LibreOffice Conference 2026

Welcome and Registration

Housekeeping Information

Welcome from TDF

Welcome from PNlug

State of the project. Statistics about development, downloads and donations.

This talk will give a status update on the OASIS & ISO ODF Standard and one of its reference implementations, the ODF Toolkit from The Document Foundation (TDF) https://github.com/tdf/odftoolkit

Welcome from Authorities

Stats about Downloads/Donations

LibreOffice is developed, tested, documented, translated and marketed by countless contributors from around the world and used by millions of users around the globe. This community of creators and users is backed by a non-profit foundation which not only publishes the software, but also acts as a trustee: It gives the community its legal personality, is the guardian of trademarks and domains, and also manages donations as well as assets. The community makes the software what it is today: the leading free and open source office suite, available free of charge to personal, business and governmental users. True to the motto “by the community, for the community” the foundation was made possible by numerous donors from all over the world who raised the necessary capital in just eight days as part of a public fundraising campaign. In this talk Florian not only explains the legal setup of The Document Foundation and how its statutes enshrine values and ideals like openness, transparency and meritocracy. He also explains how it was designed to combine the conservative elements of a foundation with the modern approach of an open source project and shares a “lessons learned“ from the past 14 years to explain the challenges a growing FLOSS project can face.

Many AI tools focus on generating an answer or a text. For administration, ministries, authorities and public IT, this is insufficient. Many AI systems focus on generating content.

However, in public administrations, enterprises and regulated environments, the central questions are often different:

Who worked on a document?
Which state was reviewed?
Was the document changed afterwards?
Can a verified state be transferred, audited or reconstructed?

KWI LIAN introduces a local governance and evidence infrastructure around LibreOffice Writer and Calc.

The Kyber extension integrates directly into LibreOffice and provides document state verification, LIAN stamps, evidence creation and workflow visibility.

The approach follows a simple principle:

The AI supports.
The human decides.
The infrastructure documents.

The talk presents how LibreOffice can become a local workspace for governable AI-supported document workflows without cloud dependency.

Topics include:

• LibreOffice Writer and Calc extension integration
• local AI support architecture
• document state verification
• Human Commit Boundary
• server-side evidence records
• hashes, snapshots and commit information
• multi-user handover scenarios
• digital sovereignty for public sector environments

KWI LIAN does not replace human decisions and does not act as an autonomous decision system. Instead, it creates a transparent layer connecting people, documents, verification states and evidence records.

The presentation demonstrates a practical approach for using LibreOffice as a foundation for sovereign, auditable and human-controlled AI workflows.

The LibreOffice project has been a champion of advocating open formats and rejecting proprietary ones for documents, in addition to its commitment to free software ideals. However, like in many communities, most of the communication channels of the LibreOffice project are hosted on Telegram, which is not based on an open protocol.

In this talk, I explore the possibility of switching to XMPP, a decentralised messaging system with considerable adoption in Free Software communities. XMPP is an open standard based on XML, which incidentally results in significant synergy with LibreOffice's own goals!

I will cover how the project can switch to XMPP, whether it is something TDF would like to self-host or use existing community run services, and present some practical advice.

LibreOffice was announced in 2010 as a desktop productivity software focused on interoperability thanks to the native support of the Open Document Format. Today, is the only technology platform for desktop productivity supporting applications for the desktop, the cloud and mobile, independent from large commercial companies, and providing ownership of contents to users, thanks to the native support of the Open Document Format.

This talk is about AI extension to create drawings in LibreOffice Draw

The LibreOffice project is (or was) active on many social media channels: Mastodon, Bluesky, X/Twitter, Reddit, LinkedIn and more. But what have we changed in recent years? What controversies have there been? And where should we be active in future?

Come hear about many of the large number of Open Source Office suit projects that have been developed over time in this space - from the early history of KDE and GNOME suites, through OpenOffice and its many derivatives, as well as newer and peripheral code-bases, and office adjacent suites running on FLOSS platforms.

Get some insight into the trajectory, ODF interoperability and feature-set of each of these, and how they compare to a recent LibreOffice.

