I am helping to organise a workshop as part of RSEcon24 specifically on maintaining Fortran code. Registration is free, and it would be wonderful to gather as many perspectives on this as possible from this community.
A huge amount of research software is, or relies on, Fortran code. However, maintaining Fortran code is fraught with difficulties for a variety of interconnected reasons, including lack of suitable tools, the prevalence of ‘hero developers’ in scientific software, and a dearth of training. In this workshop on sustaining Fortran research software, developers, software engineers, scientists and other stakeholders will work together to 1) better understand the landscape of problems 2) share good practice and 3) plan activities to support the community.
Just to add: We do have some funding to support attendance. If you want to avail of that, please get in touch with me directly (andrew.brown@qub.ac.uk)
Welcome to the forum! I have not looked in any detail yet at the website you mentioned, but I did not see this workshop announced. Is there some programme yet so that we can form a better picture?
There is no programme yet- we will be meeting with the facilitator in the next several weeks to work that out.
Importantly though, what we are planning will not comprise talks and presentations, but rather facilitator-led discussion and activities to help map out the current situation, and to plan for the future. In that sense, the contributions we’re looking for are not formal presentations, but people’s ideas and experiences.
@abrown41 the workshop is free though the parent conference has a registration fee. So do we need to register for rsecon24 to participate in the workshop?
Hi @rfarmer (I am a co-organiser): as far as I am aware, there is no requirement to register with the conference to attend the workshop. You can ‘just’ come along to the workshop.
Sometimes tools are available, but organizations don’t buy them because they undervalue the time of their researchers/programmers. Some commercial tools for checking Fortran code are below. The NAG compiler is also well-regarded for checking program correctness. Could you be more specific about “lack of suitable tools”? In general, I wonder if maintaining Fortran code is more difficult than for other languages, except for the issue of finding Fortran programmers, who are less numerous than Python or C++ programmers.
Codee: automatic code inspection platform specialized in performance for time-critical C/C++/Fortran applications
Cleanscape FortranLint: static source code analysis tool that reduces your organizational exposure to risks from latent software problems by automatically identifying problems at their source – in the Fortran code prior to compiling or executing programs.
Forcheck: oldest and most comprehensive Fortran analysis tool on the market. It performs a static analysis of a Fortran program or separate Fortran program units.
fpt - Tools for Fortran Engineering - Error Checking, Measurement and Assessment, Structural Engineering, Software Migration, Run-time Testing, and more.
plusFORT - multi-purpose suite of tools for analyzing and improving Fortran programs, free for personal, educational and academic users working on non-commercial projects.
Understand: static analysis tool focused on source code comprehension, metrics, and standards testing, supporting standards through Fortran 2008
I agree: “lack of suitable tools” doesn’t mean that the tools don’t exist. Speaking from my experience in academia, I can say that 1) most of my colleagues think that ‘tools’ == vi and 2) even where people are aware of tools, the way that funding for research is set up, either people can’t or won’t spend money on tools, because software maintenance is often not seen as a valid use of an academic’s time.
FWIW, I would consider it extremely valuable to get your perspective on how these tools might be used in the contexts I’ve described- so perhaps you should come along to the workshop
Because the format is not based around presentations, we won’t have an online option. However, we will run a follow-up event online to present the outcomes of the workshop, and to glean more input from the community more broadly.
Thanks for the invitation, but I am not an academic and do not have a travel budget.
For some time I have been maintaining a list of Fortran tools – the subset I posted were the commercial ones. I have previously mentioned it on the forum.
We developed and maintain fpt. I can attend the workshop, though Newcastle is an interesting place to get to if you start in southern England . What I would like to learn are:
i. What should we be doing that we are not doing already?
ii. Why do so few academics use the tools (fpt is free to academics)?
iii. How important are Windows and MacOS versions? Everyone we work with uses Linux but we don’t work with everyone.
iv. All kinds of things I haven’t thought of.
Will PlusFort and any of the other toolmakers attend? It would be a pleasure to meet.
I don’t want to attend the main conference, but is there advice on accommodation?
As @Jcollins mentioned, let us know what should we be doing that we are not already doing, and if you have some tips for improving, or are able to persuade more contributors to join, that would be awesome.
Hi @Jcollins: thank you for your message. These are exactly the kinds of questions that we want to explore, so I’m delighted that you would like to attend.
There is no need to attend the main conference, of course, but as we’ll be in the same venue, the recommended accommodation is all the same. (The discounted rates are only available for attendees of the main conference, unfortunately.)
and last but not least LFortran a modern, actively developed Fortran compiler, packed with features like interactive execution and code conversion to Python,C++, web assembly and many more.
I also think it would be valuable to mention some of the challenges we have faced over the years in the development and areas where we could improve. Unfortunately, I will be out of the country. Is it possible for some of these tools to be mentioned in the meeting? I would be very interested to hear feedback on what users like/feel excited about and what could we do better.
@gnikit , I’ll be there, and will make sure I bring up all these things that Fortran-Lang has been working on. It is definitely worth pointing out all the progress that’s been made recently on tooling for Fortran.
Thank you to all those who came to our workshop last month. It was really inspiring to see so many Fortran devotees in one place!
As promised, we are going to run a follow-up session online. The idea is to review the outputs from the workshop, discuss subsequent process, and gather input on how we might proceed. Thus, we hope this will be useful both for people who were there, and those who weren’t. You can thus consider this an open invitation to the meeting.
We will hold two sessions, allowing for availability (and time-differences, to some extent!)
I can’t seem to share a calendar invite without violating some data protection regulation or another, so those links will simply launch the meetings in MS Teams.
