Note to Old Welsh Guy, don't be helpful twice in a short period of time. I tried to green rep you again but it's telling me to spread the love
The thing is that I got this info from a semi trusted source and due to this a decision was made... an important one. This leads me to believe that TR is out there, but still not spread out
Now here is a thought with a sound logic. If there is positive trustrank There must be negative trust rank (suspicion rank)SR I am not talking porn, gambling bad links etc, I mean jost JO AVERAGE, who has a links page(s) and links out to loads of irrelevant links
Ok, as a result of this thread, and also because I am seeing this come up more and more, I have written an article http://www.umbrella-consultancy.co.uk/art1-trustrank.htm . I welcome feedback, good or bad, but remember that I am self educated, have no real formal educational qualifications, and as such my written English is lousy. I hope you can get the message, withour the quality spoiling it.
And I thought my article produced a head ache jasontchandler.com/google-services/google-trustrank.htm Great article it always helps to see anothers thoughts on the same subject.
Yes, it is a simple concept really, but trying to explain it in simple layman's terms is not so easy. I see that we both decided to go down the whole number fractiopn road, as trying to use realworld examples would hve made the process seem complicated. It really isn't it is simple in the extreme.
I believe that overall topic of ‘trust’ goes even farther than how it might relate to page rank. I would suggest that every new site receives a default ‘trust factor’ and that passing of TrustRank increases or decreases that initial trust factor. To explain my view more, it’s helpful to look at how pages are ordered in SERPs. G states that they are presented in order of a ‘document score’. That doc score is the number that the algo results in, for a specific KW or phrase for that page. So, let’s say my site is about New York City hotels, and let’s say my root page has a doc score of 20,000 for the specific phrase ‘NY City Hotels’. Sites with a greater doc score will appear above me, and sites with a lower score will appear below me. I believe that G multiplies that doc score by a trust factor of something between 0 and 1. In this example, let’s say my root page’s trust factor is ‘.4’. Multiplying 20,000 by .4, I get new doc score of 5,000. This would probably put me a few hundred places down in SERPs. If I could get my trust factor to ‘1’ though, my page would reach its full potential. /tom/ BTW: I think that the sandbox is nothing more than a low default trust factor.
Longcall, where have you got this idea from? There are clear white papers and patents explaining how trust rank works and it is nothing like you have said it. Google ranks pages using two main elements that it calls 'relevance' and Importance'. The relevance is a semantic relevance where the phrases are used, the importance is their Internal pagerank system. I am not sure what you mean when you say All the elements of the algorithm impact on the others in either a positive or negative way, so if your Title content is spammy, then it will likely decrease the value of other elements. As I stated in the article I wrote and linked to above, this is How trust rank works, and it is NOT my opinion of TR it is FACT from Stanford university. I am not having a go at you so please don't think that, I am just saying that what you are suggesting is the complete opposite of both the original white paper below and the patent google applied for. If they are doing it your way, then they have no patent on it, and they would NEVER take that risk IMO What you say above is completely how page rank works. Every page has a PR value of 1 (I am talking real PR value now not the silly green bar thing) The page can then vote up to 85% of its own value via links. Trust rank is ABSOLUTELY not given, it is earned. Honestly every single white paper and patent applied for confirms this. Trustrank is a MANUALLY seeded system, Trust rank can not be created in any way other than through this manual seeding. Trust rank is also independent of Pagerank, although I have suggested that it could be used as a multiplicator to the value of a page and how the elements are valued. A prime example is here. If you read in a magazine that life had been found on Mars, how would you react? The answer will without doubt depend on WHICH magazine you read it in. If it was the national enquirer then you would laugh, If it was the New Scientist you would believe it. Because your level of trust in the content is affected by the source of the content . This is a working example of trustrank.
The great thing about TrustRank is that the principles behind the concept and the end result follow a common sense approach and are designed to mirror the way 'real life' works. Any algorithm or process designed to combat spam and improve search results has to mirror the real world or it will fail in the long term. In a nutshell the way to improve your rankings is by getting important websites to link to your site. Great points made by Old Welsh Guy by the way.