LibreThinker is a free AI assistant extension for LibreOffice Writer, with over 19,000 downloads in 9 months. It started as a Python project wrapping GPT's API and grew into a multi-backend tool that supports free hosted models, connecting your own AI provider, and running models locally on your own machine. This talk covers how it was built, the engineering problems that came with running a free public endpoint, and what users actually ask for.

Many Free Software projects, non-profit organizations, and digital rights communities have opened accounts on the Fediverse—but far fewer have succeeded in making it a real communication channel.

Simply copying posts from X, Facebook, or LinkedIn rarely works. The Fediverse follows different social dynamics, rewards authentic interaction, and offers opportunities that centralized platforms cannot provide.

Drawing on years of experience running and moderating some of Italy's largest Fediverse communities, this talk presents practical strategies that have helped thousands of users discover, join, and actively participate in decentralized social networks.

Rather than advocating an abrupt departure from Big Tech platforms, the session explores realistic migration strategies that allow organizations to maintain their existing audience while steadily growing independent communities.

Attendees will learn how to increase engagement, encourage voluntary migration, organize onboarding campaigns, leverage community directories and curated recommendations, and measure success using meaningful community metrics instead of algorithm-driven follower counts.

Whether you represent a Free Software project, an open source community, a public institution, or a non-profit organization, you'll leave with practical techniques you can apply immediately.

Writing scientific papers is a requirement for lecturers each semester. Despite various limitations, including the equipment used, time constraints, and other factors.One of the challenges faced in writing scientific articles is creating formulas. Therefore, the author sought an alternative solution to overcome this problem by using LibreOffice Writer, adding the TexMaths extension and installing TexLive to ensure the formula creation requirements are met according to the template used in writing.
This knowledge is considered important to share with beginners who want to write scientific papers or other needs.
The benefit is that you can write easily without having to master complicated programming languages ​​to get results with quality and elegant layout.

LibreOffice OfficeBean enables Java applications to embed LibreOffice within their own interface. At the moment, this works through native bridge code on Linux/Unix and Windows, but macOS support is still missing.

I started looking into this through Bug 170415 by trying to add a small Cocoa/JNI bridge for macOS. The first result was partly encouraging: the native bridge connected, and the window could be created without crashing. But LibreOffice itself only showed a grey blank area.

In this lightning talk, I want to share what I tried, where it got stuck, and what I learned from maintainer feedback. The main challenge seems to be that macOS does not handle this kind of cross-process native window embedding in the same way as Linux or Windows. I will also briefly mention the direction I am exploring next: using IOSurface-based delegated rendering instead of trying to make LibreOffice draw directly into a Java-owned view.

LibreOffice Writer is a powerful tool for document editing and has many degrees of freedom for creating a document. Nevertheless, when it comes to documentation production and maintenance, some rules must apply to narrow the freedom and allow use of automation in text formatting. We explain the use of the Sanity Check extension written by B. Antonio Fernandez for creating consistent homogeneous text document format.

A few years ago, Janayugom — a newspaper based in Kerala — switched to GNU/Linux. The software they were using for printing was outdated and only supported ASCII encoding, but they needed something with a better support for complex rendering layout for printing in the Malayalam language.

When they looked for better alternatives, Adobe InDesign was the most recommended solution by the industry. However, it required paying a recurring subscription fee to use it.

With the help of the free software community, the newspaper switched to its own GNU/Linux distro instead, which was based off of Kubuntu. The stack also contained LibreOffice which the newspaper uses for its documents on a daily basis.

References:

[1] https://alphafork.com/janayugom/

[2] https://hasgeek.com/fosscell/fossmeet23/sub/desktop-publishing-and-foss-a-success-story-EQN2NSfmh9Yi15XWV5n2vu

[3] https://www.opensourceforu.com/2020/02/the-print-media-can-slash-costs-if-it-switches-to-open-source/

[4] https://news.ycombinator.com/item?id=21504721

[5] https://blog.smc.org.in/smc-monthly-report-october-2019/

Histogram Chart support work is been worked again and we now have much better and complete support for the Histogram Chart.

https://gerrit.libreoffice.org/c/core/+/203895

This was earlier part of my GSoC 2024 work but got reverted as lack of support of new namespcae in LibreOffice and New namespace by OOXML.

