In this talk, I will introduce coreboot + LinuxBoot development on Intel Xeon-SP platforms. Presenting why and how for this more open source approach for the system firmware for our servers.

講者

Johnny Lin

Currenlty I have been working on server BIOS (system firmware) for data centers. I have made some contribution to open source coreboot project and LinuxBoot (u-root) under commit author johnny_lin@wiwynn.com.

Debian是一款完全自由的作業系統，以其穩定、安全和跨平台的特性在各種設備和領域廣泛應用。無論是服務器、個人電腦還是嵌入式設備，Debian都是首選之一。其著名的衍生 Linux 發行版如 Ubuntu, Kali, Proxmox VE, Parrot OS 等。

Debian 12 "Bookworm" 是 Debian 在 2023 年推出的最新版本，在許多方面進行了改進和功能更新。其中包含超過 11200 個新軟體套件，總軟體套件已超過 59000 個。

此議程，SZ 除了分享 Debian 12 相關資訊外，也會分享如何參與 Debian 社群，一同貢獻打造自由的作業系統。

講者

SZ Lin (林上智)

SZ 具備 Debian 官方開發者身份。身為 Debian 開發者，SZ 長期貢獻於開放原始碼軟體，目前主要參與資安開放原始碼軟體團隊進行維護及開發。

此外，SZ 也曾擔任 Linux 基金會官方專案 Civil Infrastructure Platform (城市基礎設施平台) 技術指導委員會成員並為 kernel 工作小組前任主席、 Linux 基金會官方專案 OpenChain Project 前董事會成員。

SZ 曾在多個開源國際會議中發表分享，其中包含 OpenChain Japan workgroup meeting, OpenChain Taipei workshop、Debian Conference、Embedded Linux Conference、Embedded Linux Conference Europe、Open Source Summit Japan、Open Source Summit China。

RPA (Robotic Process Automation) - 機器人流程自動化。 主要是使用軟體機器人模擬人類在電腦上不同系統間的操作行為，適合高重複性、標準化、規則明確、大批量的作業流程，並且可用於跨系統、商用軟體及外部網站作業串接。 在本次演講中，我們將介紹 OpenRPA所使用的軟體技術與系統架構，並透過人力銀行履歷篩選以及系統告警通知信件這2個應用案例向大眾推廣OpenRPA 的價值和優勢，並介紹OpenRPA的後續Roadmap規劃與Open Source RPA社團。

講者

IMA(資訊經理人協會)

中華民國資訊經理人協會 （Information Management Association of R.O.C. 簡稱 IMA）成立於1982年，以維護會員使用資通訊系統之相關權益、協助會員提升資通訊系統使用績效、促進資訊經理人管理與資通訊科技之技能，以滿足國內機關、團體及企業面對不同階段挑戰進行轉型的需求，同時提升資通訊從業人員交流為宗旨。2022年後更立意以資訊人才為核心視角，以推動健康的IT產業環境、促進IT產業發展為重點工作目標，並開始更積極地參與技術社群交流，並投入開源技術的研討。Mail: ima@ima.org.tw

講者

鄧宣禮

None

大型自然語言模型LLMs是今年熱門的議題之一，本議題主要分享Google IO 2023年推出的PaLM2模型，除了模型介紹之外，也會整理與知名的GPT模型進行測試與比較。

講者

Jerry

JerryWu is a Google Developer Expert (GDE) in Machine Learning. He is also a Founder & Chief Technology Officer (CTO) in the 亞太智能機器Asia Pacific Machine Intelligence Company (APMIC). Jerry Wu's teaching and research interests include Machine Intelligence, Computer Vision (CV) and Natural Language Understanding (NLU).

Linux 核心的社群常給人一種遙不可及的感覺，想要參與卻又不得其門而入。究竟在 2023 想參與該社群是可能的嗎？到底需要會什麼？要怎麼交流？很容易被罵嗎？ —— 這些問題都會被一一解答。

講者

Shung-Hsi Yu

Kernel Engineer at SUSE working on BPF

Tensor Operator Set Architecture (TOSA) 提供了一套常用於深度神經網路的全張量操作。其目的是使得在不同處理器 (CPU, GPU, NPU) 上運行的各種實現規範，能夠在 TOSA 層面上保持一致的結果。

講者

Odin Shen

Business Development | AI edge computing specialist | SoC arch. exploration | Semiconductor functional safety expert | Cloud Native Computing | Co-founder TinyML Taipei | Embedded Maker | Kid's Coding Volunteer

This presentation explores the Matter standard, a promising solution to the interpretability issues among various smart home ecosystems, and its integration with Home Assistant. It discusses the advantages and disadvantages of using Matter products within specific ecosystems (Google, Apple, Amazon) versus linking them to Home Assistant. The talk also provides guidance for planning smart home product investments in the upcoming year.

講者

Amanda Lam

Amanda is an organiser of ProductTank Hong Kong and Women Techmakers Hong Kong, and co-hosts one of the most popular Cantonese tech podcast, HKPUG Podcast. She manages digital products in Hong Kong Disneyland as her day job.

TPM2 is not about making the computer more trustworthy to you. It is about forcing you, the user, to earn the trust of big corps by tethering your identity to the unique ID of the computer whenever you use a cloud service. The "fairness" of DRM, game anti-cheating, and exam proctoring will be more easily enforced if the big corps have rootkit-like control over everyone's computer. Reverse engineering tools such as virtual machine will not help the user regain total control of his own computer.

講者

ckhung

我的部落格之一： 玩具烏托邦

Flutter 是一個由 Google 開發的跨平台應用程式開發框架，讓開發者可以使用一套程式碼開發 iOS、Android、Web 和桌面應用程式。SonarQube 是著名開源程式碼檢測工具。隨著應用程式規模的不斷擴大，程式碼品質和安全性的重要性也日益凸顯，程式碼靜態掃描是安全程式開發工作第一步．本演講將介紹如何使用SonarQube來進行Flutter程式的靜態程式分析，以確保程式品質和可維護性。我們將深入探討SonarQube的工作原理以及如何配置SonarQube來掃描Flutter程式。此外，我們還會講解SonarQube如何提供即時反饋，並生成報告和指示，幫助開發團隊識別和解決潛在的程式問題。

Flutter is a cross-platform application development framework developed by Google, allowing developers to use a single codebase to develop applications for iOS, Android, Web, and desktop. SonarQube is a well-known open-source code analysis tool. As the scale of applications continues to expand, the importance of code quality and security becomes increasingly prominent, and static code scanning is the first step in secure software development. This talk will introduce how to use SonarQube for static code analysis of Flutter applications to ensure code quality and maintainability. We will delve into the working principles of SonarQube and how to configure it for scanning Flutter code. Additionally, we will discuss how SonarQube provides real-time feedback and generates reports and guidelines to help development teams identify and address potential code issues.

