URL: Directory like or Pages with extension

Question

If you look at the url of the stackoverflow.com you can see that the url is like:

http://stackoverflow.com/questions/ask

or

http://stackoverflow.com/questions/1284899/designing-a-website-for-both-javascript-script-support-and-not-support

or some very popular websites like wikipedia have the same format (en.wikipedia.org/wiki/MySQ)

So what's wrong with old .php .html index.php or index.php?show=ask-question ? what's the point in building the website in this format? better search rank? better surfing?

Answer 1

Reason 1:

Human readability Way easier to understand /articles/article_title instead of articles.php?article_id=31204513

Reason 2: SEO

I'm pretty sure that most search engines don't really allow querystrings to be indexed. Friendly URLS is one way around that.

Answer 2

Multiple reasons:

• Better SERP rankings. Especially a while ago the part behind the ? was more or less ignored by Google and thus made getting your entire website into the index very hard. It has changed, but I guess there still is an advantage
• Better readability. Those "pretty" URLs make it much easier for a user to understand what the page is about. Posting such a link somewhere for example makes it very easy to guess the contents of a page if it's shaped like your examples
• Making the URL of a webpage about what it does, not what its technological infrastructure is. Having a .php file in your url does not give any information to a normal user.
• Leaving information like that out of your url makes moving to a different infrastructure (e.g. switching from PHP to ASP)much easier: Good URLs never change

Answer 3

Search engine optimization. Search engines supposedly look for search keywords in URL-s and give better rank to pages that have them, and rank down "generated" pages (with ?....) and ranks up static pages ...
I say "supposedly" because nobody knows (except someone at google, or bing or whatever search engine) exactly how it works.

Answer 4

Depends on the technology. Servlets appear to look like directories. In spring you can specify whatever URL path you want to refer to a controller as well. Newer technologies tend to hid, how they are implemented. Spring does this by using .html extensions. Things like this are done for security reasons, as the less people know about the inards of your website, the better.

Answer 5

These sites are using server-site url rewriting rules. Such url is readable and understandable even for no-programmers. Security is increased, too. That way you hide your get parameters, so people can't mess with them and do sql-injection, e.g.

Answer 6

Another reason could be security - you are not explicitly revealing what technology powers the site.

Answer 7

This is more general than just building websites. This is good practice for any large set of files you have to organise. http://stackoverflow.com/ is the root of the web server, the root of the filesystem as far as the web server knows (not over-complicating with virtual hosts e.t.c.). stackoverflow.com/questions/ will go into the directory called questions. Usually these directories will contain index.html or index.php and a bunch of other files|scripts|pages that are required for the main page to run properly.

I don't think it affects your ranking. It only impacts your surfing if you often type exactly what you want in the URL bar rather than clicking on links from the main site. Even then, the impact is minimal. For the administrator/maintainer/developer of the website this system of building using folders and sub-folders is indispensable for any modern website.