If you are interested in knowing what happens with your content, who reads it and what impact it has then this is for you.
While some people say the days of RSS are numbered, the fact remains many people rely on RSS as a way of getting information, often via services like Google Reader or one of the Newsgator services. Many services use feeds. It is an excellent way to aggregate data from many different sources. That’s the basis of the sponsored posts section at AccMan. But when something goes wrong then you can be in trouble.
Earlier today I was pinged by a sponsor who said something like: ‘Heh – wassup? My feed has disappeared.’ That’s weird since I have not changed any of the code. Hopping over to the service provider desktop editor all seemed fine except for the missing feed data. The feed reference was as it should be, the feed itself was correctly validated. So what might it be? Coincidentally I had CloudAve open and noticed a post from Zoli Erdos:
We’re experiencing trouble with our Feedburner feed – apologies and we certainly hope to get it fixed soon.
(Oh, and Googlers who read this are most welcome to chip in to help )
Lo and behold and Feedburner is displaying an error message. I nipped over to Feedburner to see if the feed in question was displaying the same problem. No. Curious. What could it possibly be? My sponsor posted a test post but when I refreshed the raw feed at my end – nothing. Nada. So here is what I think happened:
At some point, Feedburner stopped taking information from the original feed. However, since it HAD been receiving information, that data was sitting in the feed reader mechanism. Systems that take and fully refresh data from Feedburner were not able to see what I was seeing, hence the fact nothing was showing up. If it doesn’t see anything then it can’t parse that back. That’s a big deal. I’ve been told that Feedburner has been crappy for months. According to Jevon MacDonald who knows a good deal about these things:
To make matters worse, Feedburner is inconsistent. Some Feedburner feeds are fine, others are screwed up. You can’t tell until someone asks or you are monitoring the service. Monitoring defeats the object of the exercise of having services that automate tasks. So – if you are using Feedburner or rely on its ability to parse data to services upon which you rely…be warned. And by the way – I have a Feedburner problem per the image below which tells me I have zero subscribers to my two main blogs.
We solved the problem by using a different feed and avoiding Feedburner altogether. That’s a bummer because it will be harder for my sponsor to get tracking data. Tracking what’s happening with feeds is important because it provides a way of knowing what works in terms of referrals to your content. Feedburner provides a number of tools with which you can do that but if the basic feed has issues then there is real trouble.
The problem is less to do with Feedburner per se but with Google which owns the service. Google has a dreadful reputation for acquiring technology and the people behind it and then leaving them to languish. That happened with Jaiku, a great version of what Twitter could have been. Google Mail gets reasonably regular updates but still lacks features that some consider essential. Google Calendar is inconsistent across different platforms. Google Docs is OK as a simple wordprocesser and spreadsheet but can’t hold a candle to Microsoft Office. I could go on ad nauseum.
Google occasionally says it is serious about business applications. Now and again we see a flurry of impressive new wins for GMail. The fact is that 97% of its revenue comes from advertising with a tiny fraction from the many services it has acquired. What do you think is likely to get attention? The ad services of course.
You can argue there is no point moaning about services that you don’t pay for in hard cash unless you’re a Google that can command many millions of eyeballs. That’s an excellent point and one that is rarely missed by professionals. However, we live in a world where ‘free’ has become the new normal, where we expect to try before we buy and where developers are encouraged to see if a service will catch on before they figure out a business model. As a business person this is the epitome of madness. Would you open a shop or start a business unless you knew there was a revenue opportunity? Would you advise a client to do that? Would you say to the guy or gal that wants to open a sandwich shop start with a mobile cart and see how it goes? Would you trade on people’s natural greed in the hope you could milk them later? Of course not. But that’s not the world of much current software.