講者

Noflag

各位大家好，我叫沈宜婷，可以叫我noflag，專攻資訊安全，體制外學生，跳脫現今教育制度，是一位致力在資安圈打拼的女性

研究項目：車聯網資安、應用程式安全、機器學習

github : https://github.com/Trinity-SYT-SECURITY

☞中華資安國際 SOC team，實習生

☞NCKU 金融資安實驗室，行動應用程式資安檢測人員

☞Google Developer Group (GDG) Taichung Organizer

講者

Fngi

曾任數家金融機構之開發者/資安工程師/資安主管，其專長於雲端資安規劃，零信任架構，混合環境資安監控及數據驅動資安，亦為 GDG Taipei (台北谷歌技術社群) 資安領域負責人

In the past few decades, event-driven architecture has become a common design paradigm for developing high-performance network servers, as it can efficiently handle concurrent requests using I/O multiplexing. However, the mitigation mechanisms for microarchitectural vulnerabilities such as Spectre and Meltdown have increased the overhead of already system call-intensive situations. Moreover, the synchronous execution blocking feature of traditional system calls makes it difficult to leverage advanced multi-core processors. To address these limitations and identify performance bottlenecks, this session will gradually analyze the evolution of the Linux I/O model and clarify the runtime of applications such as Redis and KeyDB. As a way to overcome these limitations, this session will also demonstrate the implementation of a specialized core module that improves the performance of event-driven network servers through I/O offloading. Our approach does not sacrifice high power consumption or security like kernel bypass techniques in exchange for high-performance throughput and latency. Additionally, we have conducted multiple experiments to compare our method with native systems, and the results show that our method improves bandwidth, latency, and power consumption without requiring the redesign or reimplementation of the entire application.

講者

Steven Cheng

Cheng is a computer science graduate with a master's and a bachelor's degree from National Cheng Kung University, Taiwan. His research interests are in system software and Linux kernel development. He has experience in system profiling and performance analysis of large-scale systems. He has published an academic paper in an international conference and presented several Linux-related topic in the conference held by Linux Foundation.

He has also interned as a silicon engineer at Google, where he contributed to improving the performance of simpleperf, a Android profiling tool. He is a self-motivated learner and a good team player. He is looking for a challenging and rewarding position in the field of system software engineering.

Reach out to me via linkedin: https://www.linkedin.com/in/eecheng/

本次分享將介紹如何透過 Python 使 data 與 ChatGPT 結合，透過 plugin 將台灣政府的開放數據導入到ChatGPT中，從而推動資料的民主化。邀請所有對此感興趣的朋友一同參與！

This session will introduce how I can integrate Taiwan's government open data into ChatGPT through Python and the chatGPT plugin. I invite all enthusiasts interested in this project to join me in helping put wings on open data!

講者

aha

is empowering finance with AI and engaging in the real and virtual world.

Jupyter是一個集文件、編程及執行環境於一身的作業環境，可說是數據科學家們的最愛，在機器學習的領域中同樣活躍，但有沒有想過即使在風馬牛不相及的DevOps範疇中，Jupyter也能大放異彩呢？

圖像介面簡單易記，但難以處理複雜的自動化任務；指令介面則能勝任複雜任務卻難以使用。Jupyter正好集二者之所長，提供圖像介面來操作複雜的指令，自由地擴充，加入圖文說明，還能透過連結獲得更多資訊或功能。

你甚至可至不用費神學習相關的編程智識，因為一切都可以找ChatGPT幫你代勞。雖然現階段AI難以編寫過於複雜的代碼，但用Jupyter編寫的DevOps的操作台並不會有複雜的互動，正好是能讓AI大顯身手的場合。

講者將會分享如何在AI的輔助下把Jupyter客制成DevOps用的操作面板。

講者

Ben Lau

Ben Lau是一位來自香港的開源軟件開發者，曾經擔任香港的Linux User Group的副會長，是2016及2017年的Qt Champion，他也是「封鎖內容農場」這個Chrome套件的作者，他還制作過一款叫做「Dualless」的套件，最高的用戶數達到200萬之多。

隨著機器學習這幾年在各領域的蓬勃發展，AI 開發者們也逐漸想在 MCU 上運行 AI Model ，而 TinyML 社群因此孕育而生，匯集了 MCU/MPU 硬體、系統軟體、模型開發者、應用開發者來研討如何強化 ML 在 Tiny Devices 的各式機會。另一方面， AI Benchmark 的組織 MLCommons 也注意到 AI 開發者們在 Tiny Devices 上的需求，而開發了 MLPerf Tiny ，其為目前用來測量 AI on MCU 上最權威的 benchmark，AI developer在挑選 MCU 時，首重的就是 MLPerf Tiny 跑分的成績。Skymizer 在今年六月公佈的 MLPerf Tiny 上 ONNC 的跑分成績在所有的 benchmarks上，效能與記憶體使用量均優於 MicroTVM。我們使用了兩種不同的 MCU – STM 與新唐 – 搭上兩種不同的作業系統 Zephyr 與 mbedOS，在效能上達到 6%~33% 不等的優勢。

在本篇演講裡，我們會先介紹 TinyML 社群以及市場趨勢，再來介紹 Skymizer 參與 MLPerf Tiny 的經驗、與 ONNC 如何以編譯器的角度來優化 AI Model 在硬體上運行的效能。

講者

Luba Tang

Luba Tang is the founder and CEO of Skymizer Taiwan Inc., which is in the business of providing system software to IC design teams. Skymizer’s system software solutions enable AI-on-Chip design houses to automate AI application development, improve system performance, and optimize inference accuracy. Luba Tang’s research interests include electronic system level (ESL) design, system software, and neural networks. He had focused on iterative compilers, ahead-of-time compilers, link-time optimization, neural network compilation, and neural network optimization. His most recent work focuses on exploiting various types of parallelism from different accelerators in a hyper-scale system-on-chip.

