4 ms·
I'd be curious what people on HN think of that essay. The direct link is http://works.bepress.com/cgi/viewcontent.cgi?article=1034&context=james_grimmelmann htt
by Matt_Cutts 16y ago
I'd be curious what people on HN think of that essay. The direct link is http://works.bepress.com/cgi/viewcontent.cgi?article=1034&context=james_grimmelmann http://works.bepress.com/cgi/viewcontent.cgi?article=1034... for the PDF or http://james.grimmelmann.net/essays/SearchNeutrality http://james.grimmelmann.net/essays/SearchNeutrality for the HTML version. The PDF is more in-depth, but it's a fairly long read.
- Andrew_Quentin 16y agoI agree with the main proposition that search by its very nature is not neutral, you are searching for something specific and therefore biased towards that something. However, I do not think he effectively argued that google does not manipulate its results by favouring itself, specifically in regards to paid results or paid ads. These paid results are presented to the user in the same style and fashion as the organic search results, but differentiated subtly by a very light different colour background. They are above organic results, right below the search box, making it much easier for users to click on them. Due to their prominent placement, they attract most of the attention, although web savy users might have learned they are ads and thus have learned to ignore them. Finally, for all practical purposes, the paid results are the first ranked, suggesting therefore that google finds them to be the most relevant. The conclusion is clearly that you can buy your first position on google, even if it is not the most relevant.