Being Trendy

Browsing through iGoogle widgets I have found a few quite interesting ones.
They are meant to show the latest Google Trends, unfortunately some of them seem to be broken.
I have decided to give a try to one installed below.
It shows the Hottest Google Trends, that means the most searched terms within the last hour ( or at least as often as G updates it).
The G Hot Trends page displays the top 100 searches, and in the gadget you can see the top ten.
Google Trends started in 2004, and since that time proved to be unvaluable tool for marketeers.
It allows to chart, and compare volume of particular search terms, and the data can be broken by country, region, city or language.
It is possible to refine the main graph by region and time period, and with some keywords are displayed on the timeline the news, that affected the search results in the past.
The picture shows the values of seraches for keywords hotel ( blue ) and travel ( red ).
You can easy observe seasonal drop of searches for word 'hotel' in December, and dramatic rise at the beginning of the year.

[picture from www.google/com/trends]

Free SEO site analysis at Cuwhois.com

I would like to reccomend another site with free SEO tools :
www.cuwhois.com [not a link here]
The site is in Spanish, but easy to understand.
The information about the domain is presented in a nice form,with many extended info, besides standard checks it tries to find LANG tags, and If-Modified-Since server responses.
It gives you some information about the host server IP, country,ISP and hosting name, also other places using the same IP. The only problem I've met here is that the tools preffer the www.domain.name formand have some problems with interpreting results , when your main page is simple domain.name, and you use redirection from www to non www.
The main page is given a SEO points 0-10, however the score is not explained.
The site is also checked for presence of robots file and various formats of sitemaps.
Google,AlexaTechnorati ranks are checked as well, yet it seems to miss blogs ranked in Technorati.
The extended tools allow you to check duplicated index pages, and Domain Hacks - nice tool for domain name variations like typo's and similar names.

The whole site is great, and became one of my tools ( in spite of some notes above ).

[note for myself on one of the domains I've seen that server responds with iso 8859-1 set, while meta tag on page states UTF-8 coding. I am going to check if it has any SEO importance.]

Web site promotion tools

On the site by instantPOSITION you can find some usefull tips and tricks for positioningand a couple of free tools. Besides standard IP checks one is particular interesting:
SEO Doc.
You can check you pages for keyword optimisation - online.
As usually the test is only a guideline , as search engines algorithmes are changing every few days.

White Hat way of introducing a new site

About a month ago I have set up the new site guides in Krakow and Poland .

Before the submission I'd checked the pages very carefully, as indexing the new sites takes huge amount of time, and any mistakes would last for weeks literally as G can visit the site only once a week.The other two even more seldom. The pages were checked by w3c validator for XHTML and CSS code.
Some pages I was still not certain about were given noindex,nofollow metatags and there were no empty links to non existing pages.

Another search has been conducted for duplicated content, to make sure that the text is original ( different from already published) and will be indexed by SE.
I am not certain about so called canonization effect but there were two other issues introduced.All the links to home page are referring to '/' not to 'index.php'. Also I used 301 redirection from cracow.name to www.cracow.name as we decided that www form looks better. Links on pages are referring to relative '/' not to full URL address.

I didn't put any outbound links( but the weatherunderground.com) . They are going to be added soon but slowly, one by one, to see the influence of the links, and to avoid 'bad neighbouring', as you don't know exactly what is 'bad' for big G. Of course there could be some links, but for now there aren't any, especially that I need the people to link to me , not 'vice verse'.

Next the robot friendly file robots.txt and sitemap were created. There are not necessary but sniffing bots are suspicious when they don't find any of these files. Later on the files are the 'must'.

I've submitted the site to three engines GYM and kept waiting.

Icarus flight - the headers

Now it appears that G doesn't like when the subdomain name is the same as the blogs name.
The blogs name is in headers that is similar to situation with same names for subdomain ,page name,
and h1 header in some free hosts sites.
Two or three the same names in page's key position seemed to be considered as spam.