講者

謝政道

Dr. Cheng-Tao Hsieh is the Compiler Team Lead at Skymizer, leading the development of ONNC. His research expertise includes static analysis of peak power and automatic addition of redundant circuitry to tolerate delay variation, as well as experience in low-power high-level synthesis targeting FPGAs. Prior to joining Skymizer, he held a position in EDA at Qualcomm and earned a Ph.D. in Computer Science from NTHU. His professional knowledge and leadership contribute significantly to the team’s growth and product innovation.

講者

Peter Chang

Peter Chang is the business development manager and the co-founder of Skymizer Taiwan Inc. His research interests span areas in operating systems, virtualization, and computer architecture. Currently, he focuses on topics in hardware/software co-design and benchmarking on Machine Learning. He is also devoted to participating in the MLPerf Tiny and TinyML communities. He was also the maintainer of SkyPat, an open-source performance unit-test suite, and ARMvisor, one of the Kernel-based Virtual Machine solutions on ARM architecture.

在產生密碼用於加密各種協議，但隨機產生的數怎麼產生的？這些數字真的夠「隨機」嗎？產生亂數又需要多少時間？

在這個議程，想跟大家分享討論這些議題 - 運用 LFSR 得到亂數的手法 - Linux 運用哪些資訊來產生亂數？ - Linux 產生亂數的整個架構 ，/dev/random 和 /dev/urandom 對應到內部哪些實作？ - Linux RNG 如何達到 CSPRNG 要求？ - 如何去評估亂數的好壞，如何確保他「夠亂」

講者

黃柏愷

成大資工系、陽交大資工所畢業 現任職於 Trend Micro

1. The high level introduction of Open Source FW,
   • Open System FW/OpenBIOS with different architectures (UEFI, coreboot, etc..)
   • OpenBMC for all in one of server board and modular design for different devices in Datacenter
2. How the Open Source FW become major solution for server industry
3. Current status of Open System FW and OpenBMC for the commercial server
4. The challenge and things need to be done in future of Open Source FW

講者

Hancock Chang

25+ years experience with cross domain skills such as System Solution engineering, product/program management for various product portfolio including PC, Server (For Data Centre and Telco Infrastructure) and Embedded Solution (Automotive and BMC/BIOS), success in server product management through the driect collaboration continuously with various open source communities including OCP, OpenBMC, Linux Foundation, ONF, OSFC.

KVM (Kernel-based Virtual Machine) 是 Linux 核心中的系統虛擬機器基礎建設，它是 Linux 核心模組，能讓 Linux 系統成為一個 Type-2 Hypervisor 。KVM 透過硬體虛擬化支援 (Intel VT, AMD-V) 來提供 CPU 和記憶體虛擬化功能。藉由硬體虛擬化技術，客體作業系統 (Guest OS) 不必經由軟體模擬或轉換指令，可以高效率且安全地直接執行在硬體上。使用者空間 (User Space) 的程式只要負責模擬週邊裝置、呼叫 KVM API ，即可建立並高效率地執行虛擬機器。

在這次議程中，將介紹 KVM 運作原理，並展示一個以 KVM 為基礎的精簡虛擬機器管理程式實作，它能在 x86-64 及 arm64 平台上運作，提供了基本的 VirtIO 儲存裝置、 serial 裝置 (用來作為 console) ，並且能成功在上面執行 Linux 系統。

講者

王彥傑

就讀國立臺灣大學資訊工程學系碩士班的學生，對 Linux 系統、系統軟體開發有興趣。 如有需要歡迎與我聯繫: yanjiewtw@gmail.com GitHub: https://github.com/yanjiew1

考量到對於 sandbox 執行環境的需求，一個高效能且可嵌入式的 RISC-V 模擬器對滿足這樣的需求至關重要。受益於 RISC-V 的開源指令集和龐大的系統生態系，我們的模擬器得以用較低的記憶體開銷和可接受的執行速度來模擬 RISC-V 程式。 本次將介紹如何實作一個精簡且高效指令集模擬器，另外，也將討論如何透過加入周邊模擬，改造為可運行 Linux 的 RISC-V 系統模擬器。

講者

陳彥甫

目前就讀成大資訊工程所碩士二年級並實習於晶心科技，實驗室為前瞻系統實驗室，指導教授為黃敬群教授以及涂嘉恆教授，研究題目為 RISCV 模擬器，專案名稱為 rv32emu，主要研究方向為探討不同實作技巧對編譯器以及現代處理器的影響，並用以提升模擬器效能。

講者

莊集

現職是國立成功大學資訊工程所碩士二年級的學生，所屬前瞻系統研究實驗室，指導教授為涂嘉恆教授，研究主題與量子電腦模擬器相關，因修習 Linux 核心實作課程，參與 RISC-V 系統模擬器相關專案。

作為一個系統軟體開發者常常需要在茫茫大海中找臭蟲(BUG)，需要快速閱讀大量讀程式碼，所以我們需要一個可以幫助我們的IDE，往往正確的配置IDE也是非常重要，因此這邊我會分享如何配置VScode讓他能很好的幫助我們。對於找臭蟲(BUG)與GDB的關係，這主題總是有講不完的經驗 ，使用GDB就像有上限很高的武器可是不會用的話，還是只能把它放旁邊。還有這神兵利器竟然是開源免費軟體比起商業軟體txxxx32更顯高CP工程師的價值。這次會分享如何撰寫GDB python script跟如何讓它幫助我們找到kernel 的臭蟲。這個議程會實例分享如何運用VScode及GDB幫我們找到riscv kernel BUG跟追蹤kernel 行為，聽完後會覺得GDB比printk方便節省加班時間。

講者

wayling

打雜工程師

Almost every system has BIOS firmware. Be it an individual box or enterprise-grade workload Servers. What is the need to change it? current BIOS is the result of so many innovative cycles. so just because it is proprietary firmware and we believe in OpenSource we should change it? In this talk, I will discuss How Firmware changed from time to time? Every time the driver and rationale in the Firmware growth path with the history of the Firmware? The Pace of Firmware improvement is not the same as it was sometimes back! Is it only Cloud or cyber security which has increased multi-variant releases in firmware? Like normal software, can we go for tailored firmware that can really make a direct difference for any business use case? Firmware capabilities should be scoped closely limited to booting the device only or a tailored firmware can aggregate the innovation and can help to solve many problems normal software is struggling with? This talk will have content on Libreboot firmware in terms of a tough journey and great acknowledgment for BIOS replacement.

