Welcome to the FOSSASIA Summit 2017 at the Science Centre Singapore.

Can a future of Singularity as articulated by Ray Kurtzwild and also the gloom and doom as noted by Stephen Hawkings indeed be a cataclysmic end to humanity - just as the meteoric accident that killed off dinosaurs? I will paint a picture of hope and one of collaboration. There are plenty of rules we can articulate and follow - including the 'new overlords'.

Artificial Intelligence and Science

How Singapore Government’s central information technology systems and infrastructure, drives the development and delivery of innovative public services for citizens and businesses.

Are you looking for ways to get involved with Open Source in the Cloud? Here's your chance. In this talk, I will briefly talk about Google Cloud Open Source products and related projects, including machine learning, systems, client side libraries and data analytics tools. I'll explain how you can get involved, contribute and share some tips for code reviews.

AI Assistants are able to augment the human intellect, to perform task and offer services, to have capabilities beyond the ordinary power of humans. With open AI Assistants a community can create such tools for their own need. We show that the creation of such tools can be as easy as writing a wiki article. With the proper Cognitive Assistant Technology we will be able to augment human intellect in the same way as it is possible to document the world knowledge with the wikipedia.

Natural Language Processing (NLP) is the task of the computationally understanding and production of human languages, often using a mix of hand-crafted rules and machine learning techniques. The recent hype in artificial intelligence and deep learning has brought a tsunami of media coverage, governmental interest and public enthusiast. And NLP rode the wave and pushed the computational processing of language to the current frontier. In this talk, we will go through the various NLP open source available and share my experience being a contributor to Natural Language Tool Kit (NLTK) and Moses machine translation suite.

In the recent years, EEG brain-computer interfaces are becoming popular and more accessible to consumers. Applications range from controlling computers, machines, biofeedback, to quantified self tracking. At first sight, the current generation of commercial devices seem to be decent in their functionality, and various use cases are suggested. However, neurophysiological signal quality, as well as limitations of software and hardware hackability are among the greatest issues and hurdles towards advancement in user experience. This is why we started to work on Brain-Duino, an open-source brainwave amplifier shield for the Arduino and other microcontrollers.

Brain-Duino is a high quality, low noise and affordable EEG / BCI for hackers, makers, researchers, artists and other enthusiasts. This talk will be about how the Brain-Duino project started, where it is now and where it is going. We will briefly explain how EEG brain-computer interfaces work, and also why this topic is relevant, together with the current advances in artificial intelligence and hardware.

During the FOSSASIA you will have the chance to try out the Brain-Duino and visualize your brainwaves.

As society and technology become more deeply intertwined, society is taking interest in -- and influencing -- nuanced technological issues, ranging from net neutrality, to the selection of emojis, to the enforcement of open source licenses. The FOSS community takes for granted the values of open source, but the benefits can be difficult to explain to those who are not empowered to utilize the source. This talk examines some of the cultural and technological barriers that have stymied inclusiveness, using gender imbalance as a case study. Finally, we propose a solution which we call "Love to Code", which attempts to address the issue of inclusiveness in technology.

Andrey will talk about Microsofts Open Source strategy. He will put a particular focus on scaling up open source workloads on Microsoft Azure cloud platform.

FOSSASIA - Our next steps!

Low Earth Orbit (LEO) satellites cross the sky fast enough that Doppler-shift often has to be corrected for in order to sustain communication. In this talk I describe my efforts to make use of the Novena's Field-Programmable Radio-Frequency (FPRF) and Gate Array (FPGA) components in combination with various open source Software Defined Radio (SDR) front-ends to receive UHF downlinks and to perform phase-preserving Doppler-shift correction in real-time.

As a tradition of MySQL Community meetup, we always have a platform where our community will be able to discuss everything about MySQL with our developers.

Please join us for a Group Photo of the FOSSASIA Summit 2017!http://2017.fossasia.org/

This talk introduces the Classical Language Toolkit (CLTK), a project bringing natural language processing to ancient languages. The talk focuses on two things, namely the CLTK core software and CLTK Archive, an interactive reading environment.The CLTK core is a Python library that offers natural language processing support for Classical languages (namely, those of ancient Eurasia, from earliest civilizations through the beginning of the modern era). It can be seen as an extension of NLTK, but with special consideration for the unique needs of historical languages. It provides basic functions for many languages (e.g., n-grams, word frequencies, word tokenization) and wider support for several languages (Greek and Latin) that includes text normalization, sentence tokenizer, POS tagging, and named entity recognition (NER). We plan to extend to other languages, as well.The CLTK Archive is a website I worked on during my GSoC project. It's a web app front-end to the CLTK core and also provides a modern reading environment for interacting with ancient texts. The Archive aims to provide an aesthetically-pleasing yet utilitarian environment that facilitates reading, note taking, and learning. It presents Classical texts in an easy-to-read layout with metadata like translation, commentary, and word definitions. The site is backed up by a Flask based REST API that provides access to text corpora and other functions of the core CLTK package. I will also be giving a short demos of the capabilities of the Python library and CLTK Archive website (available soon as an alpha release).

