Title: Letters to the Editor

It seems that Francis' column has provoked more than one email... The following came from Mark Grimshaw

Dear Editor,

After reading your column in April's C Vu I decided to download the MinGW software as I to have been on the look-out for a decent (free!) development environment that runs on Windows and Linux platforms.

I downloaded version 2.05 (build date 12-12-2003) and installed it on my PC at home. On first appearances the IDE seems "polished" and at first glance it felt as though I was looking at MS Visual Studio. The software does provide most of the bare essentials one would need to start writing C++ programs although I did find that the lack of a stack trace and suitable debugging facilities (apart from a basic Quick Watch) a bit annoying. Still the product is under development so I think I may have another look in the near future. I may also have a look at the Linux version to see if it has a little bit more functionality.

I think the idea of a project repository is an excellent one and something which I myself would be keen to use for learning. After using C++ about 3 years ago and feeling like my skills were improving quite well I have since been consigned (not through my own choice) to using other languages in the course of my daily work, most notably Visual Basic. To maintain my career prospects I thought about how I could go about getting back up to speed with C++ in my own spare time.

After revisiting my personal library of C++ texts (e.g. Effective/Exceptional C++ series) I found that my interest was wavering somewhat. I realised that my interest in the past was driven more by the prospect of delivering software to satisfy real users needs and not necessarily just by learning programming techniques. The best approach for me is to combine the two approaches to learning programming - i.e. the acquisition of techniques and their application to some real software solution. Coming up with suitable ideas to maintain interest in one's spare time is very challenging, to me at least. Not only that but different individuals' learning motivations can vary widely - some people are driven purely by the act of writing software for customers or creating software that relates to another (domain) activity and others purely because they find programming a satisfying mental exercise.

Others, such as college/university students, may be motivated simply by the prospect of gaining qualifications. Working as part of a team can also be quite inspiring. What might be boring to one person is not necessarily so to someone else. I think this means that to cater for as wide an audience as possible, any repository should ideally cater for different learning motivations - in the first instance it might be possible to list projects that serve to highlight good practice and techniques but as you pointed out in your article, challenging, practical projects are also invaluable. Although I haven't looked at them for some time, the mentored developers projects might be good avenues for people to learn. Although I haven't had a thorough look at it, Sourceforge seems to be a another possible outlet for people to work on something interesting and as part of a team.

In sum, I for one would be only too willing to pass on any ideas for projects that I come across.

Mark Grimshaw