"Cu capãtul arcurilor, trecînd,
rãzboinicii
mîngîie roua..."

Yosa Buson (1715-1783)

For some time the seo specialists (and a lot of other guys that have some idea about search engine optimization) has recommended that URLs on the page should contain hyphens instead underscore because the some search engines were considering hyphens as word separators but the underscores were just letters inside words. But lately, at least since 2007 for Google, all main search engines are considering both as word separators. Some of them considered underscores as separators from the beginning. There are people that still don't know that: I've seen recent articles (written at the end of last year) that still treats underscores as letter inside words.

As a web developer I started to use underscores to separate words from the beginning because of the language: I am generating the URL from the title or the article (whatever the content – no matter if it's a blog, a news portal or an online store) and there are a lot of words that uses hyphens as the grammar requires in the Romanian language (I could give you some examples, but it would be meaningless for the most of you) and even some names contains hyphens.

As I told you in the earlier articles, the best URL is the one that's readable by both the humans and the search engines and it's the clearest and easiest to understand. From my point of view, these conditions are accomplished using underscores as word separators and the hyphens only where they are needed.

Wordpress, for example, started from the beginning with the hyphens as word deliminator and they are stick with it. That's a good thing, even sometimes I am not sure if the names in the titles are of two individuals or one name created from two other names put together.