講者

Preet Sharma

Preet is a Solution Architect at NCS-Singapore with 15 years of rich experience in Solutioning, Transformation, and Digitization. Preet has witnessed a revolutionary journey of technologies from dBase to SQL/NoSQL/GraphDB, from Pascal to Python, Onprem to Cloud, waterfall to agile along with multiple Business domain knowledge.

He is always keen to learn and conceptualize new things for better customer experience management. He is a customer-obsessed and collaboration believer.

使用者往往在使用自己的硬體時並不知道firmware 與hardware是在不安全的設定下執行，例如secure boot disabled 與 firmware 沒有更新到已安全性修復版本。或在一些需要更高security level 的場合，但卻無法立即檢查security level，因此需要一種自動化的方法來評估系統的安全性。fwupd 是用來自動更新系統上的firmware，包括了BIOS, ME, 以及各式周邊的firmware。fwupd 並有能力搜集所有firmware 的版本與系統的設定，在執行期間並可以使用這些資訊來評估系統firmware與hardware 的安全性。fwupd Host Security ID (HSI) 定義了Host security level，並使用一個簡單的數字告訴使用者platform 的安全程度。這個ID是由fwupd 社群所定義的條件來決定。同時在GNOME control panel 的 “Device Security” panel 中也可以輕鬆的取得。讓使用者可以更容易的了解系統的暴險程度。

講者

Kate Hsuan

目前主要貢獻上游軟體開發，包括了 Linux Kernel, fwupd, libcamera, GNOME以及Fedora。Kate也一直使用及貢獻開源軟體，在QCT時負責ONAP (Open Network Automation Platform) 電信產業專用開源專案，並與國外電信公司一起開發與貢獻 ONAP DCAE Datalake專案。在Hopebay Technologies 時利用Ceph 開發出Scalable Software Defined Storage，也長期關注OpenStack發展，並使用它開發和沛公有雲系統。

一般播放PowerPoint簡報時都需要一組無線遙控器來控制播放下一頁動作，這裡使用運動手勢就能代替簡報遙控器。主要使用Seeed Xiao nRF52840 Sense(Arduino Nano33 BLE Sense同級產品）作成類似運動手環裝置，並利用開發板上的運動感測器及Edge Impulse TinyML雲端開發平台來訓練運動手勢模型，最後產生Arduino源碼，手動整合BLE HID功能，就能以手勢變成電腦上的按鍵按下，藉此來控制簡報播放。

講者

Jack OmniXRI

來自歐尼克斯實境互動工作室的神奇傑克，在TinyML領域另有一個小名「史蒂芬周」。

Why Solana in 2023? 在這次的分享中，我將會講解 Solana 的技術特性與 Solana 高性能的秘密，以及對圈內一些關於 Solana 的誤區進行說明（例如：SQLana / 停機問題 / FTX 事件之後的發展），最後，我也會分享一些能幫助新手快速上手的學習資源。

講者

Jim

I run DappioLb and Gen3

Tenok (在阿美族語中意指 Kernel) 是一個支援 ARM Cortex-M 架構、6000行左右、目標應用為機器人控制 (Robotics) 以及物聯網 (Internet of Things) 的實驗型微型即時作業系統。本專案是由講者所開發之專案，希望與聽眾分享關於作業系統開發的細節以及經驗。

講者

鄭聖文

畢業自陽明交通大學 (機器人碩士學程，2021年) 及靜宜大學 (資訊工程系，2019年)，主要興趣為機器人系統、嵌入式系統以及即時作業系統。曾參與開源四軸飛行器韌體開發 (UrsusPilot，因主要成員創業，已轉為閉源) 及小型實驗型自駕車系統設計 (Puyuma，基於電腦視覺及 Real-time Linux)，熟悉機器人控制以及導航，更詳細的介紹可見: http://wiki.csie.ncku.edu.tw/User/shengwen

地圖是生活中相當常用的軟體。Google 與 Apple 地圖無所不在，但它們顯然都是封閉原始碼的軟體，對開發者來說是黑箱。這些供應商對某些用途來說並不好用，例如公共衛生、政府服務與新聞地圖。

在本次演講中，我將帶您了解：

1) Leaflet 與 MapLibre 等函式庫與 GIS 有何相關之處，在哪裡使用它們，以及它們如何讓沒有地理背景的 JavaScript 程式設計師使用地圖

2) OpenStreetMap 資料集與全端互動地圖 API 之間的區別製作

3) 地圖的設計與資料注意事項，如何製作縮小的圖磚化地圖檢視、地圖偏差與國際化

最後，我將會討論開放原始碼的 Protomaps 專案，這是一個用於製作與提供圖磚化網路地圖的完整系統。我將把它與其他開放與專有的替代方案進行比較，展示它的使用者，並解釋我打算如何使其長期可持續發展。（中文翻譯，原文請見英文版）（由 OpenStreetMap x Wikidata 臺灣社群與 COSCUP 大會聯合邀請）

講者

Brandon Liu

I'm the lead developer of the Protomaps open source project - a self-hostable alternative to Google Maps and other proprietary APIs, built on open data. I used to live in the United States, now I live in Taipei, Taiwan!

Vortex is an open source Hardware and Software project to support GPGPU based on RISC-V ISA extensions. Currently Vortex supports OpenCL/CUDA and it runs on FPGA. The vortex platform is highly customizable and scalable with a complete open source compiler, driver and runtime software stack to enable research in GPU architectures/compiler/run-time systems.

