Functions are used to perform a well defined task that is normally repeated at various places within a web site, web application, or other software application. The function sometimes needs certain information before performing its task, and sometimes returns a value to the calling page or program.

An example of how to use a simple function in PHP:

 function showMessage() {
echo "{$_SESSION["message"]}"; }

The above example shows a very simple function that displays the contents of a session variable called 'message'. The idea being that while a user is using a web site, etc, various messages are generated and stored in the 'message' session variable. Whenever the showMessage function is called, the contents of the session variable are displayed. So, for example, when the user logs on, you could display a message saying that they have done so. Likewise, when they log off, a different message could be displayed.

Calling a function

To call the showMessage function, you would just need to include the line:

showMessage()

in the web page (wrapped in php tags to distinguish it from normall HTML).

Passing values to a function

Often, a function requires one or more input values in order to perform its task. For example, take the following example that takes two numbers as input, and adds them together. The result is then returned to the calling web page or program.

function addNumbers($number1,$number2) {
$answer = $number1 + $number2;
return $answer; }

In the calling program we would have something similar to the following:

$number1 = "5";
$number2 = "6";
$sumOfNumbers = addNumbers($number1,$number2);
echo "$sumOfNumbers";

Obviously, in a real program you would not have the numbers hardcoded like this, but they would be obtained from user input or by some other means. Notice also that although the name of the returned variable is $answer, the calling program makes no reference to that. We could, if we wanted to, change the calling program to:

$answer = addNumbers($number1,$number2);
echo "$answer";

which might make things slightly easier to understand.

This is a question you will have to address. Many folks just a build a website but do not think out why they are really doing it. As a result they don't take it seriously and therefore they do not get serious results. This makes logical sense right?

A website in and of itself means nothing unless it has visitors and achieves a certain objective. So you have to be clear what you want to achieve as this will help you focus on your research and build a site that can achieve your objective. Quality research will help you build a quality website.

To build a business online you are going to have to build a relationship with your site visitors. You will need to show them how to connect the dots and you can only do this by providing valuable content that is benefit driven.

Furthermore you are going to have to show them solid evidence of how your product or service can help them achieve results. Connecting the dots means moving them closer to their outcome. Once they can follow this and see how it will work for them you will naturally build credibility which will result in you having a profitable website.

As mentioned when you build your site you want to clearly have your visitor's objectives in mind so that you can address their concerns and overcome their objections. If you can help them achieve their goals it will be a win win situation for all parties concerned.

I've done quite a bit of research into web analytics lately, and I have found a couple of things. Google analytics Is a great tool, but it cannot be the only tool an organization relies on. Because it is a purely JavaScript based, it is unable to catch browsers that do not have JavaScript enabled, this includes the search engine robots, some hand held devices, as well as a percent of users that manually turn this off. That being said we need a tool to parse the log files from the server and to give us information about these users, as well as providing us with "on demand" stats where as Google's stats are next day. The server logs also contain a wealth of other information that we are collecting automatically, so we might as well take advantage of it.

I have come to the conclusion that we need to use Google Analytics but we also need a tool that reads the server logs. There are quite a variety of tools that would fill this need. Google recently purchased Urchin who's web analysis product runs Google analytics, costs about $3k and will work with the Google Mini, Urchin also gives the ability to read logs from server software other than IIS (like apache, the software used for open source applications) .

Urchin will be releasing another version shortly, and this will be the first release since the Google buyout. Another option is WebTrends Analytics it costs about $4k. It Like Urchin will read both IIS and Apache logs, and has a web interface for users. The last product is NetTracker, it runs about $4k for the software, and another grand for support. it was recently bought out by Unica. The administration of netTracker is not the smoothest, and the support seems to have declined on quality since the buy out.