Hi Thomas,

Makes sense. It wasn't clear from your original post that the technologies were spread out more and aren't in immediate need of integration.

Aloha,

+++ Rick ---


>Hi Rick,
>
>>No offense but this sounds like a massive maintenance nightmare. Best of breed is good, but when you're dealing with too many technologies it's hard to administer (especially for Web apps) and just as difficult for developers to work as they have to switch between technologies for their work. This means multiple learning curves and constant context switches during the development day (which is one of the biggest losses of productivity in dev). It's bad enough in Web dev in general with the required technologies (Web backend, HTML, JavaScript).
>
>you're absolutely right. The reason for using different languages in our company is, because our departments are specialized for 1 language each.
>That means: The department I am working in uses VFP. Our Web-Department works with Java and our department for system-near coding uses C#. So, the problem with learning curves appears only when a developers change departments which rarely happens. However, as soon as a department gets the order to use a new tool, which will happen for me in the near future, then there will be a learning curve for the new language.
>
>>Also as devs leave your team you have to replace with people who have that same range of skills which has got to be difficult as well.
>
>Well, ... yes :-) but like me most of the other developers are with us for more than 10 years (partly more than 20 to 30 years and have developed software for Nixdorf-Hardware in the 70s and 80s, changed to Unix/Linux and a new language in the 80s/90s and then switched to M$ and VFP in the mid90s.
>
>>It's good to diversify and understand technologies but IMHO mixing too many of them in single solutions or even a batch of solutions leads to big problems in maintenance disaster later.
>
>My intention for Dennis was not to bring him to using several languages at the same time. That could definitely end up in disaster. I wanted to make him aware, that he should not select a language for marketing but for its ability to develop the kind of software he has to build. For me VFP is still #1 for data centric applications, but .. 'the times they are a changing'... ;-)