There are a few things about the internal linking for a site after talking about the friendly URLs. The first is about canonical URLs. Canonical essentially means standard or authoritative, so a canonical URL for search engine marketing purposes is the URL you want people to see. Canonicalization is the process of picking the best URL when there are several choices and it usually refers to home pages. For example, most people would consider these the same urls: * www.webdesign-software-code-seo.com * webdesign-software-code-seo.com/ * www.webdesign-software-code-seo.com/index.html * webdesign-software-code-seo.com/home.aspThey look very similar, but technically all of these urls are different. A web server could return completely different content for all the urls above. When Google “canonicalizes” a URL it tries to pick the URL that seems like the best representative from that set. Depending on how your web site was programmed or how your tracking URLs are setup for marketing campaign, there may be more than one URL for a particular web page.The problem most search engine marketers run into deals with domains when they are not setup properly. In these cases the domain URL without the www prefix (e.g. webdesign-software-code-seo) and the domain URL with the www prefix (e.g. www.webdesign-software-code-seo.com) are considered individual web pages. Since both pages may be indexed by the search engines you could get hit for duplicate content and at the very least you would be splitting your link popularity.The easiest way to protect your site is to redirect all forms of URL your domain to one standard URL – a canonical URL. For example to force the use of www.webdesign-software-code-seo.com instead webdesign-software-code-seo.com or http://webdesign-software-code-seo.com you must have this lines in the .htaccess file (this is Apache specific, if you use IIS the lines should be the same using ISAPI filter). RewriteEngine On RewriteCond %{HTTP_HOST} ^webdesign-software-code-seo.com$ [NC] RewriteRule ^(.*)$ http://www.webdesign-software-code-seo.com/$1 [R=301,L]This response to the problem is resolved on the server side, just before the script is accessed. One other way to consider the problem is for the web designer to define inside the page a variable that, in our case, is http://www.webdesign-software-code-seo.com and to use is to construct all the URLs inside the page.The previous part of this article says about one way to create duplicate URLs. Another way to obtain such URLs is to when they are generated dynamically and there are flaws in the script. For example, when you generate the URLs from only the title and you put the same (or very similar) titles to your articles you can get the same URLs. I am thinking the words separators (hyphens and underscores, but this is another subject and I will write about it later).There are websites (like the ones create with the wordpress blogging scripts) that can generate multiple URLs for the same article. Some examples are the following ones: * http://www.webdesign-software-code-seo.com/?p=2 * http://www.webdesign-software-code-seo.com/my-blog-post-number1 * http://www.webdesign-software-code-seo.com/my-blog-post?print=trueAlthough it makes sense to enforce unique URLs for all your web pages inside any site as Google imposes a penalty on suspected duplicate content, its not always possible to do so. Google have been introduced a way to specify the unique URL (the canonical URL) of any web page by using a meta link tag. Setting this tag is simple: you first decide which URL you wish to use, and then add this link tag to the section of your HTML page something like the next example.Google claims that this would not be treated as a directive, but a hint that would be honored strongly. Additional URL properties, like PageRank, would be transferred as well.The next important thing are the bad URLs, I mean the URLs on the sites that can not be found, clicked, visited or submitted to social media. It may be a technical problem (server is down, the internet access with problems, etc.) but it also may be a design/maintenance problem: indexed links may disappear from time to time (temporary or for ever).The technical problems can be solved (and the most of them are solved in time), but the other problem depends entirely by you. When you are designing a page make sure that the URLs in the site does not have flaws and all them are working ok before putting the site online. Its no fun for a human to have a list of articles on the page and a lot of content in that articles and when he/she tries to access them find a 404 error page because the URL was not created all right, especially when from other part of the site the article can be accessed ok. The search engines acts in the similar ways: if there are bad URLS (404 error pages) the will not index it.When an article is deleted (from some reasons, depending of the owner of the site), there are 2 ways to react to that from the designers point of view: the URL will completely disappear (and the search engines will not find an already indexed URL) or the URL will remain the same and the content will be replace with a message similar with “this article has been erased”. From the search engines neither way is perfect, but URLs appearing and disappearing periodic with no message is not something good.From the seo point of view, bad URLs are also the one that contains special characters (spaces, apostrophes and other characters like –), but that can be corrected by generating or redirecting to friendly URLs.The internal link structure of a website works in the same way that external links does: in order to build a quality internal link structure a site has to add links on its sub-pages and these links need to contain the keyword or search term that is being targeted. There is no limit on how many links from internal sub-pages can be created... there can be 10 pages or 10 billion pages. When the search bot sees these links all pointing to a specific page (no duplicate URLs and no bad URLs) it is going to read the provided anchor text in those links. It will then assume that the page being linked to is important for the anchored keywords.Deep link building is a good way to separate great websites from the mediocre ones: linking to sub-pages of a website is one of the best ways to grow long tail search traffic. It also is a great way to built up the site authority.The first step when building deep links is to define the target (or the targets) of the site and them keep them focused all the time. For example, this site is dedicated to web design, software code and seo and you will not find here anything other than that. Usually the web sites are custom to their owner needs, and they have only a few targets (a limited number of subjects for the content and keywords). The bloggers can write about they want, but even they have to customize the subject and the contents of their articles accordingly to what people is searching.The second step in the process requires the identification of sub-pages that can be used to gain more search traffic and can help boost top level keyword rankings. Top level keywords are very competitive and they usually require substantial link building campaigns to move up in the rankings. A great way to boost the authority of pages that are targeting these competitive search terms is to increase the internal authority of those pages. One way to do this is by adding a contextual link to the sub-page that links to the competitive top level page. Adding deep links that point at these pages will boost their authority. The authority of you site will increase together with the the authority of the pages inside your site.In one sentence, make the important pages on your site as visible as possible (inside the site and on other pages). For the first part the web designer is the person that can help you (a smart one, with seo knowledge, can even give you some suggestions) and for the second one you should aks the persons that do marketing (or seo marketing) to do this job.

In the very beginning of this article I must tell you that an URL comes from Uniform Resource Locator (the initials) that is a Uniform Resource Identifier (URI) that specifies where an identified resource is available and the mechanism for retrieving it. URI is a string of characters used to identify a resource or a name on the internet. Such identification enables interaction with the representations of the resource over a network using specific protocols.

It is said that in popular usage and in many technical documents and verbal discussions the URL is often incorrectly used as a synonym for URI. I've also met a lot of people that are using the term and they don't know what's coming from.

