As is evident from several posts on my blog, I am a total food freak! I am also a regular burrper. I have written several valuable reviews at Burrp and also received written appreciation from Burrp. A couple of months back I was thinking of copying food review from Burrp to my blog. Since that was going to be tedious, I went looking for a RSS feed from Burrp. It was sad they don’t offer it, but I wrote them my feedback. Now I was left with no option but to think out of the box!

And then the Web 2.0 enthusiast in me came to life! I had been using a service called Feed43 to process several feeds and remembered that it allows creating a feed out of any page on the internet. I checked the HTML source code of my Burrp profile page to find that reviews were quite structured in terms of markup. So why not let Feed43 read out the page and create a feed for my reviews? This way I won’t have to copy anything manually. Moreover, whenever I post a review to Burrp, it will be available on my blog in less than 6 hours (that’s the refresh rate for free feeds at Feed43)

I’m going to present a small tutorial about how to get going with Feed43. You will need to sign-up for the service, and start by creating a new feed. The first input expected from you is the URL of the page or feed that you wish to use as the information source. In my case, its the Burrp profile page URL.

F43 will extract the markup from that URL which you will have to study to extract information. It’s more about pattern matching. I identified that all my review titles along with the link to the review are enclosed in <h3> tags. There is some useless information in between that I need to skip.

So I help the tool find the information by specifying delimiters or markers for start and end of information – the h3 tag in this case. Since I see a line break after the tag, I use {*} to escape any control characters that falls between the h3 and span. My review title is wrapped between the title tag, so I indicate that with a {%} which I have already specified as a ‘global match pattern’.

Please note that ONLY data matching the entire search string will be fetched. So, this won’t give you data inside other h3 tags that don’t have a span & a in between. So 1 mistake in the pattern, and you will find nothing! Click ‘extract’ and system will find all instances of matching data for all the {%} specified. We had 2, so they are shown below as %1 and %2 for all 5 items matched! (Wondering why 5? Because the Burrp profile shows last 5 reviews; you need to paginate for the rest – Feed43 can’t paginate for you, sorry!)

Now that data is extracted, you need to consume it! That is, put it together into a feed. I will be using %2 as the item title and %1 as the link in my feed. Note, that you can write your own text and append it to any field; eg. I wrote ‘Review:’ before the %title.

You’re almost there – Just give it a sensible name (or use the randomly generated number) and grab the URL for the feed.

The feed can be accessed through a reader (aggregator) like Google or in your browser (click to see how it looks)

And if you own a blog, use it in a RSS widget and show-it-off! Like I did… (What you see as the result took a more complex search pattern, and I’m more than willing to share it if someone’s interested)

Conclusion: Feed43 lets you create a feed out of any page (provided you are authorized to) and share it with people! As another example, consider a bidding site that always displays 3 items closing in the next 10 mins on the homepage; I can create a feed for this page, and read about the items in my reader, without having to go to the page each time.

Use the comments box to suggest any other cool applications or if you need help creating your feed.