Google engineer, researcher and other Google emploees in general.

[kgun]

1. Background

I have been on the internet since its start and followed the discussions on different forums.

2. The problem

One thing I have discovered is that what Matt Cutts says is taken literally and as a truth. Here

Disclaimer: Even when I joined the company in 2000, Google was doing more sophisticated link computation than you would observe from the classic PageRank papers. If you believe that Google stopped innovating in link analysis, that’s a flawed assumption. Although we still refer to it as PageRank, Google’s ability to compute reputation based on links has advanced considerably over the years. I’ll do the rest of my blog post in the framework of “classic PageRank” but bear in mind that it’s not a perfect analogy.

Source: Matt Cutts PageRank sculpting

he implicitely admits that he don't know every aspect of the algorithm. And it would be very remarkable if he did.

3. A similar example.

I worked in the research department for the Central Bank of Norway with a lot of the banks software. I can tell you, knowbody except me knew in detail the algorithms I was working on. Some of them like a quadratic programming program with a lot of restrictions based on Harry Markowitz's CAPM model, required mathematical and numerical knowledge at a certain level to be understood. A lot of information offisers and journalists would not have a chance of understanding these programs unless they had a strong mathematical and numerical background. I remeber what a new Information offiser and responsible for external information from the bank once said. He looked at our simple to understand syclical analysis data bank. Did you make these systems already in the 1980's?

I also remember a discussion on the Sitepoint forum (one of the bettter forums) about object oriented programming. The discussion was about Ruby on rails (made by a person from Denmark). Members were so impressed that this language / platform was invented as early as 2000. - Bjarne Strostrup (another Dane) that inveted C++ is earnest in his books. He admits that this and this feature he stole from Simula. Simula was invented in mid 1965.

Of course there are very few internal engineers and researchers that know Googles advanced algorithms in detail. They may be even more protected than Coca Cola's secret formulas. So the cited disclaimer by Matt Cutts should perhaps be stated more often. But if it is stated too often, that is not good either. So when Matt Cutts speaks he should know more than a Google employee outside USA and what is written on

Google's: Webmaster Central Blog

(There is an interesting top story there while this post is written: Adding associates to manage your YouTube presence)

I highly doubt that Goolges engineers and researchers write there.

4. Further reading

How Google Finds Your Needle in the Web's Haystack

Litterature:

Amy N. Langville and Carl D Meyer: (2006) Google's PageRank and Beyond: The Science of Search Engine Rankings.

Revisiting The Capital Asset Pricing Model

5. Conclusion

IMO, I don't know about a more informative person speaking on behalf of Googles spam filters and googles algorithms in general than Matt Cutts. But don't take everything he says literally. His main task is responding to feedback and discussions with webmasters. He is the person to listen to, but don't take everything he says and writes as a truth and an exact statement. Take what he says with some sceptics. When he informs about a new tool. He may on behalf of Google test the market. It is not necessarily true that that tool will ever be an important part of the algorithm, and outsiders don't know how it is used by Google. I doubt that a single person know every aspect of Google's spam filteres and algorithms.

And when a Google employee outside USA speaks and writes, don't take it litterally and put less weight on it than when Matt Cutts speaks. But neither know everything in detail. The internal secrets will most probably never be revealed to the public.

6. Private copy.

Since some of my posts here at WPW seem to be randomly deleted, especially one about semantic linking, I take a private copy of it and post it here at WPW only once.

[C0ldf1re]

... One thing I have discovered is that what Matt Cutts says is taken literally and as a truth. Here
Source: Matt Cutts PageRank sculpting
he implicitely admits that he don't know every aspect of the algorithm. And it would be very remarkable if he did.
...

It would be very remarkable if any Google employee knows the whole thing. This is not just a matter of it being very complex. There is the security aspect to consider too. No single employee should be able to divulge the hugely valuable secrets of the algorithm. No doubt the coding is divided into modules, and the coders are only allowed to see their own module.

[kgun]

Litterature:

Amy N. Langville and Carl D Meyer: (2006) Google's PageRank and Beyond: The Science of Search Engine Rankings.

Bad experience. It seems that even authorative links changes content. The last link above should of course be to the book (where I in case one of the links should be removed or change content list two links).

http://press.princeton.edu/titles/8216.html

http://www.amazon.com/Googles-PageRa.../dp/0691122024

We see it again and again. Webmasters and even scientists don't seem to know the advertising power of stable semantic links.

Here in Norway companies pay hundred thousand USDs for 5 minutes Tv advertising while they delete stable semantic links or redirect them to nonsene or deliver a 404 error page.

Cool URIs don't change

There is an onging discussion here about: Link removal... are you doing it?

A better use of your valuable time should perhaps be to study outgoing links on your own sites:

1. Are they broken?
2. Has the target changed content?
3. Does the link take you to a 404 error page?
4. Are you blocked from visiting the page?