The way you create the URLs inside you page is one of the most important methods to improve the search engine optimization. Friendly URLs means they should be readable by the humans, but this counts for the search engines as well.

There are two things that are important: the first one is the human point of view. How many of you will access an URL that looks like http://www.webdesign-software-code-seo.com/8548-954page.html and how many will access http://www.webdesign-software-code-seo.com/freelancers-wanted/ or http://www.webdesign-software-code-seo.com/freelancer-jobs/? The tendencies are overwhelming for the second type of URLs because they contains words that have meaning for us and that gives clues about what the page contains.

That's why it is better to use words in the URLs even for the search engines (the second point of view), especially if they are keywords (I mean the words that are the most relevant inside the content and the subject of the article, that the users may search on the internet). The best way to make a friendly URL is to create it based on the title of your article (and the date if this helps to identify faster the URL you are searching on a big site) and the underline the keywords using the heading tags. The search engines see the all this elements and it helps to index the page easier and to improve it's ranking.

The friendly URLs are always static URLs. There are a lot of sites that generate URLs like this: http://www.some-site.com/index.php?category=456&subcategory=12&article=44574 and that's called dynamic URL. When spaces, apostrophes and other special characters (like –) appears in this dynamic URLs it's even worse: when you try to put such link on the Facebook or StumbleUpon (or any social network) there are a lot of chances that the URL will appear broken.

From the search engines point of view the static URL will be indexed much easier than the dynamic form, and there is no confusion and no missing parts from what could be important inside the URL.

Also, one of the most important things is to keep it as shorter and as descriptive as you can (clean and simple). The shorter the URL is the more successful it will be, as for web rankings and for the visitors' use (this includes people copying the URL for link purposes). Also the twitter URL posting (for sharing information as well for backlinking) have became a habit in the last few years and a longer URL can not be posted there. More, if a URL string is long the weight and relevance of each word inside is diluted. If your URLs are keyword-rich, including too many other words or phrases means that the importance of the keyword is lost amidst all the other words.

For example consider the following URLs:

www.readmybooks.com/horror-science-fiction-temporal-travel-book.htm www.readmybooks.com/store/books/science-fiction/horror-temporal-travel/book6888767.htm

The first URL is succinct, has the relevant keywords and no surplus syntax to dilute the importance of these words, and is easy for somebody to read, copy, and paste. The second URL is more complicated and the relevance of the keywords are reduced.

URLs can be constructed using upper and lowercase or a mixture of both, and the servers make the difference between them. Mixing uppercase and lowercase (even when following standard grammar rules such as using a capital letter for a name) can make your site structure unnecessarily complicated. You also run the risk of losing visitors who forget to use the required capital letter through the URL and then can’t access the page. The most common way to post the URLs across the internet is to post them lowercase. If you are redeveloping your URLs and come across pages using uppercase you should create a permanent 301 redirect to a lowercase version to avoid confusion.

There are a few tools and methods used for the URL rewriting in order to have the friendly URLs, one of the most common is using the .htaccess file (this is the special file that sets up the deal for you, it can contain all sorts of directives for the Apache server. If you’re not using an Apache-based server, you’ll have to read your server’s manual on how to do it) put in the root directory of the site. It works perfectly with the Apache server and php (for example), so it works fine with all the php-based systems. I will give some examples how to work with .htaccess and the other tools later.

Finally, it is very important when developing and maintaining a site is that you must keep the same structure of the URL. Do not change the rules to generate the URLs after the site is online, especially if they are created dynamically for the entire site. Once one URL is online it must be the same as long as possible (until the site drops dead or the internet ends). The search engines do not like the sites whose URLs changes from one month to another (for example), so try to change (even correct when necessary) them as little as possible. Also try to keep the URLs the same on the entire site, not depending by it's sections, this will make future development much easier as there will be a standard convention to follow... if your URL structure is globally used the visitors will also find it much easier to understand how the information is organized and stored and they will find what they are searching faster.