Over the last month or so we've written about Mumps, which generated a great deal of interest and prompted Lord Norfolk to sharpen his quill to pen a follow-up feature, and now a piece on Ada.

OK, so these 'old dogs' will still have their place, but the level of interest poses an issue - is the future for these old languages more than just a slow decline into that final, write-only archive? Is there instead the possibility of a new future?

The signs for the latter are in fact encouraging. HP's porting of the old Tandem Non-Stop system to the Itanium processor looks like giving the system a new lease of life - not only giving existing users a current platform for upgrade options but also new applications. It is, for example, finding a new role as the secure back-end transaction database behind the SimDesk system - which runs unlimited numbers of cheap Linux servers and clients that can be anything that is capable of running a browser. This is already finding applications as a web-based environment where the Non-stop back end provides the bomb-proof insurance for the front end, even the front end systems collapse in a great and glorious manner.

It seems there could be potential for Ada in a similar role, and as it is on x86 processors it runs on the right hardware.

So are there some new tricks that these old languages and environments can learn - and then teach the young whippersnappers how to really do it properly? Given the growing move to online trading etc, having bomb-proof, reliable systems is now important.

Views anyone?