Link power (2)

G too much attention puts into the different aspects of links.
The links are different and could be used in malicious way, but
the solutions G takes are a bit curious.
When you link to a 'bad neighbour' you are penalized, but it seems
that G penalizes more sites in a kind of a chain.
So if the site is penalized , the one that links to it gets penalized
but another one that links to the last one is penalized as well.
It is hard to trace exact degree of penalizing, but then I think it is a bit too much.
You can control the site you are linking, but not the next one and so on.
I am a bit suspitious about being penalized when a bad site links to you,
that shouldn't happen but...
This post should clear the situation, and soon will have another part.

Reciprocal links between blogs

Recently I have created links between my two blogs , so site1.blogspot.com
linked to site2.blogspot.com and vice versa.
The sites didn't dissapear from G SEO but no new posts were indexed.
As soon as I withdraw one reciprocal link posts were indexed again and started to appear in SEO.
The blogs had no other of a few inlinks , so maybe when they get more links
that situation will not be considered as 'wrong' by G.
That will be tested soon , so far I will think about three way linking and webrings.

Name experiment

I have decided to shorten my name to Rum in some occasions,
and see if it works.
Some pages were filtered and one possible explanation is that there were considered as 'spam' ?
or smth.
the situation is the word 'rumburak' appeared in:
a) subdomain name
b) page title
c) h1 header
d) inlinks
e) somewhere on the page
I thought about my name as a kind of a clip zapping all together , not as forcing the name itself
but soon I will see if dropping some R. will benefit in rank position.
---------------
So indeed, slowly but slowly the site has been cured. It appeared that could be spam site suspition, and I see some logic, but come on ! My blog URL is rumburak2008, created by Rumburak, so I wanted to name the blog Rumburak's Webmastering , but that was too much for the poor bots. I hope they allow me to use the full name someday.

smuggle

And once again I try to post something.
It is supposed to pop up in the results around number 100.
If not, that is the end.
-------------------------
(edit) Definitely it appeared in search results , otherwise I wouldn't
write again.
The pages of the blog have certain position in search for specific keywords
and usually the new post replaces the old one on the same or similar position.
If the position is different or not at all , it is a sign that something is wrong.
Usually it is a bad link or some keywords.

Robot wars

The link to a page pulled down this blog.
Then The page linked to this blog was pulled down as well.
As the blog has a link to a page I have to remove it again,
as now two drowning sites won't have a possibility to swim again.
This gets me the idea of automatic warner that checks the SERP results
if one site goes down, the rest will have to remove the link before G bots
would got them.
Robot wars.
Another idea the dynamic page that first shows if the site is in SERps
and then decides to show the link to it.

Is it domain or subdomain ?

I was just thinking about having a domain at co.cc.
And my guess is that it is not a domain you have there but
subdomain that is treated much poorer by search enginers.
So domains are :
google.com
google.co.uk
google.CC
And the domain is CO.CC
thas is completely different than say .co.uk .
Thus google.co.cc is only a subdomain where the homepage is co.cc.

You can actually buy google.co.cc that is not connected
with the big G that has its search engine registred as google.cc.
Interesting will be to see if anyone would buy it and keep itfor some time.
And I am curious what will be reaction of G.

Finally rumburak.co.cc is only a subdomain where the homepage is CO.CC.
As now G treats subdomains as subfolders of main site
indexes them much slower and generally with less care.

Google datacenters

The blog of royal pingdom has published recently the map of google datacenters
That shows 35 locations around the world , from which 19 are in US.
Not neccesary all of them are processing SE results, but totally there are
more than 700 different IP's gruped in a few dozens of C-blocks that
could answer your question.
Official google help page mentions that requests from the countries outside US
are redirected to local pages ( like google.co.uk ) , based on searcher's IP
however you can individually set your browser preferences.
The point is there are different caches of internet web pages , and the answer
depens on the location of the searcher.
Even the requests from one computer at various time are redirected
to different G servers , so the search 'results may vary'.
The subject is very interesting and I will return to it on various occasions.