Experimenting with self-built equipmentThe "open science" movement stimulated people to understand and get closer to scientific questions that can affect their life or the environment surrounding them. This triggered the need of cheap but sophisticated equipment for everybody. The "Rock your Rainbow" project builds an open source spectrophotometer that is affordable and easy to reproduce.With this tool everybody can approach spectrophotometry:It is possible to understand what a spectrum is, which kind of light lightbulbs do emit, and how it can interact with our circadian cycle.Another interesting field is plant pigment analysis. With easy to reproduce protocols it is possible to extract pigments from leaves, fruit or vegetables, isolate them and analyze them with the self-built device.

Join us for a "top-secret" workshop and help to re-establish contact with the Mars mission crew to bring the mission back on track with your knowledge of NodeJS and the Open Source Bot Framework Emulator. Receive a bounty to solve the challenge and path the way for the future of humanity on Mars.

BRING THE MARS MISSION BACK ON TRACK 

Commissioned under the top-secret space project, our first human team had set foot months ago. This mission on the red planet begun with the quest to establish civilization by creating our first outpost on an extraterrestrial body. Not so long ago, the mission control lost contact with the crew, and we are gathering the best of mankind to help save this mission.

In this rescue mission, you will learn to create a bot using an open source framework and tools. You will be given access to our code repositories and other technical resources. We have 3 mission and 2 code challenge to solve in order to bring the Mars mission back on track.

We need you! Be the first to crack the problems and rescue the compromised mission! Your bounty awaits! Receive your mission briefing at the control centre after checking-in at the FOSSASIA Summit!

HOW TO ENTER THE CHALLENGE

1) Join us on March 18 at FOSSASIA Summit (Science Centre Singapore), Tinker Lab (Hall E) at the following timeslots:

* 9:30 * 11:30 * 13:30

2) Bring your own PC or load one from the mission control. We provide internet access at the lab room.

3) Fill up the registration form and check in with the form at the Mission Control.

4) Mission briefing will be provided, you will be given access to the github where you mission resources will be provided, and you can proceed to crack the challenges.

5) Badge of honors to be earned and bounty awaits the team with the best-time!

6) Winners to be announced at 17:30! Be there!

INSTALLATIONS NEEDED

1) NodeJS (https://nodejs.org/en/)

2) Any Code Editor (Visual Studio Code/Atom/Sublime Text etc.)

3) Open Source Bot Framework Emulator (https://emulator.botframework.com)

A lot of scientific equipment produces experimental data in proprietary data formats. We think everybody should have the possibility to analyse data from scientific experiments hence the data formats need to be open source'd. OpenChrom is a processing/analysis software that contains data import routines for virtually all Gas Chromatography / Mass Spectrometry (GC/MS) systems available and in-use. GC/MS is today one of the most widely used chemical analysis techniques in the whole world.

We would talk about two major open source initiatives - Google Summer of Code and Outreachy.

Urvika would share her experience with "Outreachy"- Outreachy is a similar program like GSoC but it helps people from under-represented groups such as females and LGBTQ, to get involved in open source software. She would like to share her journey of getting into Outreachy in the hopes of inspiring budding female developers like her.

She would talk about how  being on Outreachy team changed her persepective from “how to code" to that of “how to contribute".

Pranav would be sharing his working and learning experience during Google Summer of Code’16 with Debian.

Both Urvika and Pranav worked on the same project - Lumicall. They would talk about “Lumicall” which is a free and convenient app for encrypted phone calls from Android. It uses the SIP protocol to interoperate with other apps and corporate telephone systems. We would be explaining the work we did using Android and Java to add new features into the existing application and how to pace the learning curve.

Urvika would also give a brief about a few subjects that she worked on using Android/Java

1. Explaining what “WhiteLabelling” is and how can we implement it in Android

2. Working with productFlavors, Preferences, AudioManager, RingtoneManager, Date , SimpleDateFormat, Calendar classes in Android.

3. Pub/Sub Support

Pranav would explain about adding PUBLISH Support to Lumicall and developing a library for adding quick enroll option to Android Applications.

Debian is the Linux distribution that is composed entirely of free and open-source software and it is by far the largest Linux distribution available. The development is carried out over the Internet by a team of volunteers guided by three foundational documents: the Debian Social Contract, the Debian Constitution, and the Debian Free Software Guidelines, documents that in themselves have had a lasting influence on the free software world.With this presentation the phenomenon of the Debian project is explained, from its beginning to the current time, both on the technical side as on the organisational side. Debian has a lot more to offer than one can see on first glance, so it also shows why users should care and which advantages can be received from using Debian.Being a Debian developer for more than 20 years, the presenter will take the audience back in time and show how Debian got going and how it developed to become what it is nowadays.

It is impressive how much time and resources a team can save by using the OBS to manages their packages creation and distribution. OBS is a generic system to build and distribute packages from sources in an automatic, consistent and reproducible way. Andrew Lee will cover the benefits of using OBS, explain some of it features and workflow for all your packaging and releasing needs, like automatically build package from scratch on multiple target distros and architectures, easy access through QA to the developer's repo to generate new images with the changes for testing before integration into the production repo, vcs-like workflow as branch code, send merge requests and review submissions and flexible to connect additional resources to empower the backend worker(builders) even with different architectures. At the end tips on how to setup and optimize OBS will be provided.