講者

Blaise Tine

Blaise Tine is a Professor in Computer Science Department at University of California, Los Angeles. He is a recent PhD Student graduate in the school of Computer Science at Georgia Institute of Technology, where he has lead the Vortex project. His research interests are in the area of hardware accelerators and software co-design, focusing in the architecture design, programming languages and compiler tools to support heterogenous computing. He is also interested in domain-specific applications of accelerators such as 3D graphics, graphs analytics, and machine learning.

本議程試圖回答以下問題： 1. Side Project 的價值是什麼？ 2. 開源專案對職涯的價值是什麼？ 3. 開源公益專案能賺錢嗎？

講者

Denken Chen

《立法院大小聲》發起人，軟體開發與寫作經驗超過九年。作品可見個人網站：https://denkeni.org

Golang is widely used in cloud-native and distributed systems. To keep the observability of the system, distributed tracing is a key technology. Previously, vendors and open-source communities always recommended using SDK to instrument codes, meaning, the developers have to change codes. In this session, we are going to introduce, for the first time, a complete auto instrumentation go agent solution built out-of-box by Apache SkyWalking.

講者

吴晟

A searcher, evangelist, and developer in the observability, distributed tracing, and APM. I was a board director and member of the Apache Software Foundation in 2021, as the first Chinese in the Apache board. Love open source software and culture. Created the Apache SkyWalking project and being its VP and PMC member. Co-founder and PMC member of Apache ShardingSphere. Microsoft/Alibaba Cloud/Tencent Cloud MVP. AWS Hero As a founding engineer at Tetrate, lead the observability for service mesh and hybrid cloud.

講者

刘晗

刘晗，Tetrate工程师，SkyWalking PMC成员，SkyWalking eBPF探针，SkyWalking Go Agent核心贡献者 曾就职于滴滴负责全链路压测部门，拉勾负责教育业务线

In this talk, we delve into the realm of large language model-powered autonomous agents, the newest AI wave revolutionizing digital and physical interactions. Through examples of open-source projects like AutoGPT and BabyAGI, we will unravel the innovative aspects that make these agents stand out, including their prompt engineering techniques, memory system design, and the external toolset they use to augment their capabilities.

We will discuss their implementation, potential benefits, risks, and the ways in which open-source community can tap into their potential. This session aims to provide a comprehensive introduction to LLM-powered agents. Come join this exciting discussion about the anatomy of autonomy, which may define or destroy our future!

講者

曾亭翰 Michael Tseng

18, building for the singularity

I’m interested in ai, bio-inspired computing, and most of the emerging tech/science developments. I will be in sf this fall as a freshman at Minerva University.

🌉 love startups, cities, long articles, ocean blue, and night sky

Blog: tsengtinghan.github.io

今年是justfont的開源字型粉圓體推出的第三年，從誕生以來這三年發生了什麼事，未來又將面臨什麼樣的命運呢？！

講者

千秋

はおーっ！涼風千秋です！

前端架構師，字體設計師，業餘繪師，神使狐狸大小姐，喜歡優雅、可愛的東西

喜歡跟大家分享有趣好玩的事物\٩( 'ω' )و ///

噗浪佈景東京乃空的作者，justfont 粉圓字型的設計師。

本次 OpenStreetMap x Wikidata 議程特意開設開源 GIS 討論議程，探究成功的開源專案 QGIS，在台灣本地的面臨社群動能不足狀況，翻譯軟體或是教材進度遠遠落後開發進度，試圖在本次論壇識別問題並提出解決之道

講者

林宥伯

在提供服務過程，接觸了GIS領域約莫四年至今，累積的經驗與瓶頸，希望能藉由分享，與開放社群的夥伴共同發想，找出可行的解決方案，推動領域發展

講者

Dennis Raylin Chen

台灣 ê OpenStreetMap 圖客 kah Wikidata 貢獻者(User:Supaplex)，sī 臺北 OpenStreetMap x Wikidata 月聚 ê 共同主持人。最近攏 teh 大會用臺灣台語講演講。嘛是臺灣維基協會 ê 理事之一。

講者

林政道

None

講者

王祖修

None

Hugging Face 可以說是 AI 模型的 GitHub。最早他們是為了解決自然語言模型的高難度和不相容而誕生的，到現在上面除了自然語言的開源 AI 模型外，還有有許許多多的開源 AI 模型。根據 Google 內部流出的文件指出，開源的 AI 模型很有可能主導接下來整個產業的發展。也因此 Hugging Face 的角色愈來愈重要。

本次演講是針對 Chatbot 入門新手而設計的，會帶大家使用 Hugging Face 上的自然語言處理模型，並且一步步示範串到 LINE 裡面，做成一個聊天機器人。

講者

Ko Ko

Ko Ko 是微軟 AI 領域的最有價值專家(Microsoft AI MVP)，同時也是台灣知名的講者。 他是許多技術年會的講者，包含了 COSCUP、 .NET Conference、ModernWeb、PyCon APAC、DevDays Asia 等等的大型年會，累計已超過萬名聽眾。 他熱愛向聽眾分享他的經驗與所學，並且相信這樣分享的過程中能讓彼此都更進步。 今年(2023)三月初他辦的活動 Global AI Bootcamp 台灣線上場，圍繞在 ChatGPT 的原理與 Azure 自然語言處理的應用，門票 150 張秒殺，並且送出一百張的紀念 NFT，相當受到聽眾歡迎。

Generative models on "modern" mobile phones have the potential to revolutionize the way we interact with our devices. They could be used to generate personalized content, create new forms of entertainment, and improve the accuracy of machine translation. However, there are still a number of challenges that need to be addressed before generative models can be used effectively on mobile phones. In this talk, I will use Stable Diffusion and llama.cpp as main examples talk about how people try to address these challenges. Despite these challenges, there is a lot of potential for generative models on mobile phones. As the technology continues to develop, we can anticipate seeing a wide range of new and innovative applications for generative models on mobile devices.

講者

Koan-Sin Tan

Koan-Sin Tan is an old programmer, who learned to use “open source” stuff on VAX-11/780 running 4.3BSD before the term “open source” was coined. He is interested in running neural networks on edge devices the past 7 years. He is a TensorFlow contributor. He converted Stable Diffusion to tflite format and wrote some glue code for it [1].