Mad4, I would say that Orkut has a LOT to do with trustrank theory. Google had the perfect opportunity to look at their trusted sources (the top web marketers and specialists in other fields), then they could track how the social network would carry an invisible value. Google had all the data and social networking laid out in front of them. If trust rank was working properly, and they 'seeded' some of the most well known and trusted individuals, they could in the blink of an eye, see how much of that trust would filter through to the black hat members of the network. If none (or very little), then it was working. As Mad 4 says and I wholeheartedly believe in, Offline systems MUSt mirror offline systems. And trustrank wirrors social networking to a tee IMO. Thanks for the compliment Mad4 How is Shottingham these days
This makes me feel depressed. Just another thing to worry about now. On one of my sites I have a HUGE amount of unique articles on my niche, I wonder if that counts for anything. I have alot of good unique tutorials too. =[
LY2, if your articles are unique (IE you commissioned and own them and they are no where else). And they are of high enough quality, then they will attract quality inbound links eventually. Might I suggest that you write or release an article, and mass promote it. This will drive traffic to your site, where people will find other articles, see they can not be used elsewhere, and will then link to your site as a resource. You have no need to feel depressed, unless your is just another article site
I get this view from observation and logical thought. Like most others, I believe that PageRank is a trivial factor these days. So, white paper or not, it appears to be of little value in determining position. And what I am saying is that I believe that the seeding and passing of TrustRank is not the whole whole story. I have never seen an explanation of how TrustRank effects position in SERPs. G has discussed the issue of a doc score and I am simply saying that it appears to me that these days the doc score is multiplied by the document's trust factor (0-1). That is, a trusted site is allowed take its natural position in SERPs while a site with a trust factor of less than 1 results in a lowered doc score. /tom/
Hi Tom, I totally agree with you that page rank plays only a very small part in the actual ranking of sites in the SERP's. I can see from the tone of your post that I have offended you, I am sorry for this, such is the typed medium, it is almost impossible to convey emotions . When I spoke of white papers, I was referring to Trustrank, not pagerank though. Again I view them as does Google appear to do, as totally separate values. Maybe there is some confusion that TR is simply a supplement to page rank. I don't believe that at all. My feeling on how trust rank will affect position in the SERP's is simple. As I have stated, all the elements of the algorithm are interrelated, and are effected by each other as the base 5 sliding log scale algo kicks in. Google have stated that this sliding scale algo gives them 5 billion permutations of the basic algo. So trustrank will simply boost the values of certain elements such as the page title, Page description, and all the usual elements of on page algorithmic factors by a % IMO. As these elements clearly state what the site and page is about. I would also expect the anchor text value to be affected by the trustrank of the anchor page. Higher trust rank, the more value placed on the anchor text pro rata to the value of TR. The problem is that these are the very elements that are misused the most, but by using a multiplier of TR, then google can use all these elements again as they should have been used, but by including the multiplier of TR in the equation, they will know how MUCH they can rely on the same elements on different pages. Page rank is purely a numeric value that can be manipulated because PR can be created and channelled by unscrupulous webmasters. Trustrank can only be gained by trusted sources linking out to you. If Google contain the trusted sources, then they control the serps. Imagine a trusted source goes bad and starts controlling outbound links for gain (monetary or personal in the case of volunteer editors). Google could easily reverse engineer the source of the trustrank failure from the spammy pages that are showing up as a result of this infection. They can then simply adjust down the trust value of that source (manually), and they have dealt with not just one site, but the whole chain of TR that has been falsely created by this ex trusted site.
No offense taken at all, really. I was just in a hurry to get to work and as you say, often the wrong emotion is conveyed. I did not intend to sound angry or upset. Anyway, there is some good food for thought here. . . /tom/
1st Google page rank is in no way related to Goolge trust rank. Page rank is easy to malnipulate. Trust rank is one on the most improtant fetutes for ranking high in Goolge. This is why some sites can rank very well in competative keywords with only a few dozern backlinks. There is a website in one of my keywords that ranks in the top 3 for most of its keywords with only around 20 backlinks showing on google. MSN & Yahoo which shows most links still only shows 350 links. The reason for this is that they have around 15 links from .gov.uk site and around 8 links from ac.uk (education) links. Goverment sites (.gov or .gov.uk) carry huge trust rank. The next best for passing on trust rank is education links such as .edu or .ac.uk Other high authority sites are old sites that have 1000's of backlinks that have proberly been pointing to them for many years. As they have been around for so long they have gain good trust with Goolge After this look for authority site in your neiche. These will be mainly sites that are in the top 3 for your neiche. When i look for links of site always check for that site backlinks and see where they come from, If they have any links from .gov or .edu sites then i would always link to them. Trust rank is passed on from site to site. So for example if a .gov site pass on a trust rank of say 10 then the site that is link to the .gov site will pass on a trust rank of say 2 compared to a normal link that may pass 0 or 0.5 Figrues are used just as an example. Links will also gain trust the longer you have them. I'tis widely accepted that Goolgw don't put any weight on links unless they are a least 3 months old. After this period they will slowly start to gain trust rank. The reason why Goole uses Trust rank instead of page rank to rank sites is very easy to explain. It is esy to pruchase or get links from high PR sites say 6-7-8 which will easily raise your page rank. Trying to get links from .gov site is near impossiable and also very hard to get normal links from .edu site and high autority sites as they are all very carfull who they link to and must have a very good reason to link to other sites. To get links from these site you must have helpfull content in some way which will help the visitors on they sites. Thats what i beliver anyway for what its worth
You are confusing TrustRank with Signals of Quality. They are totally different concepts. You can read the Q&A on TrustRank I posted a year ago here. Matt Cutts and the Google engineers continually refer to Signals of Quality and although no one outside the 'plex knows exactly what these are the variables you mention are as good a guess as any. - Michael