Learn how we are bringing Android to Chrome OS devices to let users do more. This talk will cover the growing diversity of Chrome OS hardware and best practices for Android Apps on Chromebooks.

Lightning and Ignite Talks. Please check out the latest updates about lightning talks on the wiki: https://github.com/fossasia/2017.fossasia.org/wiki/Lightning-Talks 

Unconference track

LearnBot is a fully open source robot that runs on an ODROID ARM hardware. The outside is made of fully 3D printed pieces, which makes it very easy for anyone to make their own version at home. It runs GNU/Linux along with the RoboComp Framework and can be easily programmed through python components. It is meant for educational purposes, although it can be used for any other research projects. In terms of sensors it comes with a camera and ultrasounds, while regarding actuators it has motors to move around. However, due to its simple design, it is easy to add new peripherals as required.

https://www.youtube.com/watch?v=R0ftHYlYEPk

Scientific research is producing amazing results at high rate. However it is incredibly difficult for non-experts to understand such studies and their implications. The open source technology movement simplifies those concepts and allows the understanding of these works. This enables citizen to answer scientific questions affecting directly their lives. The "Rock your Rainbow" project aims to disassemble already existing open source spectrometers, and re-design the modules to be flexible and compatible with custom built devices. With this tool it is possible to approach spectrometry: simple examples are the analysis of light bulbs emission spectra and its influence on our psycophysical health, or the classification of plant pigments.

Getting a LED to blink is easy, making a strip of 60 LED produce a pattern that is not an annoying is hard. Lets have a look at the trials and tribulations and the patterns I came across when designing lights for my use at home. There will be blinking in all colors

• What our software does and how it looks like.
• The FOSS software stack we uses to build the planting room control system: Python, Django, PostgreSQL etc
• Some challenge we faced (network issue, database issue) in our special condition, and how we overcame it.

After I came home from a recent trip that happened to occur during the middle of a heat wave. I arrived to find my apartment quite too hot, it was at least 45C inside. Needless to say it wasn’t the most comfortable way to come home after 15 days out of town, I decided it was time for me to do something aboutit to address this so I didn't come home to that unpleasant surprise again.In most houses this problem is avoided by having a thermostat which controls the air conditioning. However, the problem for me was that my apartment didn'thave a thermostat for the air conditioning. So I decided to build one using open source software. This talk will cover a how I went about solving my problem using existing software and protocols like home-assistant, MQTT, and also some new software that was created for this. It'll also go over how using software and home automation I was able to not only solve my heating issue but also make cooling my apartment smarter.

Continuous Integration (CI) is crucial in modern software development. Jenkins is an open source software for this purpose. A lot of teams involve Mr. Jenkins and hope for a better quality software. But do they know that they are killing Mr. Jenkins? My talk will answer this question and introduce 3 levels of good practices to maintain a healthy, strong and long living Mr. Jenkins. One of them is a simple and cool hardware designed by ourselves. #jenkins #esp8266 #hey!2.0 #3dprinter #lasercutmachine

The Moore's law has almost come to an end and not much performance up gradation can be achieved just by upgrading hardware generations. The methods of frequency scaling and core multiplication have reached their maximum limits of exploitation and cannot be explored further. This gives rise to the need for heterogeneous computing.Heterogeneous computing refers to systems that use more than one kind of processor or cores. These systems gain performance or energy efficiency not just by adding the same type of processors, but by adding dissimilar co-processors, usually incorporating specialized processing capabilities to handle particular tasks.There are a number of heterogeneous devices that are readily available off the shelf, some of them as commercially available system on chips (SoCs) and can be easily accessible to the average embedded software programmer. These heterogeneous devices incorporate hardware accelerators, such as GPUs and FPGAs. The advantages of having multiple types of hardware accelerators within a system on chip are many, such as high-performance gains and specificity of tasks performed. One of the major issues is the exotic programming model that each hardware accelerator brings along with it. FPGAs bring along the requirement to write accelerators using hardware description language (HDL) at a very low level of abstraction (register transfer level) which has a really steep learning curve, extremely long compilation/place and route timings and are relatively difficult to debug. GPUs have their own programming languages and until a few years ago GPUs were used predominantly for graphics processing as the name suggests. However off late, with computing problems of large sizes and datasets, a lot of computations are being off-loaded to the GPU due to its benefit in parallel execution. Even with a great number of devices available, the difference in programming models makes it difficult to truly use all these devices in unison. Even with these multitudes of methods available for hardware acceleration, it is still not possible for software developers to adopt and use these devices as part of their daily routine for hardware acceleration. In order to achieve the full potential of hardware acceleration, multiple heterogeneous devices within the same SoC must be programmable using a single programming model and also the programming model must be portable across devices without loss of functionality. The aim of this talk is to throw light on the OpenCL(Open Computing Language API) and its potential for being the one single language that can be used to program multiple devices.

Making something is easy; making something that can be made is hard – very hard. As the cofounder of Highway1, a hardware accelerator, I've seen many dead bodies in the early stage hardware space. Hardware takes longer, costs a lot and requires more people. It can be perilous for a company, but the rewards can be great. This talk outlines some of the funding and product development lessons that I've learned with my startups.