[1] https://github.com/freedomtan/keras_cv_stable_diffusion_to_tflite/

In recent years, many open-source-based startups are coming from China, such as M5Stack(Hardware) and PingCap/TiDB(software). I`m an official member of KaiYuanShe(China Open Source Alliance), Leading some document translation. And my company Switch Science is an Open-Hardware distributer; I also have success with some marketing projects in Japan with Shenzhen Hardware company. I will introduce some cases of Business development about hardware and software.

講者

TAKASU Masakazu(高須正和)

TAKASU Masakazu Official Menber/正式成员, KaiYuanShe/开源社 He has three roles in Society. The first is Business Development in Shenzhen, Japan, and the Asian area. He works with open hardware startups based in Shenzhen, such as Seeed, M5Stack, iMakerbase, Heroad VC, Etc. And his company, Switch Science, is the most famous maker tool e-shop in Japan and the DIY Indie product marketplace. The Second is the Researcher. Researcher and Lecturer of WASEDA University Business school, Garage Sumida Lab, Mentor of iMakerBase Shenzhen, MTOM Shenzhen. His book ECOSYSTEM BY MAKERS is famous in Japan. He also translated THE HARDWARE HACKER book into Japanese. That book by Bunnie Huang, who is a world-famous Hardware hacker. The third is Community development. He is a Co-Founder of Nico-Tech Shenzhen (Shenzhen-based Open Hardware community), with Over 3000 members in a Facebook group. The KaiYuanShe(China Open Source Alliance) Support Maker Faire Shenzhen and Singapore, Maker carnival Shanghai and some events take us on a journey to Japan’s homeland and their exciting ways of using technology, design, and science. He is the most experienced person in Maker Faire in Asia. He joined KaiYuanShe(China Open Source Alliance) in 2021, currently only one global member. In addition, he was honored Community star in 2021 and 2022. https://www.linkedin.com/in/takasumasakazu

介绍（中文） 高須正和是新加坡和深圳製滙節(Maker Faire)及上海創客嘉年華 (Maker Carnival) 的資深成員。Maker Faire是在世界各地享負盛名的活動，向世界展示創新項目和背後具創意的頭腦。高須先生擅於揉合並引用日本首屈一指的科技、設計與科學等知識，其豐富的經驗和人際網絡連繫著亞洲多個地區的Maker Faire。他現駐居深圳。日本早稲田大学非常勤講師。开源硬件爱好家社区Nico-Tech Shenzhen共同发起人

高須先生亦是日本知名獨資科技產品製造材料公司Switch Science全球業務發展人員，其公司在DIY手作零件銷售市場上亦佔舉足輕重的地位。此外，他的著作《ECOSYSTEM BY MAKERS》一書，在日本廣受歡迎。 现在他是开源社（China Open Source Alliance）正式成員,几一个国际成員。

隨著臺灣政府開放資料的日漸完善，適合匯入 OpenStreetMap (OSM) 的地理資訊資料也相繼出現，例如各縣市地址與建物。然而 在台灣 OSM 地圖資料的成長，與開放資料的資料集內容十分龐大下，如何匯入與維護變成如今社群關注的焦點。本議程將帶大家介紹近幾年來匯入進行的狀況與目前我們遭遇的各種議題。

講者

littlebtc

None

Development of open source silicon platforms applicable to difficult real world use cases, much as Linux has become in the operating system space, quickly run into difficulty. A CPU ISA or small system on chip is only one part of a platform. The diverse and complex interfaces, signal processing, and compute tasks such a platform is called upon to provide mean a system approach is necessary.

We outline our experience building the J-Core platform, and how scratching our own itches lead to counter intuitive design decisions. Designed from the ground up to cover mixed signal processing in systems with complex and secure edge computing tasks, we also needed portability to multiple FPGA families, low cost silicon processes, and scalability for performance on advanced process nodes.

J-core is a completely open source platform, from the transistors through the OS and end user applications. Designed in the context of real world product development and deployments, the resulting platform is unique in the open hardware space.

講者

D. Jeff Dionne

Jeff is the original author of the uClinux kernel and OS. He has been designing and building open embedded systems with linux since the 1990s.

三個主題

• (1) 分享的初衷
• (2) 跨域實務經驗
• (3) 持續與人互動

從分享的初衷開始，並逐漸擴展到國際舞台上，在美國的 Mozilla Summit 擔任講者、在歐洲藍牙論壇成為與談人。透過他人分享的經驗，能夠讓自己成長，同時透過自己的分享，除了能讓他人獲益，也能整理自己思路。這種正向循環在講者的生涯發展中，是個關鍵飛輪。

在跨域實務經驗方面，講者將分享了他在開源社群不同領域的學習經驗，從擔任 COSCUP-GNOME Asia 外交官、COSCUP 銷售組、到 Mozilla Support 在地化的專案負責人，這些經驗使他有機會挑戰產品和技術整合部門主管。他從這些實踐中學會了溝通、銷售、拆解和梳理流程等技能，並將它們應用於職場工作和產品規劃中，促進了他的職業發展。

最後，講者強調了持續與人互動的重要性。他從 Mozilla 開源社群，到 AWS 技術社群，持續在不同領域中探索，例如接觸 Mozilla FirefoxOS 而開始動手玩 AOSP，再因此而在 AWS 上建立健身物聯網。這種持續的互動使他有機會能夠瞭解市場動向，並接觸到各種技術，進而職涯上逐步前進，往理想與夢想邁進。

在摸索職涯的你，是否想要知道

• 從分享的初衷如何擴展到國際舞台？
• 一名阿宅工程師，如何在美國 Mozilla Summit 和歐洲藍牙論壇上，擔任講者以及與談人？
• 透過持續分享，如何實現自我成長？
• 在開源社群中的跨域實務經驗是什麼？
• 如何應用社群經驗，促進自己的職涯發展？
• 持續與人互動有何重要性？
• 講者如何從 Mozilla 到 AWS 技術社群中探索不同領域？
• 講者如何了解市場動向並結合到自己的工作當中？

參與這場演講，一起聊聊，一起連結，一起探索職涯的各種可能性 :)

