It is vital that when you start your site that you choose the best possible platform for it. After all you are going to be "stuck with it" for the next five to ten years.

Choosing the right platform does not end with understanding not to use blogger.com but starts there.

To maintain maximum control over your site you need a platform that you can control. It must also be a platform that provides not only what you need now but what you are going to need over the next three years.

It needs to be able to grow with you.

Using a heavy and easily broken platform like Drupal or XOOPS might be fine for a portal but a simple blog or business web site needs more stability and less bloat.

The first thing to do is not to visit opensourcecms.com even though it is a great site.

The first order of business is to make a list.

While you are there make two lists.

1. Things I want or need to do with my site.
2. Things I might want or need to do next year or the year after.

Now you have a check list.

But it is not ready to be used. You need to know what parts of that list are "things it would be nice to do" and what parts are "if we don't do this we might as well give up".

Or to put it into more formal language identify your core and secondary aims.

Now you have a check list you can use.

Now you can visit opensourcecms.com or some other quality list of software. Look at the features of each CMS, Blog, Portal, wiki (or whatever you are looking for). If it does not have every "core requirement" as part of the standard version (or with add-ons or plug-ins) it fails.

It does not mater if it has ever bell and whistle ever invented and it makes the tea and gives you a back rub too - if it does not cover the basics first it is no good.

Now hose that pass that test can be scored on the basis of how many ideal extras (secondary requirements) it sports. Be firm here - if it trims your toenails for you but it wasn't on the list then that feature scores nothing.

Don't be dazzled by the shininess of the feature list. you are selecting a tool, a weapon if you wish, with which you are going to change the world. It must be ft for the job.

Now when I last did this I made a little table on a spread sheet and I rated all requirements from 1 to 10 depending on how important each was. Then I hid that list and rated how well the packages met my needs on a scale of minus one to two. Where minus one made things worse, zero did not help and one did meet the requirements. A score of two meant that it met that requirement very well. I must admit one feature scored a three it was so good.

Then I multiplied the importance and the meeting of that need and added it up for each possible solution.

A rather bland but functional product was the clear winner. The one that had scored the three had scored it on an unimportant issue but had failed to impress on some key issues and had the second worst score of the lot.

When I started to use the bland choice I found that it not only worked really well but continued to work well whee other more shiny looking solutions would have fallen on their face.

Almost all platforms that you will consider before you set up can be made to look like anything else (This Blog looks like the wikipedia!) and so the look of the "default" theme is not a consideration.

The smart question to ask is "how easy is it to deign a new look?" possibly followed by "could a non programmer do it?"

If you are thinking about setting up a company website that list of requirements and the chat of ticks you created while looking for solutions is 75% of the report you will want to submit to the CEO, MD or whoever you report to.

Get it right now and you can enjoy years of easy use and praise for your good choice.