In the present world of IoT and the rise of smart city; people are unaware of the persistent and unpredictable threat that may encounter with their daily lives. This short talk will enlighten them to be cautious and equipped with knowledge on how to evade Social Engineering attack, online and offline.

Git became really successful in the Open Source world and, more recently, in the corporate world as well. However, Git is not suited for large files by design which is a major obstacle for many adopters. GitLFS is an extension to Git which addresses this problem. In this talk I will explain how GitLFS works and how Autodesk uses it at scale for 4000+ engineers and 1000+ repositories. You will also learn how to use it properly and how to avoid the most common pitfalls. If you have worked with Git before then this talk will prepare you to handle large files without trouble in Git!

Since it was created in 2005 by Linus Torvalds, Git has become the new standard in source code and version control. With tools like GitLab and GitHub, it opened up a new era of collaborative development. Through some of its features, let's see how it can have a great impact on software quality and agility.

The OpenStack Community's CI System operates at a tremendous scale, with on average roughly 12 thousand test jobs launched every day. It's exciting to see testing happen at such a large scale. However there is an issue, it is too difficult for humans to deal with all the data being generated or analyze all the results from the testing. To deal with this the OpenStack community has developed and deployed several tools to help deal with the torrent of data. Using and building off these tools have been invaluable especially as the OpenStack project continues to grow. This talk will explain the basics of OpenStack's CI infra and cover the approach and tooling used by the OpenStack community to interact with the large amount of test result. It will explain the techniques used and the benefits derived from having both open test result data and performing analysis on that data to understand the project.

The primary language of most open source projects is English, but open source users and contributors span the globe. Non-native speakers face many communication challenges when participating in the ecosystem. In this talk, as non-native English speakers and contributors to OpenStack projects and its community, we will share our experiences and some best practices for coping as and including non-native speakers in your project.This talk will include:* Common issues for non-native English speakers* Tips for avoiding and resolving these issues* Suggestions for effective communication

Japanese enterprise companies have begun introducing HTML5 for their backbone system. However Evolution of HTML5 and JavaScript are really fast and quickly change its trends. They think, want to use long-term trusted architecture. So we have developed oss named hifive. It's HTML5 framework for enterprise companies. I will introduce current status of Japan's enterprise market and what do they need.

ManageIQ is an open-source hybrid IT management platform that allows you to discover, optimise and control different cloud, IaaS and container environments with a unified UI. We'll go through a brief history of the project, introduce the community behind it, demo some of its features, and discuss how to contribute and get involved!

Managing your Data Center and Cloud/Infra resources can be a complex and challenging task.ManageIQ, the leading Open Source hybrid cloud management platform presents the operator a consistent view of the data from his cluster, helps planning future scaling, handle unexpected events and proactively identify problems and security issues. One of the most powerful features of ManageIQ is its ability to automate the orchestration of workloads and resource. In this session, I will present recent Implementation of dedicated ManageIQ Ansible modules. Using these modules simplifies bringing your Cloud and Containers into ManageIQ, and leverage its management and orchestration abilities. I will also cover using Ansible from within ManageIQ, utilizing its Automation Engine. After a short demo you’ll know how to: add a Cloud and Containers providers to ManageIQ, manage users, policies and alerts, all through Ansible modules. The demonstration will also include writing and launching ansible playbooks from ManageIQ.

There's a lot of magic in Python -- metaclasses, asyncio, decorators -- you name it! In this talk I'd be covering a short introduction to these and also how and when to use them. I'd cover (a) metaclasses in Python 2 and 3, differences and dynamic behaviours that can be leveraged; (b) decorators; and (c) Mocking and Patching to dynamically control the way the code behaves.

This talk is about my work done during the GSoC coding period (May 23 – Aug 15). This project is an extension of an existing project named cyvlfeat.Brief introduction: We intend for this to be a light wrapper around the VLFeat toolbox. cyvlfeat will provide a mixture of pure Python and Cython code that looks to replicate the existing MATLAB toolbox. The role of Cython in this regard is to re-implement MATLAB mexfiles.The VLFeat open source library implements popular computer vision algorithms specializing in image understanding as well as local features extraction and matching. Algorithms include Fisher Vector encodings, VLAD, SIFT, MSER, k-means, hierarchical k-means, agglomerative information bottleneck, SLIC superpixels, quickshift superpixels, large scale SVM training, and many others. It is written in C for efficiency and compatibility, with interfaces in MATLAB for ease of use. The corresponding documentation is detailed throughout both parts. It supports Windows, Mac OS X, and Linux. The latest version of VLFeat is 0.9.20.Blog: https://gsoc2016.wordpress.com/

Go (often referred to as golang) is an open source programming language created at Google. Unlike Python which is an interpreted language, Go is compiled. A few projects have moved from using Python to Go lang lately. This talk summarizes observations made when a Python programmer learns Go.