Talk about the newly introduced property, slide sections in Collabora Office/Online, and their specifications in the ODF format.

PREEvision© [0] is a future-proof, holistic environment by Vector[1]. It challenges the needs of modern automotive system engineering and supports the entire E/E process. Thereby, it covers all states of the gadget planning like textual description, component design down to hardware components and its software and even signal routing and wiring harness. In each level of the process, various texts can be created and graphical diagrams may need to be persisted.
By using LibreOffice within PREEvision© as a document engine, an easy interface was created to process this data:
• For editing formatted texts, LibreOffice is used as a text editor
• LibreOffice’s various import and export filters offer the possibility to convert each document into the required data format
• Together with the Open Document Format, LibreOffice offers a powerful mechanism to create various presentations of the data

[0] https://www.vector.com/int/en/products/products-a-z/software/preevision/
[1] https://www.vector.com/int/en/

This talk reports on organizing LibreOffice Asia Conference 2025 (Tokyo, December 13-14, 2025), co-hosted with GNOME Asia Summit 2025. LibreOffice's local team normally decides everything, while GNOME Asia's local team only handles on-the-ground tasks — so our team ended up doing both. On the preparation side: venue, this dual structure, the call for papers, visas, meals, printed materials, and a post-conference tour, along with the thinking behind these decisions. On the outcomes side: discussion at the event was genuinely lively, and it did strengthen connections between communities — though it's hard to measure exactly how much. This is a first-hand, practical account of what it took to run this event.

Traditional usage of office documents stores them as binary archives hindering effective version control. This session showcases our workflow using Flat ODF and Git to enable granular, text-based versioning and seamless merging for office files. Unlike standard LibreOffice collaboration where internal change tracking can be altered, our approach leverages Git’s immutable history to ensure non-repudiable attribution for every modification.

The new LibreOffice website was implemented using Hugo static site generator. A large part of the development effort was focused on multilingual features.

Some time ago, I worked with a dataset containing textual data. At that time, I performed entropy calculations using Microsoft Excel. In this study, however, I aim to explore the use of LibreOffice Calc for data processing. Using the same dataset, I conducted sentiment labeling, topic relevance analysis, and entropy calculations for each column. The entropy calculation was performed to determine the weight value of each word.
The first step was to identify the number of unique words appearing in the dataset. After collecting all unique words, the next step was to calculate the frequency of each word. Once the frequency values had been obtained, the subsequent step was to calculate the probability of occurrence for each word. An example of this process is presented below:
Unique Word Frequency Probability
Train 2 2/4 = 0.50
Fire 1 1/4 = 0.25
Bekasi 1 1/4 = 0.25

In LibreOffice Calc, the formula used to calculate the entropy contribution of a probability value is:
=-(0.5*LN(0.5)/LN(2))
The result of this calculation is 0.5. This serves as an example of entropy calculation based on a probability value. In information theory, entropy is used to measure the amount of uncertainty or information contained in a dataset. The same calculation can be applied to other probability values, and the resulting entropy contributions can then be summed to obtain the total entropy of the dataset.
In addition, I calculated the probabilities of sentiment categories that had previously been grouped into three classes: positive, negative, and neutral. The first step was to determine the number of instances belonging to each sentiment category. To perform this calculation in LibreOffice Calc, I used the following formulas:
=COUNTIF(A:A,"positive")
=COUNTIF(A:A,"negative")
=COUNTIF(A:A,"neutral")
These formulas count the occurrences of each sentiment label within the dataset. The resulting counts are then used to calculate the probability of each sentiment category by dividing the frequency of each category by the total number of records. The results are presented in the following table:

A B
Sentiment Frequency
Positive 140
Negative 35
Neutral 16
Total 191

After obtaining the counts for the positive, negative, and neutral sentiment categories, the next step was to calculate the probability of each sentiment. Before calculating these probabilities, the total number of sentiment instances was determined by summing all sentiment categories. Subsequently, the probability of each sentiment was calculated by dividing the frequency of a sentiment category by the total number of records.
For example, the formula used in LibreOffice Calc was:
=B2/$B$5
Using this formula, the probability of the positive sentiment category was calculated as 0.73298. The same procedure was applied to the negative and neutral sentiment categories. The complete results are presented in the following table:
A B C
Sentiment Category Frequency Probability
Positive 140 0.73298
Negative 35 0.18325
Neutral 16 0.08377
Total 191

LibreOffice Calc proved to be a valuable alternative for this study due to its ease of use and user-friendly interface. The software was straightforward to understand and provided a comfortable working environment for data processing and analysis. Based on this experience, I would recommend the use of LibreOffice Calc to other students and researchers who require an accessible and effective tool for data analysis.

