11.00 a.m. Thursday 15th May, 2003
Stanley Gill Room, BCS, 11 Mansfield Street, London W1G 9NZ
(nearest underground stations: Oxford Circus, Gt Portland St, Regent's Park)
A G E N D A
(For any latecomers - this will probably be at the Dover Castle, Weymouth Mews, W1G 7EQ)
|14.00||Progress on Fortran 2000 - almost complete!|
|John Reid, WG5 convenor
The recent meeting in Las Vegas considered all the comments submitted with the ballots on the Committee Draft of Fortran 2000 and agreed on how to respond to them. The intention is that only editorial changes or very minor technical changes be made from now on. In effect, the content of the new standard has been chosen. It is planned to submit the Final Committee Draft for ballot in September and the Draft International Standard in June next year for publication before the end of 2004.
In this talk, I will summarize the changes made in Las Vegas and explain the UK ballot comments and the WG5 responses to them.
|14.30||Porting Legacy Applications from the 1970s into the Twenty First Century|
|Alistair Mills (Rutherford Laboratory), George Mitchell
and Chris Dallimore (WS Atkins)
Many of today's technical applications for science and engineering were written in the 1970s. Many of these programs are robust and reliable pieces of software. However they generally have user interfaces which do not meet the expectations of today's users.
This talk will describe a project where the speaker took an existing Fortran 77 application and with very few code changes modified it to run in a Windows Environment using Fortran 95 modules and Windows Dynamic Link Libraries (DLLs). The application then interfaced correctly with Research Systems Interactive Data Language (IDL). This made it possible to build modern interfaces to the program, using a modern software development system. The result was a software system which meets today's expectations without major changes to the legacy application.
The talk will discuss the application area and provide a demonstration of the software.
The speaker will provide download-able materials to support the presentation.
|15.30||Why we don't convert systems to Fortran 9x|
|John Bray (Meteorological Office)
The Met Office were early adopters of Fortran 90, starting to develop very large new applications for observation processing and data assimilation in Fortran 90 in 1994, with operational implementation in 1998. We developed a variety of innovative and successful techniques, and were very impressed with the new language. So why 5 years on is our main forecast model still being actively developed in Fortran 77?