Following an article by Larry Constantine, I began a thread at Cre8asite Forums discussing the usability of open source software. The subject of the original article is the advantages (and disadvantages) of open source software – one of the chief weaknesses cited by Constantine is poor usability.

Now, I don’t have the kind of knowledge of open source software to be able to make that kind of judgement. Some open source software I know is quite difficult to use, but some is very easy. Some closed source software is very difficult, some is not. I’m not capable of drawing that line. Regardless of the reality, the impression that open source software is more difficult to use than its closed source counterparts is rampant.

I’m inclined to believe that people involved in the open source usability movement are those who are knowledgeable about the problems in open source projects. After all, it’s probably quite true that most open source projects are developed by engineers and code geeks – not usability experts. Some of these engineers are undoubtedly thinking about usability: but certainly not all. Establishing a movement to improve usability can hardly be a bad thing!

But is this usability challenge an illusion?

Ruud, at Cre8asite, states:

My guess is that poor usability isn’t always recognized as such in closed source though. The difficulty a company has to use that $75,000 CMS is expected: of course such an expensive system is “complex”…

But the strange clickthrough path in a freebie open source project? Bad usability, cloaked functionality…

To what degree do our expectations cloud our judgement? Is open source more difficult to use, or is it just that training and thorough documentation is less available? After all, the thousand hours of training required to install, administer, and use some complex software packages is simply not available for an open source equivalent: there isn’t necessarily a company established behind the product to provide these services.

On the other hand, Rashmi Sinha states:

Open source software projects are often feature driven. People join a project, they want to contribute. How can they contribute – by adding a feature. The user experience becomes more and more complex over time, as more and more features get added.

So open source projects are subject to feature bloat. Well…I’d argue this is equally true of many closed source projects. Many, many software packages add features in order to be able to ship a new version: in order to sell again, they need to offer something more. Perhaps this is less a division between open source and closed source and more a division between good project planning and oversight and bad.

I’ve just done some moderately substantial updates to my script for fulltext searches in PHP with MySQL. Although I don’t think the changes merit a whole new version number, I’ve made enough changes to justify releasing it to the public, I think! Mostly, I’ve focused on making it easier to implement, although I’ve also made changes to make the actual searching better, as well.

The principal changes include:

• Extended the Boolean capabilities to include the full range of MySQL boolean query options, including the “less desirable” and “more desirable” modifiers (<, >), complex queries using parentheses, and wildcard searching using the asterisk.
• Added options to display results in either a definition list or a table format.
• Made it far easier to customize the script to your own database: it’s no longer necessary to dive into the script and rewrite the MySQL SELECT statements or customize variables.
• Added an option to allow the script to search on email addresses. This has been requested a couple of times, so I figured I might as well provide it as a choice.
• Made a lot of settings controllable without diving into the script itself: heading levels, extracted text length, whether to give me attribution…

My next goal is to improve the overall security of the script: but that will have to wait until later. Next week will definitely not be a week for programming: instead, it will be a week for moving.

I welcome bug reports: I’m sure they’re there – they just need to be chased down.

Read further details about the script.

Download the script now.

Kelly Ford, from the IE Accessibility Team, posted today in the IEBlog about IE7′s expected behavior with a variety of commercial screen readers and screen magnifiers.

The general sound of things is that IE7 will be compatible (or mostly compatible) with the most recent versions of most assistive technology software.

Although the number of products detailed on the IEblog is quite small, it does cover the better known products in assistive technology, all of which either are currently compatible with IE7 or will be patched for compatibility within the next month.

The information in the post is brief and provides little information about any kind of advanced functionality: however, if you want to verify whether your assistive technology will be compatible with IE7 it’s probably best that you check this out — before you install IE7!

This is cool. Basically, this is a tool which studies the tag structure of your site and communicates it as a color-coded series of interconnected tags. Although this has little to do with web accessibility, it is an interesting way to visually communicate generalities about the structure of a website.

If you read the original post by Aharef (get the funny joke in his handle?), you’ll get the opportunity to look at a variety of interesting graphs of major sites: BoingBoing, Yahoo, Google, and so on.

Although part of what I like about this is the elegance of watching the tool gradually display the structure of your page, I also find that it provides an interesting view into the overall style of building a page. A very focused site, such as BoingBoing, has a very straightforward graph. A vast density of links, but little else. The sprawling monster which is Yahoo.com provides an equally sprawling graph.

Here’s Joe Dolson Accessible Web Design:

Here, you can see this site in all it’s glory: sparing of <div> tags (green), the site predominantly consists of blue and gray tags: links and “other” tags. The “other” category includes <p>, so that’s probably the majority of what you see there. I feel pretty good about this graph: it’s straightforward and simple.

Still, the tool could be made even better: I’d love to be able to produce graphs where I chose the color coding pattern for particular tags. I could set all non-semantic tags to be bright red, to easily spot the condition of a site in that respect. I could focus my attentions on inline versus block elements, or I could differentiate between different levels of headings.

Or, of course, I could look with an eye to the prettiest graph I could come up with for my site.

This post has nothing to do with IE7‘s support for CSS, adherence to web standards, or general functionality. Instead, it’s a general complaint about one specific annoyance: why did Microsoft’s IE development team want IE7 to “click”?

I mean, as a marketing metaphor, it’s great. Everybody wants their product to “click”. However, in this case I think they’ve taken it a bit too literally.

Read more: Internet Explorer 7: Irritation of the Day

My little boolean search engine has been receiving a fairly substantial amount of traffic, so I thought I should make it a little bit more user friendly. I just uploaded a new version of the zip package which includes a template file to connect to your MySQL database and, most importantly, a little readme file with some very basic instructions for installing and modifying the search engine for your site.

Finally, I put up a example implementation so you can try it out for real. Also made some minor changes to the script itself, but no real spectacular changes – it should be a TINY TINY bit faster now, but that’s about it.

Edit: put up the wrong link to the search script. Oops!

Well, it’s a work in progress, so it keeps mutating bit by bit. Today, I fixed a minor bug which caused some images to be produced at the wrong length and added the ability to use a wider variety of colors for your poll images. No longer are you restricted to monochromatic! The whole rainbow is available. Aww, how cute.

At any rate, the new package is available and you can download it as you please.

Afterthought: Oh yeah, I forgot. I also added a minor feature which prints the vote percentages after the images.