Enterprises are not supporting the LibreOffice project as much as individual users, and this can represent a threat for the sustainability of the project. Although we can further increase the number and amount of individual donations, we have to educate enterprises about giving back to FOSS, to ensure the long term sustainability of the LibreOffice project.

Placeholder for the ESC meeting

Collabora Online and Collabora Office are using LibreOffice technology, which has enabled us to contribute lots of work and improvements to LibreOffice over the past year. Come and hear about the many feature improvements from the browser, ODF & interop, as well as the core code - and about the team of volunteers and staff that made them happen, and how you can get involved.

Lots of knitters use spreadsheets in their knitting projects for everything from tracking yarn usage to designing knitting charts. This talk will be a whirlwind introduction to using Python macros to augment your toolbox of functionality in Calc, with a specific example use-case of improving the workflow for knitters. Even non-knitters may find it interesting!

Report on the recent progress of ODF at the technical committee, the current road to ODF 1.5

The traditional LibreOffice interface, while familiar, often presents a rigid layout that struggles to adapt to modern, diverse user workflows. For beginners transitioning from proprietary ecosystems like Microsoft Office, a jarring UI shift can hinder adoption. This presentation introduces a conceptual framework for a customizable, tileable user interface system within LibreOffice, designed to give users ultimate control over their workspace through intuitive drag and drop mechanics.
Rather than proposing a heavy, modern web-app overhaul that compromises system resources, this design philosophy focuses strictly on leveraging LibreOffice's existing native libraries. By utilizing the suite’s established codebase, we can achieve high performance, lightweight UI responsiveness without the memory bloat typical of modern framework shifts.
This 30 minute session will explore the UX layout logic of a modular button and panel library, the technical advantages of keeping UI rendering low-level and native, and how a highly adaptable interface can lower the barrier to entry for new users. Whether you are a developer interested in lightweight UI architecture or a UX enthusiast focused on software accessibility, this talk offers a fresh perspective on evolving the open-source desktop experience.

This talk will give an overview of accessibility related changes in LibreOffice since last year, from both, a
user and a developer perspective.

For over 20 years LibreOffice’s Mac-native interface was created by drawing pictures of buttons and checkboxes with outdated functions. While this has served us well, it still doesn’t look quite native, doesn’t provide the same standard behavior as first-party applications, and these outdated functions are going away. We can do better! From real native Mac controls, to third-party helpers, to Apple’s new standard writing and creation tools, this talk explores new possibilities to help LibreOffice look and work better on Apple platforms.

Introduction to the LibreOps Extension
* Purpose and Objective of the Extension
* Development Process Using Python
* Bookmarks with RDF Design
* Scope of This Extension
* Solutions to Problems Encountered During Development
* Related Bugs

We will present changes in 2026 that benefit user experience, improve the design, and make LibreOffice shine. The talk addresses ordinary users as well as people who are interested to volunteer with easy hacks.

LibreOffice contains over 80,000 strings, or over 270,000 words in English, of user interface text. In this presentation I will look at what makes a good UI string, drawing from human-computer interaction and my experience in translating LibreOffice. I will also discuss ways in which designers, developers, and translators could work together to improve the quality of UI text.

In this talk, I discuss LibreOffice user interface architecture.

The classic ‘blank page’ in word processors and standard user interfaces are not neutral: for users with low vision, photophobia or specific learning difficulties, poor management of contrast and typography represents a genuine digital accessibility barrier.

This talk explores the ‘behind the scenes’ of visual accessibility, using LibreOffice as a reference suite and examining the official development specifications for colour and font management. We will analyse how the software interacts with the operating system to automatically detect and inherit ‘High Contrast’ settings, dynamically adapting the interface without requiring complex manual configuration by the user.

