91–100 of 217 posts
Earlier quoted context omitted.
An extremely easy way for a bunch of people to get together and destroy someone's ranking? That doesn't sound like such a good idea.
Given that Google gets hundred of millions of search and visitors, then you'll need a hundred of thousands of down votes to get black-listed. No black hatter can really do that (create a 100K account/IP to avoid Google radar and down vote the website).
Earlier quoted context omitted.
Do you think 99% of users are too stupid to click 'report spam' when they get a spam email?
Not "too stupid" no. I think 99% of email users have not been adequately trained in why or how they should report spam, and even if they were I think most of them would still not care enough to actually do it with any regularity. When pushed many may acknowledge that they know it exists, they will probably even be able to find the button when asked if given a chance. But they won't remember to do it when they see spa…
Yes that would be good. They could then look at the number of people blocking certain domains and de-weight them in the global results. Traditionally google seem against human powered editing (as this would be), but I think as the black hat SEOs run rings around them, its needed badly .
An extremely easy way for a bunch of people to get together and destroy someone's ranking? That doesn't sound like such a good idea.
Earlier quoted context omitted.
Not "too stupid" no. I think 99% of email users have not been adequately trained in why or how they should report spam, and even if they were I think most of them would still not care enough to actually do it with any regularity. When pushed many may acknowledge that they know it exists, they will probably even be able to find the button when asked if given a chance. But they won't remember to do it when they see spa…
Do you think this could be improved by alternate wording? Instead of reporting 'spam', ask the user 'was this useful?' or 'did you want this?'. After all, the real goal is giving people a better, more relevant experience, detecting and removing spam is just one facet of that. Whether it's email or search.
However I don't think it's as simple as changing the button text. Even a process driven UI, like a wizard style interface where you have to click next to progress through each step might work, but users very quickly become immune to dialogs. They don't read them, they just evolve the actions that get them to their goal the fastest, and the users goal does not include reporting their spam.
No, it's not particular hard, but it will make the problem worse. Why? 99% of users are non-tech oriented. Those users will not really be aware of the specific problems with the search results, they won't understand the concept of a good vs bad result and they certainly won't bother to tweak/ban/filter their results. The 1% that do care and are currently being vocal about it will start filtering their results and the…
Perfection being the enemy of good enough, and a common and valued and traditional mechanism to delay product shipment.
And Google might well be able to utilize information from that 1% of users that have sorted that out - 1% of a Really Big Number of searches, factoring for the folks looking to game the search results (downward, in this case) - to provide feedback back into their search results.
I think the worst culprits are the ones that skim StackOverflow questions and rehash them into their own supposed original "question and answer" site
What do you think most StackOverflow answers are? It's a karma-paid labor pool where you can post questions and a lot of under-employed people will rush out and do the necessary Google searches, collating and slightly rewriting the results to yield the most votes. Everyone is ripping off someone's content. And just to be accurate here, SO content is creative commons (created by the community). Are those just cheap wo…
In my experience though the sites that are taking the content are ad ridden messes which remove value rather than add anything.
Earlier quoted context omitted.
Not all spam is so obvious, both in email and search results. These are all assumptions, but I think it's safe to say the average person identifies the email 'Grow your Pe Nis like a Woman' (true story) as spam. However, the less obvious 'Enter to win a trip to Hawaii' still fools many people. Think about how effective the 'I am a Nigerian Prince that needs help out of my country and then I will give you 1.2 million…
Frankly, I'm not fussed whether the majority of users would use it right. I posted this as an entirely self-serving request for a way to make my results cleaner. Also: If the majority of users do like eHow, then that's a sign it's not spam. But a lot of 'power users' consider it spam. This is an argument FOR personalised blacklists, not against them.
Earlier quoted context omitted.
Given that Google gets hundred of millions of search and visitors, then you'll need a hundred of thousands of down votes to get black-listed. No black hatter can really do that (create a 100K account/IP to avoid Google radar and down vote the website).
Hundreds of millions of searches and visitors in aggregate: yes. Hundreds of millions of searches and visitors in any one keyword niche: not so often. Many websites live off a handful of visitors a day coming from a few core keywords and associated long-tail traffic. For a keyword that only gets 100 searches a day, it wouldn't take many down-votes to affect the rankings of the relevant sites.
Why do you assume a flawed implementation?
Naturally there would be thresholds. There is no reason to devalue a site that's only displayed in 100 search-results/day at all.
The sites we want to hit are orders of magnitude worse at polluting the results. We're talking about the Mahalo's and expert sexchanges.
And no, this doesn't solve the problem.