Nikolai Popkov

не стоит злоупотреблять цитатами

сознательно занижаются сайты на которые есть текстовые ссылки, причем занижение идет именно по точному соответствию этим ссылкам - вот такая вот пессимизация. То есть даже ссылающийся сайт есть в выдаче, а того на который ссылается - нет.

"Google SEO (as SEO for any other search engine) comes automagically if you do the right things. Google SEO is not what you want to be doing - you want to be making good content for your user base."

Brett and I have differed on a several things in the past, but there's some pure wisdom waiting for those who are willing to study what he's saying. Just to try to make it more explicit (from my viewpoint--Brett might not agree with everything I'm saying):
- showing up for one big phrase is probably not the best philosophy. Target a bunch of different phrases. Instead of worrying about rankings, spend a whole day looking at your server logs and the phrases that people have used to reach your site this year. Being #1 for a high-profile keyphrase is nice, but much better to show up for a broad base of queries.
- SEO = change. At least at Google, you can assume that there is a constant (or even accelerating) amount of work being done to improve quality and our other services. Sometimes that's visible to the outside world. But even when it looks like nothing major is happening at Google, lots of work is going on. Sometimes we're building infrastructure for the next generation of quality improvements. If you think there's a magic formula to SEO, you're going to be disappointed. The only magic formula that I know of is to build a diversified, content-rich site that users love and link to on their own--the sort of site that you can promote with newsletters, or forums, or any of a hundred different ways such that search engine traffic is nice but not essential. It's harder to do that than to fiddle with PR bars/buy links/sign guestbooks/blogs/whatever, but ultimately if you follow Brett's "build a site in 12 months" guide you'll sleep better and your time will be better spent. Even if Google re-ranks your page, as long as you're building your site primarily for users you know that Google shares the same goal of ranking your site well because users find good information and like your site. Following the sign guestbooks/buy links/PR bar watching path is much more volatile, because you never know when we'll introduce a new classifier, or introduce algorithm improvements that buying links or watching PR doesn't apply to, or introduce new signals of quality.
- Take everything with a grain of salt. When someone does research, they consider the source before they believe it. If someone claims that an H1 tag hurts any site, or that internal links with specific anchortext hurts a site, stop and consider it for yourself. Does something mesh with common sense? rfgdxm1 is pretty good at doing this, for example.
Um, there's probably more that I'm not thinking of. :) If something comes to mind, I'll post it again.

"People resort to cheating ways when they are unsure of what to do."
Sadly, nileshkurhade, sometimes they also resort to cheating ways even more when they are sure of what to do. We've got guidelines on our site at www.google.com/webmasters/ and Brett wrote a lot about how to do this right. You'll also find tons of good info in the archives for this site. But my advice would be something along the lines of 1) make a site map 2) make sure your site architecture such that the spider can find every page if it wants to, 3) pick a small niche where you can be the undisputed best resource. Don't start with something ultracompetitive--look for a fringe or niche or related area where you can present something new, esp. if it's new or useful information that users can't find elsewhere. 4) Once you have a good handle on your niche, expand your horizons. Now that every knows to come to you for reviews/info/analysis about fuzzy used widgets, you can set your sights on just used widgets in general. 5) Notice that I haven't mentioned anything about links, cloaking, etc. Take that lack of mention to heart, and just add a page of useful hand-written content about your niche to your site every day.

Those would be my off-the-cuff recommendations to you.

I saw a similar post by someone. Basically they said that anyone doing any "insurance"-related search clearly wanted to buy insurance. What do you think, James_Dales? Should every insurance search return "buy my policy"? Or would you rather learn about the different types of insurance, and which type of insurance is best for my situation, and what the limits of a policy is? You mentioned inkjet cartridge. Yes, most people typing inkjet cartridges probably want to buy one, and the search results give plenty of places where you can. But to take an example of "U bend" pipes earlier--not everybody searching for that want to buy one. Maybe they want to know how to install one. Or how to take one out if stuff gets rusted. If I'm searching for dryer vents, I might want a dryer vent cleaning service or product, or I might want information about how dryer vents get clogged up and how I'd clean it myself. Good webmasters in commercial areas will sell things, of course, but they'll also have unique information that helps users and differentiates their site.

So you say: If a user types in cheeseburger, it's Dilbert thinking if you believe that the user wants anything other than the cheeseburger. Personally, I think a user who just types in cheeseburger might want information on how many calories are in a Burger King cheeseburger, or a Wendy's cheeseburger, etc. Or they're having a BBQ and would like to cook the perfect cheeseburger and would like a recipe.

superscript, why wouldn't you type buy cheeseburger or cheeseburger prices? Why should just the query cheeseburger carry an implicit "buy" keyword along with it? You don't think an average user is probably looking for recipes, esp. given that it's hard to order a cheeseburger online and have it arrive fresh? ;)

superscript/marin, I'm not trying to argue, and I apologize if it comes off that way. I'm just saying for some specific searches (using "cheeseburger" as an example search), a user typing in "cheeseburger" probably isn't really looking to buy one from their Google results page. If I tried to put on my user hat, I would honestly guess that recipes would be the best match if that's all I knew about their query. Sometimes there's user-centered philosophy and SEO-centered philosophy. The best SEOs in my opinion are exactly the ones that practice user-centered philosophy.
Let me try to give a different example; check out this thread:
http://www.webmasterworld.com/forum3/20027.htm
Read the first post. Notice how the first post only talks about rankings and different ways to manage backlinks? That post at first read sounds only like "how can I do better in rankings for this particular phrase?", without mentioning underlying quality or site utility at all. Instead, they're comparing backlink counts and asking how the keyword should be in the domain, etc. That poster would probably do well to read Brett's posts rather than concentrate on what he thinks of as his big keyword. Checking server logs would probably find many other more productive ways to build his site than spending all his time on that single phrase.

