Link building question

M.A.D said:

Ranking is per document.
Thus not for the whole domain.
But if you have a good internal structure of pagerank flow, the root domain is the one you could best link to.
Unless your trying to get the page high up in the ranking.

I made a post ones about internal linking. Maybe it can be of help;
http://www.blackhatworld.com/blackh...ce-drawn-external-links-only.html#post1732613

Good luck,

Click to expand...

when i post on different blogs for link building purpose should i use only the main url www.xxx.com or all pages like www.xxx.com/xxx ?
does it make any difference ?

Click to expand...

M.A.D said:

I wouldn't know if it is true. But don't expect that it is.
Make sure your internal structure in order.

The simplest way i can explain is as follows.
Keep it clean, try to have everything organized with the least amount of links.
An example;

Imagine the image is a website structure.
In the image you see the links from the homepage flow down into a category and the category down into your pages.
home-->category-->pages.
Perfect structure, but you would also like people to go back to the home page, and/or category page ones they have reached the bottom of the tree.
The following example, represent a website with a top bar.
The top bar contains a home button and a pull down category button.
This way you can access every page and go back to every page.

Now this is the best structure, the least amount of links.
And all the links are equally divided, meaning not one page gets more pr then the other.
I will referrer to this as the base structure.

Now say i would add a block in my sidebar with some of the latest posts.
Handy for the users, but not for the search engines.
What happens is you link to some pages double, and it becomes one big blur.

This is just with two pages, imagin if you had a hundred.
Also some pages are way heavier on pr then others.

It used to be that search engines would stop crawling when they were confronted by double pages.
They'd say we've been here before lets move on.
You'd think why. Simply money. Crawling costs bandwidth and bandwidth costs money.
For you it might be a couple pages, but if it's on every website a search engine starts loosing money.
This also the reason why google couldn't crawl javascript in the old days, although it was more a cpu power thing.
One of the reasons sitemaps became so popular. It has that tree structure.
Which is easier for search engines to crawl.
This is what i tried to make clear in the original posts with the image that wasn't exactly clear to you.

So what do you do to resolve it. You use nofollow links on internal links not vital to your base structure.
This includes login tabs/widgets/pages, submit forms, adds and the latest post widget example i gave you.
Also notable is that a lot of top bars use javascipt.
This can cause errors in the crawling. Make sure that if they are interactive, they consist 100% of css1,css2 or css3. (NO JAVASCRIPT!)
So normal Href links with css styling, or just standard href links.

WordPress has solved this problem really well.
In there standard Kubrick theme they don't have a topbar.
They arrange everything from the sidebar.
Although not perfect, it just needs slight modifying.
I'll just give you a quick example.

Also from default wordpress has your ten latest posts on the homepage.
You should change that to just displaying your latest and one sticky "about your website" for example, so it doesn't look empty.
Hope that clears some stuff up for you.

Cheers

Click to expand...

M.A.D said:

Linking from content for a internal structure is useless.
I attached three image to give you a basic idea.

Image 1 ; homepage, content and a category block.
You click on the category culture which leads you to a page with the posts that belong to "culture" category, this is Image 2.
When you click on one the post titles you go to the post content page being image 3.
This is perfect for search engines, however slightly less useful for normal visitors.
Image 4, a revised home page. Adding a nofollow latest post widget to make your users happy.

Very crude but in a nutshell thats it.

Click to expand...

M.A.D said:

Yeah should be fine;

HTML:

<a href="http://www.example.com"><img src="image.jpg" alt="my image"></a>

Although i'm not big fan image links.
Each image you add is another http request.
So the connection kind of goes like this;

first http request is for your code.
then the browser sees you use images and it has to do requests for each individual image.
Say you got five button images, thats five more http connection.
If use javascript, for something like google analytics then thats another http request.


This is very annoying when someone is on wifi, shared networks or iphone.
If i were you i would look into styling your <a href" "></a> with css.
There loads of tutorials out there on how to buttons from simple links using css.
It's less size, less http connection and easier to crawl for search-engines.

Click to expand...

M.A.D said:

Google can read flash too.
They can click a link, follow it,run it, and thus index it.

Click to expand...

M.A.D said:

Maybe because it isn't a actual hypertext link.
Actionscript is based on ECMAScript, same as jscript and javascript.
It is object oriented. In this case the button is the object and not a hypertext link, so there is no hypertext for google to index.
Rather then being a link, a event is attached to the button. In this case it's the onclick() event.
When the onclick event is dispatched, by a click on the button, a function is executed.
In this case a new page is loaded.

Now google can follow the actionscript code, but there is no actual hypertext google can index.
You can go to the webmaster help forum on google and ask the question.
Maybe they have a answer to resolve. And else they will probably add it to the bug queue.

But still........ flash can be read by google.
Not other search engines.
Thats why css is so awesome.
You don't need a player like flash. Every browser supports it out of the box.
Seriously, you can learn this within a hour;
http://www.w3schools.com/css/css_intro.asp

Click to expand...

M.A.D said:

No, it's not bullshit. You can definitely get a penalty for pagerank sculpting.
If you do pagerank sculpting, you are trying to flow the pagerank of all pages and concentrate into one.
For instance you got navbar links, comment and content links on multiple pages. But you only leave homepage links doflow and the rest noflow.
All the pr will concentrate on that page.
Using noflow to get a better site structure is not something you have to worry about, if you use it to improve your structure.

Click to expand...