All our hack are belong to us.

Active projects and challenges as of 21.12.2024 15:09.

Hide full text Print Download CSV Data Package

#parldata

AI in Parliament count()

Question from the audience: how many proposals and sessions currently deal with AI topics? Off the cuff, we were told 100 proposals and 30 dossiers. Stimmt es? Let's check the open data!


~ PITCH ~

🅰️ℹ️ Imagine a future where AI seamlessly enhances the functioning of parliamentary services while keeping humans at the forefront. We invite you to explore innovative solutions that leverage AI to support parliamentary research and streamline processes without compromising transparency and security. Focus on developing tools that can contribute to the pilot chatbot aimed at delivering comprehensive thematic documentation from resources like Parlament.ch, ParlData, and Fedlex.

Additionally, consider exploring semi-automated indexing of press articles and parliamentary interventions, ensuring that these advancements empower legislators and citizens alike. Your ideas can redefine how we view the integration of AI in governance, fostering a more effective, inclusive, and responsive parliamentary system. Work collaboratively to unleash creativity, respect diverse perspectives, and contribute meaningful solutions in the spirit of fairness and democracy!

Curia statistics

Thema Geschäfte Total Erledigt Hängig Überwiesen Amtliches Bulletin
Datenschutz 966 810 49 45 742
Software 477 413 19 17 312
Algorithmen 65 47 15 3 27
Cloud 59 50 7 2 61
KI-Systemen 21 12 6 3 4

Screenshot

~ README ~

The presentation will give an insight into the current and future projects focused on the integration of AI in parliamentary services. It will highlight AI's role as a co-pilot, emphasizing that humans remain central to all processes. The presentation will detail the upcoming launch of a pilot chatbot which will support parliamentary research by providing thematic documentation using resources like Parlament.ch, ParlData, and Fedlex. The Proof of Concept (PoC) will be tested, along with a pilot project for the semi-automated indexing of press articles and parliamentary interventions. The presentation will conclude by emphasizing the transformative potential of AI in enhancing parliamentary processes while maintaining a human-centered, transparent, and secure approach to its development and application.

#dinacon #hacknight #ideasarefree

DINAcon 2024

Quick guide on suggesting a challenge from a talk


~ PITCH ~

While attending DINAcon, you may want to highlight a project or idea that was mentioned by a speaker. This is easy to do:

  1. Click on the 🟩 green cube on the event page, or just use this link to add something to the hexagrid.
  2. If it is a talk, just copy and paste the Pretalx link. We've helpfully embedded the Program for you 👇 below. Or if there is an open source repository, use that link — otherwise, skip and add a title and summary.
  3. Fill it out with any 🧠 thoughts to reflect on what you heard, open questions, and additional HACKnight challenges.
  4. Share! Tell your friend in the next chair at the conference to Join your suggestion, or use the Invite function to email a link.

P.S. if you want to see closer integration of our platforms, leave a comment here or vote up for our GitHub Issue

DINAcon 2024 Program

~ README ~

The DINAcon HACKnight is a way to learn about and contribute to Open Source projects in a hands-on way. This event is open to all, and caters to all levels of technical skill. Our focus will be on all projects nominated and showcased at DINAcon.ch, the Swiss digital sustainability conference - other open source/data/hardware projects are welcome. Committers and experts are on hand to guide you through to your first bug report or code commit!

All Challenges posted at hacknight.dinacon.ch are announced at the start. Visit the site to browse activities from previous years.

Download the official wallpaper, checkout the Issues list, and get HACKing! See this blog post for a recap of our first event.

Terms

The most important rule of the HACKnight event is:

Be Excellent to each other.

For further guidelines of conduct, please refer to confcodeofconduct. As this is an open source event, we will encourage all teams to publish their work under open licenses in open repositories, such as but not restricted to GitHub. The organizers, sponsors, and event staff shall not claim or request any endorsement or special rights and privileges to any work you do at the event. All project documentation created or shared during the event for projects published as above will be republished and promoted under a Creative Commons 4.0 Attribution license.