講者

Ernest Chiang

Ernest Chiang 於 2008 年開始了他的 AWS 之旅。他熱衷於參與在地技術社群，以及透過部落格文章，分享各種 AWS 的思路，將 AWS 拆解後的技術架構與產品、商業思維連結起來。自 2011 年以來，在 Ernest 服務於 PAFERS Tech ，參與各領域的產品與技術整合，將手邊許多軟硬體及雲端整合專案，運作於 AWS 全球和 AWS 中國地區，服務世界各地的客戶。

Ernest 對於技術社群參與擁有豐富的經驗。2009-2014 年，參與籌備台灣最大的 FLOSS 會議【開源人年會 COSCUP】。2010-2019 年之間，Mozilla 代表。

他喜歡跨領域的產品開發和技術管理工作，目前在 PAFERS Tech 拆解健身健康產業，提供歐美客戶軟硬整合、物聯網解決方案、以及提升營運效率的組織作業流程顧問服務。他曾在台積公司擔任半導體領域的製程整合工程師。

2020 年，他獲邀成為 AWS 社群英雄，繼續他的 AWS 學習旅程。

---

如果你跟他一樣喜歡將事情複雜化，歡迎找他一起交流 PKM 個人知識系統 工作流程 :)

本文將介紹如何利用本地端顯卡建立Meta新推出的大語言模型LLaMA後的c/c++輕量版本Llama.cpp，並利用Micromamba管理其運行環境。 首先，Llama.cpp是一種先進的文本預測模型，類似於GPT-2與GPT-3，但未經過微調，適合於大範疇的問答應用。使用本地端顯卡進行模型訓練能更有效地管理資源。 我們將利用Micromamba，這是一個簡潔、快速且隔離的Python環境管理工具，用於安裝和管理需要的Python庫。 接著，模型訓練的原理主要在於使模型學習如何根據上下文預測下一個文字。訓練完成後，模型將能生成與訓練數據相似的新文本。最後，關於Llama.cpp處理中文的能力，儘管其主要以英文為主，但透過適當的訓練與微調，仍有潛力達到理想的中文生成水平。

講者

Will

hello,我是張奕爲(Will),喜歡挑戰不同領域不同事物，目前Java具備建構全端網站經驗、專案開發方面經歷,約4年的開發經驗,往後將著重於Web Application專案開發。業餘會開發一些自己感興趣的side project

身處資訊社群中，我們往往會不經意的試圖自動化許多流程——從工作分配的方式、會議時間的敲定、工作人員的招募與財務報支，有太多行政手續是可以用電子化表單甚至系統取代的了。然而在追求便利、統一、效率的同時，我們又交換掉了什麼呢？這場議程將會用類似讀書心得分享的方式，跟大家談談對數位系統自動化的反思與過程中應當納入考量的各式正當程序 (due process) 保障。

講者

Poren Chiang (@RSChiang)

姜柏任（Poren Chiang，​@RSChiang），自由軟體開發者、SITCON 共同發起人，目前在中央研究院法律學研究所擔任研究助理。除了專注於數位政府與開放資料研究外，也關注資料監控、自由軟體授權、社群媒體管制、電子裝置搜索及電子投票等資訊法與憲法領域議題。

We are Boledu Foundation. Our missions are (1) Target and Extend the Google Open Source Silicon program and (2) to Design hands-on lab materials for training, (3) Promote IC education. For the Caravel Harness, We have developed a Caravel FPGA validation platform. In addition, we also develop a software debugging framework which includes GDBWave, and a Riscv-Tracer to facilitate the verification process. The GDBWave parses the waveform after RTL simulation to manipulate GDB debugging with VexRiscv CPU. The Riscv-Tracer helps to translate waveform display to Riscv instructions representation. The Caravel FPGA validation system enables users quickly prototype their user project design on the Xilinx PYNQ board. In addition, the FPGA design and hardware bitstream can be developed using the Vitis tool free from Xilinx, Then be validated on Boledu online PYNQ board with Jupyter Python3 testbench.

講者

Jiin Lai

We are Boledu Foundation. Our missions are (1) Target and Extend the Google Open Source Silicon program and (2) to Design hands-on lab materials for training. (3) Promote IC education Boledu funder, Jiin Lai is the Chief Technology Officer of VIA Electronics. He has over 30 years of experience in the PC industry and the past 12 years in the storage space. Earlier in his career, he was a software engineer developing EDA tools. He later co-founded VIA Electronics. He led the engineering team that developed Intel and AMD compatible chipsets and x86 compatible processor. For the past decade, he developed SSD controllers and later shifted his focus to developing distributed computing storage systems. His responsibilities include product and architecture development with an eye toward future computing architecture needs. Has more than 50 US patents. Since 2020, he has taught "Applied Acceleration and Advanced Synthesis" courses at NTU, NTHU and NYCU. He also founded the Life Bridge Educational Foundation to promote technology education in schools.

2013年因Docker技術出現後，意識到容器將顛覆IT產業，開始積極投入開源社群，無心插柳成為譯者，技術專欄作家，到金融業技術架構師；透過Docker，學習和使用開源技術已成為資訊人員職涯發展的重要一環。 回顧十年來，如何透過開源技術及開放態度逐步培養個人能力及創造自身價值，如何從程式開發到企業架構規劃的職能轉變，深化硬技能的基礎到培養必備的軟技能，希望這十年的經驗可供開源同好借鏡及反思。

講者

philipz

Philipz鄭淳尹，Docker.Taipei社群共同發起人，國泰金控技術架構師，曾任台北富邦銀行雲端系統部架構師、微軟MVP、momo購物網架構師、臺北榮民總醫院資訊工程師、玉山銀行資訊處專員、宏碁eDC維運工程師。開源技術愛好者，曾在多間大學資工系擔任Docker容器技術講師，並翻譯審閱多本容器技術書籍。

The need for efficient and scalable data management and messaging protocols in distributed systems and IoT applications has become increasingly important in recent years. Zenoh is a new open-source protocol that addresses these issues by providing a lightweight, efficient, and scalable solution. It takes a data-centric approach to communication, which focuses on the data itself rather than the location or identity of the devices producing or consuming the data.

Moreover, Zenoh is built on Rust, a systems programming language known for its high performance, security, and reliability. Rust's unique features make Zenoh a fast and robust solution for managing and processing data in distributed systems and IoT applications.

