My son has the dubious pleasure of being the primary beta tester of T-shirt messages. This time it is the first draft of the model that will be called “Twat”. Turned out pretty OK…
Archive for the 'Testing' Category
The broken state of EU legal information on the web
In my pet project eurlex.nu I find a lot of weird stuff when scraping documents from the official website eur-lex.europa.eu. The most recent specimen – Final adoption of amending budget No 4 of the European Union for the financial year 2008 – has the publish date 80/80/2200. That’s almost two hundred years into the future [...]
Quick site performance improvement
I have been playing with YSlow, Yahoo’s tool for web site profiling, for a while. If you haven’t tried YSlow (which is a Firefox addon to Firebug) I recommend you try it right away. Install the Firebug extension first and then add YSlow. It is amazing how much you can improve the percieved site speed [...]
Does your webserver give HEAD?
In the process of constructing a crawler that finds and checks PDF documents on a website I discovered a lot of sites that don’t return information for HEAD requests. A HEAD request should return the same set of HTTP headers as a normal GET request only without the actual payload. The typical response seem to [...]
New release of the Ruby Accessibility Analysis Kit and online interface
The current version has some minor bug fixes that will speed up testing. The online test interface has been updated to support direct input of markup. This is for those of you unable to install Raakt locally. This means that there is no reason to skip basic accessibility testing of whatever you are developing! To [...]
A new version of the Ruby Accessibility Analysis Kit
This is to announce that RAAKT (The Ruby Accessibility Analysis Kit) has been updated. This release includes more accessibility tests and an initial mapping of tests to the Unified Web Evaluation Methodology (UWEM). Also, thanks to Derek Perrault RAAKT now uses Hpricot to parse the HTML document. This solves the problem where the previous parser [...]
Parsing ASP.NET sites with WWW::Mechanize and Hpricot
Users of Hpricot (which WWW::Mechanize is using as the default html parser) may have discovered that the buffer size for attribute values is set to 16384 bytes default. Typically this isn’t a problem, I mean who would put 16Kb of data into an HTML attribute? Well, ASP.NET uses a hidden input field to store view [...]