#openlegaltemplate

Model agreements for open source software

A repository of sample contracts that companies can use in business transactions for the use and development of open source software.


~ PITCH ~

The authors present a repository of sample contracts that companies can use in business transactions for the use and development of open source software. Our talk presents the idea of a repository of openly licensed contracts and other legal templates, created primarily for use among and by the open source community. This discussion has circulated for some years within CH Open, reignited by recent publications of nearby communities.

Our talk will go beyond just the idea, but present a prototype of this repository, built on open data best practices, and an open source platform designed for legal work. In our demo, we will guide you through a questionnaire, quickly producing a contract in standard formats. Let's discuss:

  • What about liability for the contracts?
  • Is it even possible to work with templates, which situations would that be? What do the existing templates cover?
  • Why do we think this will improve the resilience and sustainability of development?
  • How do we plan to enable collaboration among the legal/technical community?
  • Where will A.I. play into our designs?

Please add further questions and your comments into our GitHub discussion

Proxeus Demo

Demo server for contract generator prototype: sepolia.proxeus.org

Screenshots

Just for fun

An ancient legal piecemeal:

https://youtu.be/CUleKnUUaGI

~ README ~

Model agreements for open source software

(Musterverträge für Open-Source-Software)

A repository of sample contracts that companies can use in business transactions for the use and development of open source software.

Presented at DINAcon 2024

About

This is a prototype of a repository of openly licensed contracts and other legal templates, created primarily for use among and by the open source community in Switzerland. This discussion has circulated for some years within CH Open, reignited by recent publications of nearby communities. Our project aims to build on open legal data best practices, and promote open source platforms designed for legal work.

Formatted in Markdown using legal-markdown

The proxeus folder contains exports for a demo Proxeus workflow

References

In Switzerland

Globally

Licenses & Tools

Technical Projects

Public Sector

License

CC0 1.0 Universal

Retro PGF

Retroactive Public Goods Funding


~ PITCH ~

This challenge is based on an interview by Philip Heltweg with David Gasquez about funding models and open data in the Web3 ecosystem.

Retroactive Public Goods Funding (Retro Funding) is based on the idea that it's easier to agree on what was useful in the past than what might be useful in the future -- Optimism community

{ hacknight challenges }

Read the full conversation or our summary below, and check out the recommended learning resources:

✅ Get the book: Designing Data-Intensive Applications by Martin Kleppmann

✅ Try out Kaggle and AIcrowd, platforms for data science competitions

Find a person to interview in the context of open source/data ecosystems in your near vicinity, or online. Publish this interview as a recording, blog or social media post. Share the link here and on social media.

Design and build a platform or application that leverages open data portals to create value and generate revenue. Inspired by Drip Lists, GitHub Sponsors, Gitcoin Grants and Filecoin Open Grants, we challenge you to create a innovative solution that showcases the potential of open data ecosystems.

~ README ~

The article features an interview with David Gasquez, who works independently on open data projects within the Web3 community. The conversation focuses on "Retroactive Public Good Funding" (Retro PDF), a novel funding approach in Web3. David's Background: David Gasquez works self-employed on open data projects, focusing on the Web3 ecosystem, aiming to create open pipelines with open data.

Retroactive Public Good Funding (Retro PDF) is defined as a grant received after completing a project. It involves building a fund for existing projects, community voting, and retroactive distribution of funds to selected projects. Implementation examples include voting via Google spreadsheets or the Filecoin data portal.

The interview speculates on whether this funding model could work outside Web3, potentially for open data portals or even public goods like parks, with participatory budgeting as a similar concept. There is mention of possible experimentation by the Linux Foundation, which needs to be confirmed. David mentions the importance of learning from previous rounds, as seen in Gitcoin's approach, to improve aspects like project surfacing, impact measurement, and more.