Python programming language has the potential to change a dull teaching learning process in to more stimulating active learning. Equations in physics and maths always look dull and boring. With the power of python one can see the beauty behind those equations and can find real world connections. With the effective use of python programs a teacher can actively engage the students and help them visualize the concepts and have better understanding.In this presentation I will share my teaching and learning experiments with Python in Science and Maths with live demo of some experiments .Join me to enjoy the memorizing patterns and mash-ups with Light, Sound, Physics and Python!

In this workshop we will learn about MicroPython. The flow of the workshop:

* Super quick review of Python

* Introduction to MicroPython

* Update our NodeMCU devices with MicroPython firmware

* Using sensors with NodeMCU* Basics of MQTT and using it for our first IoT device

The Prosthetic arm project is about building prosthetic arms that are functional, can be controlled by the mind directly or indirectly, and that can provide a tactile feeling by sending signals back to the brain. The World Health Organization estimates that there are 40 million amputees throughout the developing world, but only about 5% of them have access to any form of prosthetic devices or assistance. The average prosthetic arm can cost between 3000 and 30000 dollars and many of these devices are not functional and are purely cosmetic. Thus I am trying to develop low cost prosthetic devices that are effective and are low in cost.

The project deals with developing various mobile applications (Apps) using free and open-source web application for hearing impaired children. The work has been done in association with a Non-Governmental Organization (NGO) called SAHI, Society to Aid the Hearing Impaired which is part of Apollo Hospitals, Hyderabad. SAHI supports the hearing impaired kids for Cochlear implant surgery so that the kids audibility increases. After the surgery audio therapy is given to kids for better understanding of the surrounding world. By using free and open source web applications, MIT App Inventor, various mobile Apps have been developed for the better learning of the kids. The learning of the impaired kids have been cleared explained by the special teachers to the developers. This made to develop Apps more clearly, concisely and perfectly. Different Apps were done keeping in the view of international standard of teaching to special kids who are hearing impaired. The Apps developed covered variety of topics such as Early Vocabulary,Vehicle and Voice Sounds, Syllable Length Word (in English and local languages - Hindi and Telugu), Discriminate Consonant, Same Length Words (in English and local languages - Hindi and Telugu), Leaning by Manner, Learning of Colors, Learning various Poems (in English and local languages - Hindi and Telugu) and explaining short Stories(in English and local languages - Hindi and Telugu). The Apps have been done in both English and native languages too keeping in view of the therapy requested by the kid's parents. Tests are also included in the Apps to know the level of learning. During initial stages, suggestions were given by audiologists, special teachers and counselors. After continuous and rigorous reviews and feedbacks, the apps developed had become satisfactory to all the special trainers and was ready to use. They were given to the challenged kids for testing as a part of their learning. The results found by the teachers were astounding. Children are showing more interest to learn on mobile rather than book. The duration of attention towards learning of the kid has drastically improved. Teachers training the special kids are very happy with this m-learning (Mobile learning). Parents too found it to be amazing as their kids show more interest towards learning through mobile. Also, now they are very much comfortable to train the kids at their homes by using these Apps. The parents are very excited to get all these learning Apps shared to their mobiles. It is a great feeling of satisfaction to see kids interested to learn and the parents with bright smile on their face. This all happened by developing learning Apps of open-source software. Over all it's an awesome experience !!

Build using the latest open technology to address critical user needs on the open web. A look into how Accelerated Mobile Pages (AMP), Progressive Web Apps (PWA) and beyond is shaping the future and truly creating a best in class mobile experience

In this workshop Frank will explain how to install and setup an Nextcloud server. Basic knowledge of Linux and webservers is recommended. Nextcloud can be installed on a small Raspberry Pi like device or on a bigger Linux system. Another topic of this workshop will be how to start collaborating with other users using the Nextcloud federation feature

Working from home (or a cafe)? Working with geographically distributed teams? This is the new normal.Tools come and go, yet you'll always be thinking about the base - collaboration, interaction, decision making, conflict resolution, and most importantly communication. For people with an open source software development background, much of this might come naturally; mixed mode environments will require some tweaking.Having never worked in an office my entire career (consulting, MySQL, Sun Microsystems, MariaDB Corporation, Percona), here are some lessons that I've managed to learn.

We provide updates on our OSS project, Legalese.com, which is developing a programming language to allow programmers (and eventually robots/IoT devices) to draft simple legal agreements using a domain-specific-language.

This presentation shows why the software created by OSS communities is so important for business users. It gives some motivations for the immense growth in enterprise space OSS experienced in recent years. With OSS not only fulfilling the technical requirements, but also the operations requirements it quickly became a strategical choice for a lot of companies, be it for internal IT or as the basis of their own products

