This document discusses trends and developments in deep learning research presented by Jeff Dean from Google. It covers improvements in accuracy from increased data sizes and model scales, the prevalence of neural network approaches over other machine learning methods, and new areas of focus like reinforcement learning optimization and deployment. Application areas mentioned include computer vision, speech, language processing and vertical-specific domains.
Chaos Engineering on Microservices - ???, AWS ???????? Channy Yun
?
This document contains information about chaos engineering and experimenting with latency injection between microservices. It discusses distributing traffic between production, control, and experimental versions of a service called Service A. 98% of traffic would go to the production version, 1% to the control version, and 1% to the experimental version where latency is injected between Service A and downstream services using an injector. This allows experimenting with how systems react to different latency conditions to test resilience and identify problems.
Chaos engineering involves intentionally introducing failures into systems to test resilience. It reveals issues that may otherwise remain undetected. Some key aspects of chaos engineering include using tools like Chaos Toolkit and Gremlin to fail components, running experiments at specified times using a configuration file, and introducing faults into service meshes between microservices. The goal is to build confidence in systems by identifying weaknesses before real failures occur.
This document summarizes a presentation about using machine learning and cloud computing for geospatial services. It discusses Amazon SageMaker as a fully managed deep learning service, case studies of companies like Digital Globe and Development Seed using it for geospatial machine learning tasks. It also outlines how Amazon uses geospatial AI for applications like fulfillment and delivery drones. Finally, it promotes the Earth on AWS and research credits program for geospatial data and computing resources.
How to Measure DevRel's Perfomances: From Community to Business - Channy Yun ...Channy Yun
?
Developer relations are an impactable to generate business values in many software companies who hope to gain mindshare of developers in various approaches from contributing open sources to gaining meaningful sales leads. In this session, you¡¯ll learn about how to measure the perfomrmance of developer relations for building community, increasing impacts and generating leads for sales.
https://tokyo-2018.devrel.net/speakers/yun/
This document summarizes several companies' approaches to microservices architecture. It describes how companies like Twitter, Gilt, and Hailo implement microservices for configuration, tooling, discovery, routing, and observability. It also notes that the microservices ecosystem is rapidly evolving, with many choices for development and operational tools, orchestration, and datastores. Next-generation applications may assemble components from a Docker Hub "app store" and leverage ephemeral, orchestrated, or database-as-a-service solutions.
Mozilla Firefox OS, its Technical Platform and Future - ISET 2014Channy Yun
?
The document discusses Mozilla Firefox OS, an open source mobile operating system based on web technologies, outlines its technical structure including the Gonk, Gecko and Gaia components, and details Firefox OS's developer platform and future including expanding to new device types and emerging mobile markets with a focus on low-cost hardware.
Webware - from Document to Operating System Channy Yun
?
This document discusses the past, present and future of web technologies. It covers early web documents from the 1990s that used technologies like blink and marquee. It also discusses the browser wars of that time and the standards movement. The document then covers modern web technologies like HTML5, CSS3, Canvas, WebGL and APIs for multimedia, geolocation, offline apps and more. It discusses the evolution of the web to a runtime for applications rather than just documents. Finally, it discusses potential futures for the web like browser-based operating systems and the web as a business platform.
7. It¡¯s ¡°impossible to avoid¡±.
By 2011, 80% of all commercial software
will contain open source code.
Open source impossible to avoid, Gartner says¡±, Network World
http://www.networkworld.com/news/2007/092007-open-source-
unavoidable.html
13. Benefits for students
Why do you join Open Source?
? Intellectually stimulating : 44.9%
? Personal Skill-up : 41.3%
? Supporting to be public source code : 33.1%
? Non-work functionality : 29.7%
? Work functionality : 33.8%
? Pleasure on community : 20.3%
? Improving career : 17.5%
? Reputation : 11.0%
? Hating commerce software : 11.1%
Reference: Boston Consulting Group (2002)
http://www.bcg.com/publications/publication_view.jsp?pubID=935&language=English
28. Linux 2.0.26
? 50% of the changes where made by 2.5%
of the developers
? 50% of the changes were made by 97.5%
of the developers
Who Wrote 2.6.20? http://lwn.net/Articles/222773/ by corbet
33. ??? Github http://github.com
Open Source Developer¡¯s Social Networks
34. Project management
Code review, Pull request
Code hosting
Online editing, annotation, IDE integration
Community
Developer Karma system
Social graph
52. The Story of Mozilla
?? ? ???! http://www.mozilla.or.kr
http://www.youtube.com/watch?v=YibglpsnfNQ
http://www.youtube.com/watch?v=wjHQ7NRGQL4
http://www.youtube.com/watch?v=AQuOhrXINiM
53. Thanks for Listening!
Q&A
@mozillakr
@channyun
channy@mozilla.or.kr
http://forums.mozilla.or.kr
http://facebook.com/mozillakr