(Summarized will help from Llama 3.1)

Source

Souveräne Mini-Cloud

Cloud Service Souveränität mit Laptop und RaspberryPi


~ PITCH ~

Cloud Services gibt es aktuell nur von den Grossen (BigTech). Warum eigentlich? Ist es wirklich so schwierig eine Cloud zu bauen und Services bereitzustellen? Wir meinen das war es einmal. Technologien wie Ansible/Kubernetes und erschwingliche Hardware haben die Spielregeln geändert. Mit diesem Hack wollen wir beweisen, dass man Heute sehr wohl eine souveräne (zumindest kleine) Cloud bauen kann.

Der Organisator (Janik) bringt: 5 Raspberry PIs

Voraussetzungen:

  • Ein Laptop vozugsweise mit Linux und alternativ Windows/WSL

Vorgehen:

  • Installation der 5 Raspberries
  • Einrichten Ansible und Kubectl auf dem Laptop- Konfiguration Ansible K3S-Rolle und Deployment Kubernetes-Cluster
  • Definition der Applikationen für das Deployment
  • Erstellen Deployment-Manifest-Dateien
  • Bereitstellung der Cloud Services im Kubernetes-Cluster
  • Erstellen Demo für Lifecycle eines Cloud Services

Noch Offen:

  • Wie können SSL-Zertifikate für die Services im lokalen Netzwerk generiert werden?
  • Welche Cloud Services sollen bereitgestellt werden?
  • Ist genug Kubernetes KnowHow vorhanden?

See Also:

~ README ~

raspi-and-friends

HACKnight project at DINAcon 2024

The group in the AULA Hacknight is working on the Souveräne Mini-Cloud with a bunch of Raspberry Pi.

Wifi

SSID: AULAthenet Password: inter-nett

Setup Raspi

Tom

Hostname: raspberry-tom IP: 192.168.102.136

Username: tom Password: dinacon

Mischa

Hostname: raspberry-mischa IP: 192.168.102.151

Username: mischa Password: dinacon

Ozzy

Hostname: ozzberry IP: 192.168.102.152

Username: ozzy Password: HackNight-2024

Foss

Hostname: foss (.local)

Username: janikvonrotz Password: janikvonrotz

SSH: Enabled SSH-Login: Password

ssh: ssh janikvonrotz@

Setup K3S

Link: https://docs.k3s.io/quick-start

In the terminal of your laptop:

sudo curl -sfL https://get.k3s.io | K3S_TOKEN=dinacon sh -

SSH to the rasperry pi.

Enable cgroup v2: https://maxdon.tech/posts/k3s-raspberry-pi/

echo ' cgroup_memory=1 cgroup_enable=memory' | sudo tee -a /boot/firmware/cmdline.txt
sudo reboot

Setup worker:

sudo curl -sfL https://get.k3s.io | K3S_URL=https://HOST_IP_ADDRESS:6443 K3S_TOKEN=dinacon sh -

Copy the kubconfig on your laptop.

sudo chmod 644 /etc/rancher/k3s/k3s.yaml
cp /etc/rancher/k3s/k3s.yaml ~/.kube/config

Check if node is ready on your laptop.

k3s kubectl get nodes

Challenges

#docathon

Docathon

Give documentation your love and attention!


~ PITCH ~

Technical writers are the unsung heroes of digital sustainability. Let's check the quality of READMEs, manuals and in-line documentation, set up workflows for testing documentation coverage in open source projects, and promote the idea of a "Documentathon": a hackathon for documentation writing - or Docathon, for short.

This challenge came out of a recent CH Open workshop that is described in Oleg's blog. The generative image above of a 'lonely documentation king in the data center' is also from there. It is inspired by Docathons in the Python data science community.

🧊 Quickstart

In just 15 minutes, you could contribute a new example or a screencast - that's the 🟢 green challenge below. If you are willing to dig into the source code, you could try to improve 🔵 docstrings and code comments, as well as giving the authors and maintainers valuable feedback. You could even work on automated tooling to help with creating more ⚫ effective documentation.