// Micro AI - AI has been one of the hottest technology trends leading up to 2017 - AlphaGo beating Lee Se Do (March) - Massive acquisitions: https://www.cbinsights.com/blog/top-acquirers-ai-startups-ma-timeline/ - FB opened Messenger platform, over 40,000 chat bots developed - Uber, Google, Tesla in self driving cars - Data open sourcing, OpenAI, research datasets- It’s easy to sit back and relax watching all this happening (Image: TV potato couch), but how can entrepreneurs take action (Mine gold action)?- Micro AI product and services from the trenches (Image: call of duty trench fight) - Customer: Identify the right customer (enterprise or consumer or a subset of the offering you give - usual customer development) - KeyReply chat and engagement platform: communication and marketing team for Tech news outlet - TechinAsia - Product: Research, test and learn (product development) - Joke: Buzzword at NIPS - RocketAI - AI Differentiation: Deep research or the novelty, quality and/or quantity of the data they have access to - Healthcare data, financial performance, chat logs with open data sets like Gigaword, Enron email corpus - Share research - Mendeley app (pull in research papers from Arxiv) (Image: App photos) - Started with a demo of the chatbot, Trello board conversational flows, research on the various libraries and tool kits we need to use to deliver the optimal experience - Go to market - Hyping innovation vs expectation management - Promise amazing performance vs determining the right success metric - Charging random pricing vs value based pricing - Team - Know your strengths, build capabilities and optimise for learning - Have everyone be on the same page, read and research - Recruit both in Asia and the US (Singapore as an example - universities, AI research, no. of papers published in EMNLP China vs US vs rest of world)

The talk focuses on how developers and startups (Web / Machine Learning / Data Science / Services) can use Python/Node/Javascript along with the PaaS, IaaS offerings of Microsoft Azure to build applications and easily deploy them in a few clicks to Microsoft Azure. The talk will showcase rapid web development with flask and angular and setup of the project, development and deployment of the web applications to azure websites. It’ll also showcase running background tasks and time based tasks like triggers using Azure webjobs and webservices. This talk will be beneficial to startups and enterprises which aim to build robust and reliable engineering systems and will be using the Azure Application Insights service to showcase how the tool can help with quick fault detections and out of the box features for user demographics and behaviour. Having such telemetry and notification features in place will help find faults as they happen. The talk will be a showcase on how Microsoft does IT for maintaining and running applications with high availability and low errors.The entire talk and demos would be done using open source tools like Python tools for Visual Studio (PTVS), Node tools for Visual Studio (NTVS) while showcasing capabilities like using multiple virtual environments, built in debuggers a powerful IDE. All the SDKs for Application insights are available open source on the Microsoft github.

The aim of the workshop is more to have a sprint on Fedora Infrastructure projects. Fedora Hubs will be a primary project as the project is a budding project under Fedora Project as a platform for collaborators and contributors. We would also have other projects to sprint on like Pagure, Tahrir, Bodhi etc

This talk invites anyone who is interested in the Wikimedia's vision and in learning ways to get involved in our technical community. This talk would particularly be useful for university students, underrepresented minorities such as women in open source who are seeking to contribute to a beginner-friendly and diversity-embracing open source community. The focus would be to lay deeper emphasis on different areas through which one could contribute in our projects, such as design, documentation, research, translation, etc. More here: https://www.mediawiki.org/wiki/How_to_contribute.

We’ll talk about different outreach programs (such as GSOC, Outreachy) that we run every year and few projects (such as Community Wishlist Survey https://meta.wikimedia.org/wiki/2016_Community_Wishlist_Survey) that we promote through these programs. Some of these venues could be a good starting point for contributors to follow and get involved.

I will be giving a introduction of KDE and it's applications, followed by the procedure to apply for SoK (Season of KDE). Then I will be briefing my GSOC project LabPlot and my experience working on it.

Not a lot tends to get said (or done!) about FOSS desktop technologies around these parts. In this talk I'll give an introductory overview of KDE software, including the pieces and architecture that make up the popular Linux desktop environment, as well as the underlying Qt-based KDE development platform on which KDE software is built. I'll also talk about the how's and why's of contributing to KDE, referencing my own experiences as a former KDE developer/contributor.

An introduction to distributed version control systems with Git. We will start from the initialization of the first git repository, and continue with committing, creating branches, merging, rebasing, resolving conflicts, using GUI tools, using remotes, connecting to GitHub/GitLab, and conclude with team working practices. The talk is mostly aimed for beginners to kick start their git-izenship but I'm sure that even experienced git-ers can find something useful.

Amazon Web Services announced a new PostgreSQL-Compatible edition of its Amazon Aurora Database at the re:Invent conference in Las Vegas on November 30, 2016. With Amazon Aurora’s new PostgreSQL support, customers can get several times better performance than the typical PostgreSQL database and take advantage of the scalability, durability, and security capabilities of Amazon Aurora. This session will explain the underlying architecture of the Aurora platform and how it benefits users of PostgreSQL. It will also discuss industry trends on migrations to PostgreSQL and feature a customer's experience of migrating from Microsoft SQL Server to Amazon RDS PostgreSQL.

A lot of pros and cons are regularly dicussed when it comes to Open Source usage in business IT. There are a lot of reasons why Open Source is ideal for business critical system, but there are also a lot of uncertainties and doubts by people who are not really involved with Open Source communities. This is in particular a topic for PostgeSQL, which is the most important Open Source solution for businesses. This presentation evaluates those pros and cons and shows how to get the most out of PostgreSQL's Open Source nature for enterprise IT.

