Marcel Marchon’s Blog

Hehe, this is pretty funny - remember those games that used to have a panic button in case the boss came by to see what you were doing? Lastminute.com has one of those, too - look at the top right:

If you click on it, it takes you to this page.

FeedJumbler now caches HTML and JavaScript output for 30 minutes and doesn’t refresh the source feeds each time anymore. This should make it quite a bit faster for everybody that is using those versions.

Also, here is a first release of the underlying “generic feed library” layer.

I’m using analog for analyzing the logfiles for my websites and I’ve recently noticed, that California Bits is getting an increased number of hits from Google Image Search. Now, those referer URLs are kind of tricky, because the referrer is usually not actually the search results page itself, but rather the “image detail page” (e.g. this). So in order to find out the terms that the user had searched for for your “Search Query” report, you have to do some creative aliasing of the referrer URLs. This seems to work pretty well for me:

REFALIAS http://images.google.*/imgres?*&prev=/images%3Fq%3D*%26* http://images.google.$1/images?q=$3&$4
SEARCHENGINE http://images.google.*/* q

(note - the above should be on two lines - it may unintentionally wrap here. One lines starts with REFALIAS and the other with SEARCHENGINE)

PS. This guy seems to have had a similar problem.

Update: see also this update.

At least for their error pages, it looks like they do …:

See http://feedjumbler.com for details.

Technorati Tags: rss | atom | feed | syndication