diberry - 3:25 pm on Apr 4, 2013 (gmt 0)
My source is that I've actually been part of several communities seeking out non-mainstream views on politics, sociology, medicine, etc. Not everyone finds mainstream ideas/opinions/theory satisfying - especially if you know how often medical "facts" actually haven't been tested. I.E., for decades doctors advised us that eating saturated fats would give us high cholesterol. This conclusion was actually based on a very limited study on a few heart patients - its conclusions were never intended as general advice for all humans. Finally, between 2008 and 2012, someone decided to do studies to confirm the conclusions for everyone, and they got the surprise of their lives: saturated fat consistently raised the GOOD cholesterol in human test subjects. Which means we're back to not really knowing what causes a bad cholesterol profile, nor how to fix it - in fact, we're not even sure we know what constitutes a bad cholesterol profile, as doctors worldwide are struggling with various theories.
This is the real info, that comes from real studies. You find it on small, independent sites with citations from academic studies and textbooks, not Livestrong.
Granted, this is a minority niche, but when the query clearly indicates the searcher is looking for alternative views, mainstream sites should not come up first. That's all I'm saying. And again, I'm not accusing Google of intentionally boosting brands. It just works out that way sometimes when it shouldn't, so there is room for improvement.
tedster, medical searches are a good example of where Google are REALLY bad - they put way too much weight onto a very small cluster of sites - livestrong, mayoclinic, webmd etc - they pop up every time no matter what symptoms / medical query you put into Google. They offer generalised information at best that you often feel is written by someone who is NOT an expert on the subject, which is often not what you want when you want EVERY detail on a particular ailment.
Precisely. They give a nice overview, but they don't link to actual studies - and that's really important if you have an unusual health issue or one that's poorly understood.
Again: I'm not saying alternative sites should trump MayoClinic when someone's looking up a common disorder. But when the query is clearly seeking alternative medical information or the studies themselves, sites with that info should come up before sites that don't have that info, but have more trust signals.
Is this really such a controversial view? Do some people actually think that a query should bring up irrelevant results first, just because they're on popular sites? That certainly WOULD constitute brands being given an artificial boost by Google, just because they're brands.
I actually believe Matt entirely. I don't think the algo is designed to give brands a boost, but I think sometimes it works out that way because the algo is imperfect at determining how much "trust" should come into a query.
A better method, IMO, would be if Google looked for sites that actually link to and cite genuine medical studies from recognized study sites like PubMed/NCBI. That's where you'll get the most proven, validated information available in current medical science.