In Distributed System, such as Centralized Retail System like Hypermarket, there are 2 approaches for data collection and integration. Use Webservices or use Distributed Databases. By enabling Distributed Databases, and maintained it as if single integrated system, we can achieve simplicity, no context producer and subscriber anymore. this saves a lot of effort, time and resources of course. There are no worry about data synchronization, no need to maintain data Integrity with much effort. In this talk, we will cover about why and how BDR (Bidirectional Replications) implemted in high performance transaction system (which we developed, i.e POS Retail System). Why that matter and how this can be a great lesson for any other implementation.

Unconventional query optimization in PostgreSQL, or how to teach an elephant to dance rock'n'roll.Query performance is usually tuned by simply adding indexes and rewriting the query minimally, while the basic idea of the query stays the same. For many query, this is not the best way to optimize them: you need to rethink and rewrite them completely. To do this, you need to know how optimizer and executor work, and keep that in mind while composing your queries. In this talk, I will present an overview of a few less known and rarely used techniques such as:* using and implementing loose index scan for different scenarios;* using IOS in situation where it isn't applicable directly;* replacing IN (long list of values) with JOIN VALUES ();* moving a limit or offset clause inside joins;* and other tricks with row types and arrays of row types.Using these techniques and some other ideas, we will learn how to optimize a 'friends feed/subscription news' query, which is a typical task in many projects. We will diretcly compare performance of four different approaches: direct approach (simplest query), optimized direct approach (more complicated version of it), pl/pgsql with different algorithms, and implementation of this pl/plsql code in plain sql. I will show how this approach can convert almost any read-only non-dynamic (no EXECUTE statements and no exceptions) pl/pgsql code to a plain sql query.

If we look back and see the changes that have happened in last few years, we will realize that how things have changed very rapidly. A decade back, most of the organizations either would have their in-house database servers running on certain hardware and storage configuration. Over the period, when the servers become obsolete, then server/storage migration would itself become separate project and the procurement process for new hardware in any organization is usually time consuming process. In last two years, there has been rise in adoption of Cloud that tries to address this and many other issues. Now you do not have to worry about the underlying hardware and storage, once you know CPU, RAM and storage requirement, you can customize the server and build your own server in cloud. Also, migrating Postgres cluster to cloud is a viable option for safety (no data loss, including if rollback), minimal downtime (switchover and rollback), minimal effort for the team.Considering one of the above reasons for moving to cloud, one of our client’s in FinTech space decided to move their PostgreSQL instance to AWS RDS since RDS offered some of the following benefits like:• Automated failover of primary in case of failure • Automated creation of replicas• Automated updates• A purely declarative setup, via Cloud formationHere I will talk about how technical aspects of moving PostgreSQL server to AWS RDS.• How to configure, tune the parameters in config file while moving onto RDS. • How you can select the storage depending upon growth estimate and IOPS requirement• How you can configure HA setup in RDS.• How you can distribute your reporting workload.• How to monitor Postgres instance and alert configuration

In recent years, applicable area of PostgreSQL has quickly extended into the enterprise sector thanks to continuing effort of the community to improve performance and functionality. As a result, there is an emerging demand to use PostgreSQL in more security-critical circumstances.In this presentation, I will talk about the following two topics. * Considerations for securing a database system. * Current status of database audit on PostgreSQLI am working for the open source software center (OSS center) of NTT which is the largest telecommunications company group in Japan. We have encouraged many of our customers to migrate a lot of database systems to PostgreSQL so far and this contributed much to cost reduction.Some projects need to conform to security standards, for example PCI DSS (Payment Card Industry Data Security Standard). It is one of the most prevailing security standard in the world. However it is not easy to build and operate a PostgreSQL-based system conformant to these security standards. I'd like to describe some aspects required for secure database systems in general, such as encryption, key management, identity management and auditing.Then I explain considerable points for building a secure database system using PostgreSQL, and show the remaining challenges for secure database systems using PostgreSQL.Finally, I introduce a forked version of pgaudit that we are maintaining, then explain how to use it. pgaudit is developed by 2ndquadrant and Crunchy Data, especially by David Steel with a great contribution. However, it does not meet our customer's requirements. For example, It cannot output the audit log and server log separately, it cannot audit Superuser fully, etc. So we forked it and added some changes.

Clonezilla is a free, open source partition and disk imaging/cloning program similar to Acronis True Image®. It saves and restores only used blocks in hard drive. Two types of Clonezilla are available, Clonezilla live and Clonezilla SE (Server Edition). It could be used to Image or clone GNU/Linux, MS windows, Mac OS, FreeBSD, NetBSD, OpenBSD, and Minix. In this talk we will speak about how to deal with backup with Clonezilla.

At FOSSAsia 2015 I gave a short talk on using Puppet and Ansible to manage the configuration of desktops. In the past two years we've gained a lot more insight into the complexities of managing a fleet of machines that aren't always powered on.

Introduce linux-sunxi community. This is community behind linux kernel and u-boot contributions.I'll talk about current status on contributions and which SOCs are supported on mainline.