{ hacknight challenges }

To begin with, find a project that could use some 💙 documentation love. Look up something you heard about at DINAcon - check the comments below for ideas, go through the other challenges here, look up open source projects you use & follow, community tags like #help-wanted, etc.

Go through the documentation yourself. See if there is already a discussion open about improving the documentation. Leave a comment, or start a new issue with improvement suggestions. Practice sharing great feedback with a constructive message about how the project speaks to you.

Work on some open source documentation, start a Pull Request with your contributions. Leave a comment here with the link. See PyTorch blog and Encouraging users at the original Docathon page for more tips on getting started.

See how the interrogate project helps to identify gaps in the documentation, codecrumbs beautifully annotates it, while the hermes tool collects structured feedback. Try this, or another automated method - a CI tool like Deepsource, an open source LLM, or one of the many other awesome docs tools. Ask if your docs are "beautifully integrated" (Breathe) into your source. Think about ways to analyze, highlight and recognize the documentation work in open source.

Share your experiences with us at DINAcon - please leave a comment if you'd like to organise a Docathon in the future at your workplace, and we'll be happy to support you in any way.

Screenshot of Deepsource

Screenshot of a Deepsource analysis of Dribdat

Dashboard of Docathon projects

Drag[en]gine

Open source indie game engine


~ README ~

Drag[en]gine Game Engine

dragengine_logo

Example projects can be found on Development Wiki or in this repository: - Example Projects Repository: https://github.com/LordOfDragons/deexamples

Nightly

Game developer information

The Drag[en]gine is based on a non-blackbox principle. In particular this means for developing games using the Drag[en]gine you do not need to compile the game engine nor do you have to link against it. If you just want to develop games using Drag[en]gine installing the pre-compiled release files is all you need.

The source code repository is interesting only for people interested in the inner workings and those wanting to create engine modules or editor modules.

License Information

The Drag[en]gine Game Engine is licensed under MIT for clarity. Optionally the L-GPL can be used. This affects only game engine modules, launchers and IGDE editors. Game projects do not link against the game engine and thus are not affected by the license. They can use any license they see fit.

In a nutshell this means: If you create a game, no matter if commercial or not , you do not have to worry about licensing at all.

GIT Branches

The master branch is the current development branch with the most recent work. It is potentially unstable and should be only used by developers working on bleeding edge software build against the upcoming release.

The stable branch is the stable branch pointing to the latest release. This branch is for public use by distribution builders and developers working on software to work with the latest stable release.

All other branches are internal feature development and platform support branches not intended for public use. If you want to work on these branches please get in contact with the development team first.

Building Linux

To customize build either copy "custom.py.dist" to "custom.py" and edit or add the parameters in custom.py.dist as command line arguments. For example:

scons with_debug=yes with_system_openal=no

To see the list of supported build targets run

scons -h

To compile and install run

scons

By default the build tries to find as many dependency libraries as possiible in the host system. Not found libraries are compiled from in-tree sourced under "extern". You can force building certain libraries either in-tree or from system by setting the respective withsystem* parameter to 'yes' (to force using the library from host system and fail if not found) or 'no' (to force compile from in-tree version) or 'auto' (use from system if found otherwise fall back to in-tree version).

The default installation prefix is "/usr". To change the installation prefix use "custom.py" or use

scons prefix=/usr/local

To build distribution archives run

scons archive

The archives will be located under "archive/build"

To build installers run

scons installer

The installers will be located under "installer/build"

In general all compiling takes place inside "build" directories underneath the respective SConscript files. To make a full clean you can delete all directories named "build" safely

Every target can be cleared by appending "-c" to the command line. For example to clear build files run

scons -c

To clear only archive files run

scons -c archive

Clearing only removes files not directories. This is the way SCons behaves.

When running SCons as user it is recommended to use "prefix" parameter to define the installation directory otherwise installing files fails with permission denied errors. It is not recommended to run SCons as root unless you intent to install files directly into the system.