Bad link

I think I've finally traced the G filter, that prewents this site from appearance in search results.
I've removed it.In the case of blog it should work within few hours.
--------------
And this worked.
Apparently one of my sites is filtered by G.
It is visited by bots everyday but:
Not shown in results under full domain name.
Not shown for rumburak keyword.
No info: only site: results.
I am in the process of cleaning the mess with that particular site.
So far the page on which is the link to bad site is banned from SE results as well.
Now I have removed another link to the site that dropped in SE results some 100 places down, and see if it has the influence on the overall position of the blog pages themselves.

Webmastering Tools

Recently I have submitted some of my places to G Webmasters tools
immediately they dropped in SE results , probably they are checked
more thoroughly, or loosing a position is suppose to equalize the advantage
you gain from using the tools.
First the tools are useful with checking any technical errors, as googlebot reports unreachable server pages or errors in robots.txt file.
Among the statistics the most useful is the one that shows the position of the site that appeared for the most popular questions people are addresing the search engine.
Sometimes you can be suprised under which keywords you can find your pages.
The rest is very slow.
I use subdomains on freehosts and although some are reached by bots everyday
the change in results appears after a week or so. The cache of the indexed page is changed after a month or even more seldom.
Conclusion : useful tools, but hard to judge now how much is better with using them than without.

Is 'robots.txt' file really neccessary ?

Robots.txt is a small file placed in home directory, that is suppose to tell the crawler bots to
index or not given pages from the site.
Examples:
#robots.txt that tells the bots to go away
User-agent: *
Disallow: /
#robots.txt that prevents google from visiting /secret directory
User-agent: Googlebot
Disallow: /secret/
#robots.txt that allows google to see everything
User-agent: Googlebot
Disallow:
If there is no robots file or there is the empty one all web bots are welcome to index all pages from the site.

So the file is usefull in the case you have duplicated content on some pages
that you do not want to be crawled,blocking these pages would improve overall rank of the site.
Also it works when you have some temporary or 'secret' files you do not want to be indexed.
But beware, this file become the prime interest of vicious people who are trying to find hidden directories on the attacked
site. Using WordPress you don't want to expose the content of say '/wp-admin/' directory, but when you disallow it in robots.txt file , suddenly you give information of presence of this directory, that otherwise would be hard to guess,
as there are no direct links leading to it.
To disallow the robots from the specific page you can use
meta tag placed in the HEAD section of an HTML page
such as
<name="ROBOTS" content="NOINDEX, NOFOLLOW">
Then if you don't want to exclude the whole directory from the robots view, you don't need the file robots.txt .
That was the old technique of positioning based on different pages prepared for different bots, that could be
directed by lines in the robots.txt file, so every webbot would reach the different pages, and don't see the other prepared for other crawlers
but it looks that this technique now is not very effective.
That lead us to the topic question how this file in present form could be usefull nowadays, and is it neccessary to keep this file on the site ?

subdomain vs. domain

The advantage of sites with full domain name over sites that use subdomain is obvious.
For web spiders main page from http://rumburak.hostsite.com is treated as a subpage of domain which homepage is http://hostsite.com , so has lower position from that it could have if it would be a separate domain http://rumburak.com.
Now subdomain is treated as a subpage of a domain, thus subdomain is as strong as the main site, that could be bad if the domain hosts other vicious subdomains affecting overall ranking results.
On the other hand, one positive thing is the age of domain.
When you introduce a brand new domain say , rumburak.com it takes some time when it will be recognized by main SE. Search engines like old, mature domains and tend to rank them higher.
When you create new subdomain rumburak.hostsite.com it gets the age from the host, and is treated by SE like the old domain.

Link power

As always not everything is clear, but the overall positon of the site has the influence on the links connected with that site.
In theory you are not supposed to be punished for the fouls of the others but if the low quality site places a link to your page it could affect your position as well ( lower it).
More important is not to have on your site any links to low ranked or banned places.
These links will fuit in penalization of your site.
Now before I place a link on my page first I check if the site is in the G rank and how is it's general reputation.
No proofs of the theory that the text surrounding the link has any influence, on the other hand the anchored text does, so there are the suggestions to put some keywords into it.