Some working knowledge of J2EE or JSF is assumed for this article.

Like some of you I've been frustrated with this technology known as JSF or Java Server Faces. There are several different flavors out there that are built on the shoulders of JSF. For instance Oracle's ADF (Application Development Framework). Oracle ADF Faces Components is a set of over a 100 JSF components that let you build a richer user interface for your Java EE applications. Oracle ADF Faces also includes many of the framework features most needed by JSF developers today.

That is great, and in many ways it will make life easier to develop in a JSF environment. Some items you will find available in these "add on" packages have a real benefit. For instance, as of the date of this article, I was very surprised that a File Upload is a feature still not implemented in JSF in respect to using natural jsf tags. There are ways to accomplish this task in JSF but they are not native JSF approaches. The process is a "no brainer" in just about every other framework available today, including asp.net.

Another simple task (I thought) was having a data table present the results of a query in an editable format. Possibly to update a user record or shopping cart. After working in other technologies it was very efficient to return a result set to a data table object and let that object take care of some of the trivial behaviors and characteristics of the table itself. When I started exploring JSF I was frequently and at every turn becoming more and more frustrated in trying to duplicate some of the most basic of processes similar to managing records through data tables.

There are not a whole lot of resources out there yet on JSF although it is growing steadily, and I found that all too often the resources that I was finding on the internet either didn't apply to the more simple tasks or the information was just completely wrong. One example of that was that it was stated in one article I read that you cannot use command buttons inside of a JSF data table. The recommendation was to use JSF hyperlinks instead when trying to perform an action from a data table due to a bug in the framework that prevented command button actions to fire if the button resided inside of a data table.

At first I thought "you've got to be kidding me"! Then I remembered that I've been finding a fair amount of "bogus" information in regards to JSF development so I decided to do further research and discovered that information to be less than accurate as well.

I simply had to find a way to populate a data table through a result set and get a command button to fire an action and pass all of the data in the data table to the backing bean to update the record. Multiple command buttons would exist as well as hidden fields pertaining to id numbers and so forth. Pretty basic stuff and we've all done it before with relative ease.

It turned out that the solution was in fact a simple one. "Binding". You've heard about it and read about it. But this approach was something a little different as far as I could find.

Many of the blogs and articles that I read dealt with passing the values as parameters and following the steps to define the parameters in faces.config files etc., then retrieving the parameters in a backing bean. Processing the passed data required another set of procedures to utilize mapping to each of the field parameters passed and then processing could begin.

That seemed like a whole heck of a lot to me just to retrieve form data. Then it occurred to me that I should be able to "bind" a text field component on a page to a backing bean. Once it is bound then all I have to do is extract the data. And that's all there is to it. My query returned results and pre-populated a data table including text fields with the values of the query pre-populated in the text fields.

Each one of those text fields was bound to a "HtmlInputText" type in my backing bean. It was not a String type like other approaches define. Doing that does require you to map parameters and populate that String variable through your set methods once the form is submitted.

What I found is that if I bind my text field to a property of text field type that it solved my problem of passing values from a data table, and I didn't have to define parameter fields anywhere in any xml file. Now that I had that figured that out I needed to figure out how to get the value of that property that I've bound my form field to, well why not getValue()?

I personally hadn't seen any examples online or in books for retrieving the value from a form binding it to the type of form element it was and simply use the getValue() to pull out the value of the object. Used like this getParameterFromForm().getValue().

Let me clarify that I am NOT saying that trying this approach isn't documented anywhere, I find it hard to believe that I've had some stroke of genius that no one has had before in the world of java, especially since I'm coming from .NET and ColdFusion.

After doing that I no longer had any problems passing form data to my backing bean. I was able to dynamically populate data tables with any number of records including any number of command buttons within that data table and I didn't have to concern myself with remembering to define parameters in any other areas of the application.

So put simply you can bind your form fields to properties of the same type in your backing bean, and then extract the value of that object using getValue() if you prefer over utilizing parameter string mapping and similar approaches for processing form data. I found it to be easier and less time consuming which has costs associated to it as well.