We will delve into the rendering logic designed for maximum readability: from the automatic removal of fill colours and text in Impress and Draw when used in high-contrast mode, to the distinct highlighting of the current row and column in Calc, right through to the forced overwriting of document background colours. We will also see how the architecture enables scalable resizing of interface fonts and the disabling of animations to reduce cognitive load.

The Bookshelf project was created to allow quick access of LibreOffice guides in all relevant media: PDF, ODF, web pages and any other file format. We explain the process of creating an entry in the bookshelf for your book.

The main issue for document interoperability is a trivial one, and are fonts which are proprietary on Windows/macOS and free on Linux. LibreOffice is capable of handling on the fly font replacement to solve the issue, but the feature has to be configured and managed to work properly.

Reproducible builds are a way to increase trust in the distribution channels. If you can verify that you get the same results if you compile with the same set of tools, you can also be sure that the source hasn't been modified / is what it claims to be.

This talks gives a short overview where we stand in terms with reproducible builds, challenges to complete the process and how other interested parties can participate.

After sanitation of the chapter formatting of the LibreOffice Guides, we explain how to convert a text document into a web page, with enhancements of web navigation and more features.

"Strength in numbers." "Sticks in a bundle are unbreakable." "United we stand, divided we fall." There exist numerous cliches about the importance of working together, and no one knows the value of collaboration more than FOSS contributors. However, when focusing on the goals of one's own community, we may overlook the importance of cross-community collaboration.

In this talk I will take a deeper look at the collaborations which were part of KDE Eco's "Opt Green" project, in particular the "End Of 10" campaign. The focus will be on how building alliances with groups outside of KDE opened doors to new audiences and new opportunities. There are two main types of collaborations that I'll explore, namely (i) those with other FOSS communities such as LibreOffice and (ii) those with adjacent communities such as Repair Cafes and environmental groups. What approaches were taken? What were the advantages? And disadvantages? What other opportunities for collaboration are there looking forward?

We will look at the whole context of distributing LO as AppImage: the official position about AppImage from ESC, the community effort, the actual process of building a tailored AppImage for the necessities of users and testers, the automatic build process for the releases, feedback from userbase and how to channel it and some opportunities and threats for future development.

The intent of the talk is to get clarity on the status quo and propose solutions how to improve the actual and future situations.

The presentation provides insights into findings from the project “Digital Accessibility in the Workspace through Participatory Evaluation” with a focus on developer perspectives, open-source communities and accessibility knowledge within the community. The project analyses show that digital accessibility is not only a technical issue, but also an organizational and communicative one. The findings offer ideas and inspiration for increasing awareness of and engagement with accessibility across the LibreOffice community. The goal is to discuss accessibility as a shared responsibility within the open-source community.

ODF is an important open standard for office documents. It helps people and organizations keep access to their documents over a long time, and it allows different applications to work with the same document formats. This is especially important when documents must remain usable across many years, systems, and organizations.

Still, the history and current state of ODF are not always easy to understand. Each ODF version has added or changed features. Each application supports different versions and features. Some tools can read or write ODF files directly, while others can work with ODF data through data-processing workflows. As a result, useful knowledge about ODF is often spread across many places.

In this session, Koji Annoura will show a way to manage this knowledge using LibreOffice itself. The main example will be ODF versions, ODF-compatible applications, related tools, and interoperability. The session will also be a case study in a more general method: collecting information, checking sources, structuring data, and keeping the result in a form that can be maintained with LibreOffice and ODF.

What started as a practical need to ensure visual consistency between LibreOffice extensions and crash reporting tools quickly uncovered a larger architectural challenge: the lack of a unified web identity. Navigating the world of web development reveals that a true design system goes far beyond cosmetic presentation; it acts as a foundational framework for accessibility, inclusivity, and user experience.

This talk explores the ongoing journey of developing a comprehensive web design system tailored for the LibreOffice ecosystem. We will discuss the evolution of our design guidelines, moving from simple aesthetic alignment to a robust framework that prioritizes accessibility for all users. Key focus areas include building comprehensive multi-language support to respect our global community and ensuring rapid load times for lightweight, high-performance web experiences.

