This bit needs rewriting
Oh, sure I will!
Will do 
Sure!
Thanks for taking the lead.
in line with @JC001ās and @Lucretiaās comments, perhaps mention nvidiaās ātransition from C/C++ to SPARK for their security-critical software and firmware components.ā
4 Likes
Here is my rewrite of the āNotesā and āWhy does this proposal fit FOSDEMā
sections. I have also restructured it a bit and streamlined some parts.
Also, once again, should we ask for half a day and potentially increase our
chances? And is someone else interested in helping Dirk and I with the DevRoom?
Notes: Ada is a general-purpose programming language originally designed for
safety- and mission-critical software engineering. It is used extensively in air
traffic control, rail transportation, aerospace, nuclear, financial services,
medical devices, etc. It is also perfectly suited for open source development
with a fully open compiler (part of GCC), formal verification system and a
knowledgeable and vibrant community.
Awareness of safety and security issues in software systems is increasing. The
NSA recently published [1] a list of programming languages that are recommended
for the development of new software due to their memory safety and Ada was one
of the list (one of the three compiledb non-garbage collected
languages!). Therefore it should be no surprise that NVIDIA has started using
Ada/SPARK [2, 3] for their highest critical parts in their GPUs!
Multi-core platforms are now abundant and small, embedded devices are growing
exponentially. These are some of the reasons that the Ada programming language
and technology attracts more and more attention due to Adaās support for
programming by contract, performant and efficient code, high- and low-level
abstractions and support for multi-core targets. The latest Ada language
definition was updated last year, giving birth to the Ada 2022 standard. Work on
new features is ongoing, such as improved support for fine-grained parallelism,
which were introduced in the new standard. The Ada-related technology, SPARK,
provides a complete solution for the safety and security aspects stated above
while being fully open-source, making it stand out from other formal
verification tools, as Ada/SPARK code is compiled directly into ready-to-run
programs, which can even run on embedded systems.
More and more tools are available, many are open source, including for small and
modern platforms. Interest in Ada keeps increasing, also in the open source
community, and many exciting projects have been created with Ada.
[1] https://www.nsa.gov/Press-Room/News-Highlights/Article/Article/3215760/nsa-releases-guidance-on-how-to-protect-against-software-memory-safety-issues/
[2] NVIDIA: Adoption of SPARK Ushers in a New Era inā¦ | AdaCore
[3] When Formal Verification with SPARK is theā¦ | The AdaCore Blog
Why does this proposal fit FOSDEM: The Ada DevRoom aims to facilitate a space
for Ada and non-Ada programmers to learn and share the features offered by the
Ada language (such as for object-oriented, multicore, formal verification or
embedded programming) as well as some of the many exciting tools and projects
using Ada.
Ada is one of the languages supported by GCC with a focus on correctness,
maintainability and easy of use. It also comes with a sister language, SPARK,
which focuses on formal proofs and analysis. These topics are becoming more
relevant as time passes. Which is why a lot of new languages, such as Rust or
Zig, have been created and try to improve the state of software development.
The tooling around Ada is open and has seen a great deal of development in the
past few years. Developers have also been creating new programs with it and
FOSDEM would be a wonderful choice for them to present and share their
work. FOSDEM is also used as a meeting opportunity for the Ada community to come
together and share their passion among fellow users and curious viewers who may
want to learn more about the Ada language and ecosystem.
4 Likes
Even though I havenāt been able to (physically) attend FOSDEM in the past, I have considered it each time thereās been an Ada DevRoom. And although FOSDEM itself is free, plane tickets and hotel bills arenāt. For me, a full day Ada DevRoom will help to justify the cost of attending, as I rarely find more than one or two talks in other DevRooms interesting enough
2 Likes
Thank you for your reply. I agree that the cost of going to FOSDEM, specially for people who live abroad (like me), is quite high. And of course, something that adds value to the trip makes everything more palatable.
However, I have a worry hereā¦ If we try to get a full day DevRoom and we do not get it, the drawbacks to go to FOSDEM, because of/for Ada, are equally high. And if we do get a full day DevRoom but then people still do not show up (as you said, you have not been able to attend physically in the past), then there is no positive gainā¦
I see this as a risk vs. reward problem. A half-day DevRoom makes it more likely that we will get it, but at the expense of making it less attractive to the Ada community. The full-day option I see it as less likely to happen but it makes Ada stand out quite a bit more. There is also the topic of filling a full day DevRoom with talks. With a half-day DevRoom it is much easier to get it full of high quality topics. A full day one may be more difficult or may require more niche/technical topics that may have lower impact on the general public (but will most likely be loved by Ada programmers).
Best regards,
Fer
When you bear in mind that last year we werenāt even afforded a stand, I feel like it might be worth playing it safe and trying to get a half day.
If a half day is a success, we can use that as a reason in 2026 to justify a full day (e.g. āwe are requesting a full day this year since last year not all of the presentations were able to make it in for a half-dayā etc).
2 Likes
I would make sure to point to things like the forum, for which the numbers and engagement is consistently growing.
2 Likes
What makes you say that? In the past the majority of all DevRooms were full-day. And Ada has a track record of 10 full-day DevRooms and even one 2-day DevRoom, all filled with the maximum number of talks, and well attended.
Is there any indication that a half-day proposal has a higher chance of being accepted?
And as you say, a half-day DevRoom is less attractive to the Ada community. That includes potential presenters! We could get fewer proposals, because presenters might be reluctant to travel to Brussels when there are only a few other Ada talks instead of a full dayā¦
IMHO, we should at least try to get a full day. In that respect, the final sentence of the current proposal is a good compromise. It states: Special Requirements: If a full day is not possible, half a day could also be useful for us. But a full day is very much preferred.
Dirk
1 Like
I think AJ said it best
I wrote that Special Requirement these past two editions and I got no reply or question if it would be okay or something elseā¦ I am unsure if they are even reading that special requirementā¦ I do not know how the choices are done at the FOSDEM board level, but I have the feeling that they all read the description an the amount of time requested, with not many noticing the note. Then they make a decision while building the final scheduleā¦
I do not want the community to think that I am not in favour of a full-day DevRoom. But I want to increase the likelyhood that we will have something 
It could be worth risking a full day and if not, having a stand and talks related to Ada in other Rooms though
Best,
Fer
We werenāt afforded one in the last two years though (and we werenāt even given a stand for '24). It wasnāt just Ada though, Haskell was also denied one so itās possible that FOSDEM24 is having to make hard choices from an overabundance of requests or theyāre just trying to move away from language-centric rooms.
I donāt know for certain if requesting a half day would increase our chances (itās quite possible we would be rejected for half-day too), but I do know that requesting a full day has not worked in the past 2 years.
We cannot directly influence whether they give us a dev room (though we could ā theoretically ā indirectly influence it by requesting a half day).
What we can directly influence is our showing (I even flew out from the USA for FOSDEM24 despite everything, just to show my support for the community and show them we still exist!).
Weāve tried this in the last two years but to no avail. Perhaps we should be putting it in a different section of the application and still apply for a full day. Worst case scenario is that we simply donāt get a dev room and we put our focus into presentations (which is what this topic originally prompted).
Iām not against trying for a full day, I just believe that half a day may increase the chances of getting our foot back in the door (at which point we can fight for a full day next year armed with evidence that the half day was successful).
It may be worth talking through all of this at the ada meeting this weekend, perhaps?
This is a great point.