There are two approaches to DIY Website Construction.

The first is to go about gathering and purchasing all of the components separately. The second is to buy a package that has all of the components and build your website yourself using the tools at your disposal.

When you think about what you will need for a DIY Website Construction, consider the following:

• Domain Name
• Web Hosting
• Templates or HTML program
• Keyword Research
• Content generation
• Autoresponder
• Sitemap submission
• Community or forum for support

If you don't do this on a regular basis, putting all of the pieces together for a DIY website may seem overwhelming.

But you don't need to put all of the separate components together to do your website yourself. You can buy a package that combines the best of everything.

One thing, though. Your best bet is NOT one of the domain name services. These services allow you to get a "brochure site" up quickly. But they don't help you generate a website that sells. And, unless you are putting up a website just for friends and family, there's no use in a website that doesn't sell.

Instead, you should look for a package company whose goal is to make your site sell. They're not there to sell lots of additional components, they want you to succeed.

For one price, you should get your domain name, web hosting, templates, keyword research, an autoresponder, email addresses, sitemap submission, statistics monitoring, and more. You also want to make sure that there is some level of support for the community whether it is phone based technical support or a welcoming forum.

If you're not ready to put all of the components of a DIY website together yourself yet, I recommend Site Build It for a package deal. It includes all of the necessary components and has a very supportive forum community.

To get the SBI package deal, click through to my Small Business Website Marketing Blog.

A question I often ask my target market is, "what's your biggest challenge with building your business online?" And one answer I get frequently is this:

How do I make my website earn money?

If you've got a business online, then it's likely you have some sort of web presence. Perhaps it's a one-page 'sign up for my list' kind of site, or a full-blown brochure site with a menu of choices, or maybe your site is in the form of a blog. Regardless of what kind of site (or sites) you have, if they're not doing what you want them to - and since we're in business, ultimately that means making some sort of a profit - then it's time to make some changes.

But how do you REALLY know whether your site is working or not? Here are four strategies to test and track your pages to know what needs fixing:

1. Let the numbers speak

I'm sure if the sky was the limit, you'd hire the most expensive web designer with the fanciest tools to create the most spectacular site imaginable for your business - and someday you just may do that.

For now, though, even an ugly site can make money. I can think of at least two sites right now that I personally think are not that pleasing to the eye, yet I know they are raking in the bucks each and every month, year after year.

So don't get caught up in having the most polished and professionally looking site. Focus more on having the pieces in place that will bring you the cash, too.

2. Give your site only one job

Think of each page of your website as a separate entity with one main purpose. It may be to sell something, or to sign up for your list, or to get people to call you for an appointment. Whatever it is, make it clear that that one thing is the action your visitor should take from being on that page.

3. Track your numbers

Most web hosts offer statistics that you can use to watch your numbers, or there are other web stat programs that you could use as well (such as Google Analytics). However you are keeping an eye on your numbers, you should be looking for two things: how many people visit your site and how many people take the action you want them to take.

From those two numbers, you can figure out your conversion rate, which tells you how many of those visitors took the action you wanted them to - like sign up for your list. If you make small changes to your page AND watch these numbers at the same time, you'll be able to tweak things to increase your conversion rate.

4. Make one change at a time

This is a very effective way to increase the conversion rates on sales pages and sign-up pages. Change just one thing - for example, the headline - and watch your numbers. Compare those statistics to your previous ones and decide whether or not to keep the change. (You can also do this via a split-test in your shopping cart, where the software does the number crunching for you.)

I see so many sites that try to be all things to all people by offering everything under the sun in too small a space - a website - and all that does is confuse people and encourage them to click away. Remember, 'a confused mind always says no', so always go back to the main questions when considering making changes to your website: What's the purpose of this page? What is the one thing I want my visitor to do here? And then design your copy around the answer. Keep it simple and you'll get better results every time.