But I'd like to see people more worried about measuring the quality of CLOSED source code. Now *that* is a challenge for you...

A spokesdroid for the BSA said "How are we supposed to get rich if the government stop using proprietary software that uses proprietary standards that we own and change on a regular basis. Since the dollar went on the devaluation slide the European operators having to pay to be on the upgrade escalator for our software is the only thing keeping our revenues going. We have no patience at all with the stupid and naiive argument that software is there do perform a task for the user, the user is there, under license and threat of prosecution, to fund the software companies. The sooner governments join corporates in realising how a market economy is supposed to work the sooner we can all buy Fijian islands and retire. How are we supposed to manipulate, distort and control a market that has open standards not based on proprietary IPR? Compete for customers? Don't be ridiculous!"

I was expecting some old turfage along the lines of "I thought you said that OSS had no bugs!!!!oneoneelevenelevenone1!!"

How in God's name do they get SQO-OSS to be pronounced "squash"???

hmm, am planning on giving this a go. from a quick look at the INSTALL

etc it looks like its as easy to get running as a typical OSS - ie overly

complex and requiring far more hours of life than is worth it ;-)

seriously, looks cool. would be nice to have a central site in which

.tar.gz files could be uploaded for evaluation etc!

How does this produce any metric which can reasonably be associated with quality (whether fitness for purpose or adherence to a quality system)?

It's nothing more than a handful of trivial stats, all of which could be garnered by one-line perl scripts!

Have you looked at the online demo? there's a link on the homepage, on the right in the 'software' box.

Worth it. Basically it shows you that the UI is so badly designed that the 'thing' is almost impossible to use.

You have first to go to the Metrics page (see link on the right), select some metrics, go back to the files, enter the right directory then nothing shows up! Because, you must also click on a smallish icon on the top left of the window to activate the metrics. Who designed that? Here is a hint guys: such tools must be efficient to use, not a click-fest so I get a few result. When I use such tools I want a comprehensive result, I won't spend my time clicking through all the directories to get an overview, I want the tool to do it for me. That's why we use tools!

I've used a program (Cobertura) that provided code coverage metrics on my previous Java project. The tool was a bit difficult to set up but at least provided a few interesting results: starting at the root level, then you could enter in the details of the various packages/classes. Go there http://cobertura.sourceforge.net/sample/ to see an example.

And don't get me started on the user documentation. They don't even mention that you must click on the results icon.

So sad...