miss understood, there's nothing like that--that's the nice thing about algorithmic changes; new scores can be computed without any human intervention. So as far as the person who not only claimed that pages were penalized but that you should make the exact same page with a completely new url just to test changes.. that's barking up the wrong tree. Building good content sites is the best thing for users, and we're trying to bring those results to users. So no, I wouldn't worry about "penalties" from the most recent changes

monsterisp, we've been looking for examples of searches like that, because there's always room to improve searches, and we want users to get the best results (both if they're looking for info or if they're looking to buy). If you have examples like that, we'd like to hear them. Send them to the spamreport or webmaster [at] google.com with the keyword floridaquality instead of floridaupdate. I couldn't agree more with your last paragraph, and I'd expect that going forward we want to make sure that users are able to find info if they want info and are able to find good product/service sites if they want to buy.

dawlish, the idea of sabotaging another site by linking to them with whatever anchortext has always been barking up the wrong tree, and still is. :) I wouldn't worry about that. "Yes" as answer to your last question.
dazzlindonna, I agree that keywords can imply that you want to buy things without explicitly saying so.

Okay, I really am going to try to hang with my family now. :)

We want search results to be relevant for both info seekers and product seekers, and everything in between. If there's a search that you consider less relevant than before, send it to us at webmaster [at] google.com with the keywords floridaupdate or floridaquality. Emails like "where is my site" probably won't get the same weight as "I did a search for X. The industry leaders in this space are A, B, and C. A & B are fine, but C dropped to the second page. In addition, new page D is less topical or relevant because ..." That's the sort of email that would be the most helpful for us to get. I'll ask a user support person to collect (constructive) emails to the webmaster address so that folks inside Google see them.

Those who think no KW filter is being applied are misleading themselves I think, because I have done tests for similar keywords (which are all used in the same context).

Here are the results of those tests:

Of the several KW's tested, they all rank "top 10" except for the one KW which originally had a slightly higher density than the others. The higher density KW is no longer ranked at all (nowhere to be seen within the top 500) but the similar KW's (which are used in the same context and have many of the same words) are now ranked top 10.

This kind of throws a wrench into the theory that google is using only "context" based searches and no KW filtering whatsoever.

I do know that google is also using "context" filters and some form of "word stemming" as you have stated (that is correct), but they are ALSO using KW filters as Kackle has said. It's a broad new landscape, it includes all of these things.

Further to Benda_J's point about a new landscape - Google used to have a 2D view (up and down) of the web - now it has a 3D view (up, down and *SIDEWAYS*). I had succombed to the idea that Google was filtering some sorts of searches - I no longer believe that. Its just that webmasters have gotten used to the 2D view, the view from their one particular focus, especially when that focus is commercial.
Google is now presenting lots of ideas to the user, in an effort to match the context of the results with the idea(s) within the users head. Googles results are now showing abstractions of the ideas within the search phrase which is why they look irrelevant when viewed with your particular focus.

When you look at the Top 10 results for searches now you see that Google is offering variety - its not really an ordered list. Perhaps this has always been the case with pure informational areas, but in the money areas we are used to seeing exact matches with the search phrase - webmasters had learned to feed the algorithm what it wanted to climb up the ladder. In part, Google were responsible for this - Page Rank and its Toolbar representation actually encouraged the 2D view. By learning the elements of the algorithm, coupling it with PR, commercial webmasters had managed to impose implicit commercial ideas on most search terms so more and more search results began to have an implicit commercial aspect.

Google have just changed all that. Your website can no longer rise to the top on specific keyword relevancy alone. There is going to have to be a context within which the search terms exist on your site. Google can't know the exact context of the users search (especially for short 2-3 word phrases) - your webpages are going to have to help it offer different contexts to the user. Not only will Google use on page factors to determine the conetxt of your webpage its going to use inbound links and anchor text - BUT here is the real difference - the keywords being in the anchor text isn't going to help - its the idea within the anchor text, which probably includes ideas of where the link originates as well.

If your webpage is useful in lots of different ways, then you will stand a good chance of being visible. This is going to make it much harder to gain an advantage using reciprocal links - how do you know what text to put in the links? That's why your're going to have to build content that will be naturally linked to for a whole lot of reasons - some you wouldn't even have considered.

All considerations about is H1, H2, keyword density, URL, etc, are now trivial - they just aren't going to help get *just another set of words* elevated up the rankings.

The impact of this on commercial websites, particularly affiliates and widget resellers is going to be devastating. I imagine that the rankings will now change frequently - unless you are a mega-brand or niche authority expect lots of constant turbulence. If you're hawking what 100,000 other websites are hawking then find something new to hawk. If your promoting products that 100,000 other websites are promoting then you'd better have something interesting to say about them.

If you can't find anything new to hawk and you don't have anything interesting to say then enjoy your hobby website.

Google is now in a different league to any of the other SEs. AV is simply where Google was last month - it will take them years to catch up.

The index pages aren't coming back. Think about it.

Скажите как расчитывается Page Rank?

Наблюдаю массу контрпримеров

Наблюдаю массу контрпримеров