correctly as the RSS readers get sent running around in circles. Even if FeedBurner doesn't redirect back to FeedBlitz, FeedBlitz never gets access to the blog's original feed, and so is not only limited to showing what's available through FeedBurner, but also suffers from any delays that FeedBurner has in reflecting changes made to the underlying blog. Yuck!
FeedBurner itself obviously doesn't have these problems, even when such a redirect is in place. It clearly isn't being redirected. How does that work? It's simple: The changes made by FeedSmith and similar redirecting strategies specifically exclude FeedBurner from the redirection. As a result, FeedBurner can access the blog's source RSS feed without getting itself tied up in knots.
While it is possible to edit filters, remove plugins to enable FeedBlitz to work the same way, as observed at the top of this post the process can be difficult, burdensome and adds to the risk of human error breaking feeds for everyone. In short, it's complex work for most to do. It adds a lot of friction and risk to the process of evaluating FeedBlitz as a FeedBurner alternative. And that's just not right.
Making It Work
Problem solved. FeedBlitz's "FeedBurner Compatibility Mode" ensures that the exclusions these tools use for FeedBurner also apply to FeedBlitz - all automatically. No need to change filters, modify your blog's settings or learn regular expressions. It just works. For this reason alone, FeedBurner Compatibility Mode is enabled by default for all new feeds. It can be switched off (or back on) using the RSS / Settings screen.
Better yet, FeedBlitz's FeedBurner Compatibility Mode does not affect FeedBurner in any way, and so it is possible to view and test the two services side by side, allowing publishers full control over the timing and details of their eventual transition.
This new setting should work for any plugin or service checking user agent strings to decide how and when to apply exclusions. Even without "FeedBurner Compatibility Mode" being enabled, FeedBlitz will always break out of any recursive call it detects and return any cached data it has (if any).
The Exception That Proves the Rule
Once recursive setup that this can't help with is if a publisher sets up set up FeedBlitz as the source URL for FeedBurner, and FeedBurner as the source for FeedBlitz (this has happened a couple of times). While clearly circular in definition, FeedBurner and FeedBlitz cache feeds to serve to visitors. The recursion isn't in real-time; there are no HTTP redirects for FeedBlitz to detect, trap and break out of.
Instead, each service updates its cached copy periodically, each using the other as a reference. Your feeds will be served fine, but over time (several hours) they'll come with an increasing amount of baggage (e.g. an increasing number of flares in the same article) as they slowly update themselves. If you see this happening, it's not a bug, just a very slow infinite loop that has been set up.
Fortunately, assuming you are using both services to test / migrate to FeedBlitz while still temporarily serving your existing subscriber base at FeedBurner, it's also very easy to fix. Here's how:
- Go to RSS / Settings at FeedBlitz.
- Ensure that you have FeedBurner Compatibility Mode enabled.
- Update FeedBlitz to point to your blog.
- Save the changes.
FeedBurner Compatibility Mode - It Just Works
With the introduction of FeedBurner Compatibility Mode, FeedBlitz has eliminated some of the more technically challenging steps a publisher may need to take to switch services.
FeedBlitz - It just works! Start a 30-day trial today.
Labels: features, FeedBlitz, feedburner compatibility, RSS
|
2 Comments:
Is Feedblitz going to get a Feedburner like upgrade? Are you Google owned yet? :-)
I want to hire someone to help me convert from Feedburner to FeedBlitz. I am nervous about doing it myself, because I don't want to lose subscribers. (I lost about 5,000 subscribers in converting from FeedBlitz to Feedburner. I know, I know, I shouldn't have done it!)
I am willing to pay someone to help me. I just need someone who really knows what they are doing. Thanks.
Post a Comment
Note: Only a member of this blog may post a comment.
<< Home