Furthermore, we will introduce a highly opinionated web component library specifically engineered to empower contributors. By abstracting technical complexities, this tool allows non-developers and community members to easily build accessible, official-looking websites. Attendees will leave with practical insights into bridging desktop and web ecosystems, the principles of building lightweight design systems, and a clear vision of how these tools can accelerate community growth and web independence.

Digital documents are executable files interpreted by the office suite and represented according to the XML code instructions. There is a fundamental difference between how we are accustomed to considering documents (as innocent pieces of paper) and the reality of digital documents. This difference should be taught in schools to make students aware of the consequences of being locked into proprietary document formats.

LibreOffice Base remains an important but sometimes under-discussed module of the LibreOffice suite. This talk will present the current state of Base, including the embedded database backend, the Firebird transition, the migration path from HSQLDB, and the remaining dependencies on legacy Java-based components. It will also touch on key technical and community challenges, including SQL compatibility, forms and reporting workflows, and usability and reliability issues that affect everyday users. Rather than proposing a fixed roadmap, the talk will focus on the key issues facing Base and outline the main priorities and actionable next steps for strengthening Base as a sustainable part of the LibreOffice ecosystem.

As governments move toward digital sovereignty, open document formats such as ODF are becoming more important: they allow documents to remain usable without depending on a particular software vendor. But adopting ODF is only the beginning.

Twelve years of ODF adoption in Taiwan has taught us that a document can pass a formal ODF schema validator and still cause problems in real-world use — from layout-grid drift and hand-crafted headings and tables of contents to automatic-style bloat and other document anti-patterns.

This talk introduces odfvalidator-lib, an MPL-2.0 project developed under a project of Taiwan's Ministry of Digital Affairs. Built on top of TDF ODF Toolkit's specification validator, it adds format-quality checkers designed to catch problems that schema validation cannot. We will discuss what we learned from applying these checks in practice, and show how the embeddable C++/C ABI library, language bindings, structured JSON output, and batch mode can be integrated into document-processing pipelines.

Because an open format is only useful if we can trust the documents we put into it.

TBD

The way users work with right-to-left languages has changed significantly in 26.8. This talk will give a big-picture look at how the RTL authoring workflow has improved, by comparison with earlier versions. This talk will also review and explain the specific new features that contribute to this workflow improvement. Background will be provided for those not familiar with RTL languages.

Placeholder for the Google of Code pannel

This talk continues my "State of CJK issues" series. Last year I covered Gaiji (special user-defined characters). This year has two parts.

Part 1 looks closely at Japanese line-breaking rules. Certain characters cannot start or end a line. Past talks only showed what happens when you switch a setting on or off. This time, I explain why the rules work that way. I use JLREQ, a W3C document that ODF itself refers to. Then I check how closely LibreOffice follows it.

Part 2 gives a broader update on CJK bugs. I show a few bugs fixed this past year, and a few still open. All of them come from tdf#83066, the main CJK bug tracker on Bugzilla.

TDF New Logo Proposals

LibreOffice Certification: the updated program

Twenty-one years ago, on September 1, 2005, all Italian-language schools in South Tyrol switched to using Free Software on all computers used for educational purposes. In this talk, I will briefly review the history of the FUSS project (Free Upgrade in South Tyrol’s Schools), the reasons behind the choice of digital sovereignty we made, and the important role covered by LibreOffice and the Open Document Format.

Many activities these days, be it sports, social work, arts or free and open source software, are organized in some sort of community. If backed by a respective organization, this not only helps with setting up a structure, but also puts statutes and rules in place that set forth the values and ideals all contributors should share. Inside these communities, there can be various roles. Some of them are formalized, like the board of directors or the supervisory board, others are within dynamically grown groups that can change frequently. Either of those are ideally composed of experienced and enthusiastic community members who take leadership and responsibility.

Based on my experience from the last two and a half decades, I’d like to point out common problems when working in an international free and open source community, share mistakes I’ve made and tell how you can hopefully avoid them.

How to define a true open standard format for office documents

Questions and Answers

Closing TDF Address