Package maintainers requiring sandboxing to build packages can use the SCons sandbox support to place installed files in a different location than the "prefix" parameter defines like this:

scons --install-sandbox=path/to/sandbox

See the debian branch for an example of how this is used.

Build Dependencies

  • Git Lfs
  • SCons 2.5.1+
  • Python 3.5+
  • GCC 8+
  • CMake (for in-tree building of external software)
  • libtool (for in-tree building of external software)
  • gperf (for in-tree building of external software)

Building Windows

Install MinGW 64-Bit cross-compiler for your system or use a docker image with it.

To build use the same commands as under the "Building Linux" section but append "tools=mingw64" to the command line like this:

scons tools=mingw64 archive

Always use a build target otherwise scons tries to install into your linux system using windows path names.

The "build_windows.sh" script is a little helper.

To build the windows installers you need a docker image supporting the InnoSetup compiler. You can build the docker image yourself using the "windowsSetupCompiler.dockerfile" docker file:

docker build --file=windowsSetupCompiler.dockerfile --tag=compile-windows-installer .

To build the installer first build the archives using

scons tools=mingw64 archive

Then change into the "installer/windows" directory and run

./create_installer.sh

The installer is now located inside "installer/windows/build"

It should be possible to build the sources also on a windows machine using Windows MinGW-64. VisualStudio compile is not supported officially.

Run in "Local Directory Mode"

The Drag[en]gine Game Engine, Launchers and the IGDE can be run in the "Local Directory Mode".

The Live Launcher is an example of a launcher using this mode to run games from any directory without installing.

It is possible to run also the IGDE and all launchers in this mode using a source-able file setting up the required Environment Variables with appropriate values. You have to first prepare a directory with the game engine files. For this download the linux binary distribution file and unpack it into a directory of your choice. Then copy the preparelocaldir_mode.source file into the directory. Change into the directory then source this file like this:

source prepare_local_dir_mode.source

or like this (for consoles not supporting the 'source' builtin command):

. prepare_local_dir_mode.source

The terminal is now ready to run delauncher-gui, delauncher-console and deigde binaries from the local directory. All user configuration is also store under the user directory. You can also change the source-able file to choose different directory if needed.

Hackfinder

A new tool to look up and recommend hackathons in Switzerland


~ PITCH ~

Bringing high-skill refugees and jobs together through diverse hackathon teams

The Hackfinder is a new tool under development at the BFH (DSL / Social Work) as part of a new Innosuisse project. The tool was recently released, and we will share more info (and the source code) by the next DINAcon, as the project gets going in 2025. Here is an alpha preview, with comments and participants welcome:

In this project high-skilled refugees are referred to hackathons, multi-day participatory events where participants work in teams on real-world problems. These events are an opportunity to showcase one’s skills and expand professional networks, hopefully leading to better employment. In this project we test whether, as research would suggest, hackathon participation can really improve employment and income.

See also a recent Hack:Org:X presentation - and ask Oleg about "Hack the Hackathon", a research conference running in parallel to DINAcon.

IntelliProcureAI

How can large language models be used for sustainable procurement?


~ PITCH ~

Screenshot

🅰️ℹ️ In a world increasingly focused on sustainability, there exists a pressing need to revolutionize procurement processes through innovative technology. Leveraging insights from keynotes and panel discussions at DINAcon, we invite participants to explore the integration of Large Language Models (LLMs) in sustainable procurement.

Your challenge: mock up a conversation with a chatbot that uses the information visible in the demo. This could be a valuable tool for streamlining the procurement process and ensuring that sustainability requirements are met. By providing quick access to information and resources, it can help reduce the workload on procurement departments and suppliers, while also promoting best practices for sustainable procurement.

~ README ~

In our talk, we will present two current research projects at the Bern University of Applied Sciences. In one SNSF project, we are investigating how Large Language Models (LLMs) can be used to assess public procurement data for sustainability.
At the same time, we are working on an InnoSuisse project to explore how LLMs can better support both the contracting authority and the contractor sides in the procurement process.