In this talk, we will explore the architecture and features of Zenoh and discuss real-world use cases.

講者

Yuyuan Yuan

Rust enthusiast who likes to study language design and its applications in various domains. I am currently working on Eclipse Zenoh as a ZettaScale Taipei Team member.

OpenAI泛指利用生成式人工智慧，來產出程式碼、影像、文字等各類應用，並採開放式近用的方式提供服務。然而，從上游的訓練素材，有取用他人素材的侵權爭議，例如GitHub Copilot；而從素材生成的中游，有相關產出得否主張創意保護的著作權適格爭議；而從產出應用的下游，有模組授權拘束規範的排除要求，本講座將綜合探討這些爭議問題，並從智財權的基礎框架，及AI監管法令，開放AI模組授權的分析，來引導聽眾學習並進一步掌握相關基礎知識。

講者

Lucien C.H. Lin 林誠夏

林誠夏 (Lucien Lin) ，網路暱稱為 “Lucien” 或 “lucien.cc”，知識背景為科技法律，05-16年間任職於中央研究院從事公眾授權模式的研究，14-16年擔任台灣創用CC計畫法律項目主持人。近年的研究成果包括：協助歐洲自由軟體基金會，擔任自由開源軟體法律參考書台灣專章的編撰作者、參與國家發展委員會就「政府資料開放授權條款」第一版本的討論與編撰，並應各界參與者之諮詢，回應與釐清Open Source、Open Data，以及CC授權等智慧財產權及公眾授權應用問題。目前於鈞理知識產權事務所，以及開放文化基金會擔任法制顧問，撥付工作之餘的心力營建國內開源授權知識的分享網絡(Open Source Legal Network, Taiwan)，並於16年8月開始，擔任行政院與文化部資料開放諮詢小組委員。

隨著微服務架構的不斷發展，Protocol Buffers v3 作為針對定義溝通介面的解決方案獲得了不少軟體產業的青睞，講者自己也在現職公司中充分享受到它帶來的生產力提升。

講者期望透過此次演講做個簡單的 Protocol Buffers v3 入門介紹，並從軟體產品開發的角度來討論他在開發中的合適使用情境。

講者

Rain

目前在 Netskope 擔任 Senior Software Engineer，心血來潮時會用 Python 和 Go 為自己喜歡的開源工具貢獻微薄之力（但不一定會被接受），同時也是名桌遊愛好者。

在本議程中，Weithenn 將以自身從事 IT 行業至今 20 年的經驗，分享 Infrastructure Architect 養成技能樹，從一開始的專科畢業，憲兵退伍，油漆工，網工班……等，直到開始接觸 FreeBSD 之後，整個職涯卻產生化學變化開始了不同的走向。

此外，也將分享 Weithenn 在職涯上的個人作法和選擇，例如，角色定位、學歷無用論？ 餘裕、避免落入沉沒成本謬誤、時間管理、制約、自我學習…等，如何從寫部落格開始，到撰寫雜誌技術專欄，出版多本書籍，公開活動演講，進而獲選 Microsoft MVP 並成為台灣第一位 VMware vExpert……等。

最近 Weithenn 的 2023 年目標，則是希望更深入了解 Google Cloud 雲端環境，並朝著 GDE(Google Developers Experts) 之路邁進……

講者

Weithenn

連續十二年成功獲選 Microsoft MVP 和 VMware vExpert 技術專家，並長期擔任網管人雜誌技術專欄作家，專長為打造 SDDC 軟體定義資料中心、HCI 超融合解決方案、IaC 基礎架構即程式碼環境……等，著有微軟 S2D 軟體定義儲存技術實戰……等共 19 本著作。

曾任 Cloud Summit Taiwan 2023, Kubernetes Summit 2022、DevOpsDays Taipei 2022、Cloud Edge Summit Taiwan 2022、SRE Conference 2022、DevOpsDays Taipei 2021、Cloud & Edge Summit Taiwan 2021、Cloud & Edge Summit Taiwan 2020、Global Azure 2020、StorTrends 2020 儲存趨勢論壇、VMware vForum Taiwan 2019、OpenInfra Days Taiwan 2019、Cloud & Edge Summit Taiwan 2019、Kubernetes Summit 2019、Cloud Native Forum 2019、Dell Technologies Forum 2019、Windows Server Summit 2018 Taiwan、台灣微軟 TechDays、MVP 技術關卡破解日……等研討會講師。

建置私有雲的 Kubernetes，需要解決的就是監控問題，本議程將會講解怎麼組合 Prometheus + Grafana，除此之外，還有帶你理解常用的 CRDs，打造出屬於你的監控報表。

講者

梯口 tico88612

哈囉，我叫做梯口，目前在花蓮讀碩士寫論文，但同時也在工作，自從大四實習踏入 IT/SRE 領域後，持續研究任何 Cloud Native 的新技術、DevOps 的思維，喜歡分享自己所看到的人事物。

This talk covers integration patterns of WebAssembly through backend software, like CLIs and side cars. It shows how to embed WebAssembly into Go applications, and what to watch out for when programming. When we’re done, you’ll have a good sense of how WebAssembly fits into what you’re working on.

講者

Adrian Cole

Adrian is an engineer working at Tetrate on Open Source projects. He’s been a routine contributor to open source for over a dozen years. Lately, he spends most of his time on wazero: the zero dependency WebAssembly runtime for Go developers. His past notable project work includes Zipkin, OpenFeign, and Apache jclouds.

Go 1.18 brought Generics. When we heard this official announcement initially, we thought we could finally write Go programs quickly and run them efficiently. But after one year, the characteristics of Generics suffer us in many ways. It doesn't become Swiss knife as we think. So I will share some experiences of using Generics, point out what is going on in the design of Generics, show how to use Generics well, and when to apply Generics.

講者

Hsueh-Tsung Kuo

• A programmer from game company in Taiwan.
• Backend (and temporary frontend) engineer.
• Usually develop something related to my work in Python, Ruby, ECMAScript, Golang, C#.
• Built CDN-aware game asset update system.
• Business large passenger vehicle driver.
• Care about chaotic traffic in Taiwan.
• Draw cute anime character in spare time.