Ok, here is a rant on a pet hate of mine…

It seems that most major web projects these days are more about integrating existing bits and pieces than building from the ground up. While it’s not as much fun, I do agree that this is the way to go.

Here’s what really bugs me though, and has bugged me for years. It seems the vast majority of software seems to be poorly thought out. However, that in itself isn’t the biggest problem. What really bugs me is that most people just accept it.

For example, I’ve just finished a project based on the open source CMS, DotNetNuke. While it’s not perfect, I quite like DotNetNuke. Some of the defaults on install are less than ideal (and this isn’t simple opinion, it’s common sense based on anyone who has done any usability testing), but they are easy enough to change. But here’s the catch. Most people don’t.

As an example (and not wanting to pick on these guys who do some good work), http://www.kodhedz.net/ is a company that specialises in making DotNetNuke add ons, and they have some good products. If you look at their homepage (which is of course built using DotNetNuke), you’ll see that the left and right boxes have a little “-” icon top right which allows you to collapse those boxes. This function adds no value to the page, increases download time, and will confuse less experienced users - all downsides, no upsides. So why leave it there? Because it’s on by default I’m guessing. Most dotnetnuke sites have this left on, yet it’s so simple to turn off.

Similarly, the default settings for registering on a dotnetnuke site ask all sorts of questions - address, phone number, etc, which are totally inappropriate 90% of the time. It’s about a 1 minute job to turn this off in the admin interface. So why do at least 80% if sites have the options left at default, and would be reducing their membership as a result?

Similarly, I’ve been assessing several e-commerce shopping carts recently for a project I’m working on. I won’t mention names, but one well respected piece of software I assessed made the user go through 6 pages from the point they decided to buy till the point they enter their credit card. 6 pages! I can guarantee you that you store sales will suffer as a result. It is possible, although not immediately obvious, that it’s possible to reduce 2 steps from that process. Yet, when I visited 4 or 5 stores using that software, every single one of them had the process unmodified. Why? Why do people leave these substandard user experiences in place, and ones that are probably reducing your sales by a significant amount?

Now, you could write this off as “oh, but they are small projects with limited budgets and inexperienced staff”. I wish that were true. I have quite a bit of experience using a high end content management software package (depending on options, can easily be over $100,000), and frankly, what I saw in deployments of that was no better - poor defaults which were readily accepted, and on projects with budgets running into millions of dollars in some cases.

I just don’t get it.

Rant over.