In the BIOS time, the firmware was usually obscure and the OS developers had to deal with it by experience. Fortunately, nowadays most of the firmware/hardware vendors share the code from the open source project, edk2, in their own UEFI implementations, so it's much easier to understand the how the firmware works, especially the hardware-independent software stack. What's even better is that edk2 contains a sub-project, OVMF, to provide the firmware for QEMU, the virtual machine, we don't need a bare metal to implement features or debug hardware independent issues anymore. No more expensive flash programmer, just copy the file to update the firmware immediately. This talk will present how to develop and debug OVMF with openSUSE and introduce the new UEFI features in openSUSE Leap 42.2.

Coccinelle is a program matching and transformation tool which provides the language SmPL (Semantic Patch Language) for specifying desired matches and transformations in C code. It has been extensively used for finding bugs and performing source code evolutions in the Linux kernel. Coccinelle is designed around a language for expressing matching and transformation rules in terms of fragments of C code. Coccinelle has been used in the development of almost 4000 patches found in the Linux kernel. More than 40 Coccinelle scripts are distributed with the Linux kernel source code.The talk will highlight the capabilities and limitations of Coccinelle along with introducing various basic features of the SmPL.

This presentation introduces the FreeBSD operating system and the community that produces it.FreeBSD is an advanced computer operating system used to power modern servers, desktops, and embedded platforms. A large community has continually developed it for more than thirty years. Its advanced networking, security, and storage features have made FreeBSD the platform of choice for many of the busiest web sites and most pervasive embedded networking and storage devices.

Speaker will go into how to develop an Linux Industrial I/O subsystem driver for an entirely new sensor, and how userspace HALs can process data with reduce system I/O than a pure "userspace driver". Some of the things that will be covered is iio channel consumers + definition, SW + HW triggers, and ring buffer interface. There will be some in-depth overview of the development API to implement a new driver, or add functionality to an existing iio driver. Also will demo of several sensors using the various functionality of iio, including but not limited to hrtimer sw triggers, and buffered data to userspace HALs.

Have you ever wanted to understand how Qemu and KVM work together ? Wanted to contribute to KVM and/or Qemu but didn't know where to start ? I am here to help! :) This workshop is a deep/fast dive to the KVM and Qemu code base to better understand interaction between them. We will cover the important functions of KVM followed by Qemu. Although, the ecosystem extends way up above Qemu and KVM, we will limit our discussion to the lower layers only.

Nested Virtualization allows a kvm guest to behave as a hypervisor and launch its own guests. Since being introduced, Nested Virtualization has seen feature additions and optimizations, sometimes hardware supported. This talk is focused on x86 nested virtualization; its inner workings and optimizations that make it usable. We will take a look at some current usecases of Nested Virtualization and ponder over the future of this interesting technology with respect to adoption and uses.

Finally in 2017 Wayland is shipping in Linux distributions. I'ts already in use on some Embedded devices and that's growing. Why Wayland? Why drop X11? What's better? Come and find out. See Wayland working on tiny and big devices.

Plymouth is a bootsplash for Linux supporting animations using Direct Rendering Manager (DRM) and KMS driver. It gets packed into the initrd. Plymouth's theme is a difficult task, and will display every time when  system boot with a wonderful animation.

A lot has happened in TensorFlow since its open-source release a year ago.  TensorFlow is an open source system for machine intelligence. It's Google's second-generation deep learning system, flexible enough for research yet performant enough for production. In this talk we provide an overview of what TensorFlow is and how it works, a summary of the new developments in the library including high-speed compilation, better ecosystem support, and easier to use high-level libraries.  We also discuss how the open source community has shaped its development and success.

What if all open source software can be treated as a dataset? In this session we will discuss what can be derived form such a dataset using modern machine learning tools such as DeepLearning, clustering, etc.

The Microsoft Cognitive Toolkit trains and evaluates deep learning algorithms, scaling efficiently in a range of environments—from a CPU, to GPUs, to multiple machines—while maintaining accuracy.In this talk, we'll discuss the basics of the toolkit, common use-cases, and go through a quick demo.Landing Page: microsoft.com/en-us/research/product/cognitive-toolkitSource Code: github.com/microsoft/cntk

AI Assistant Development for everyone! We do a colaborative creation of Susi Skills which is as easy as writing a wikipedia article. You don’t need to be a programmer to create an artificial intelligence which provides knowledge and IoT steering abilities.

The most popular ecosystem for Machine Learning and Data Science is the Free and Open Source Statistical Language  R. Over the past 20 years R has grown from a specialist tool for a handful of statisticians to an estimated 4 million users. This talk will introduce the R ecosystem for Machine Leaning. We will learn how Decision Trees are built through the accessible Rattle graphical user interface and showcase some of today’s developments in Big Data deploying Artificial Intelligence and Machine Learning technologies based on open source software.

Open source game playing AIs have come a long way since the first closedsourced programs defeated the best human players. Stockfish, an open sourceChess AI, is currently the top ranked chess program. In the game of Go, opensource programs efforts are underway to replicate the results published by theDeepMind team on the AlphaGo AI. This talk introduces the algorithmsemployed by game playing AIs and the various open source projects youcan participate in.

We present a novel approach to intelligent computer assisted language learning (iCALL) systems, using deep syntactic parsers and semantic based machine translation in the diagnosing and providing explicit feedback on users’ language errors, using Mandarin Chinese (L2) as a case study. In this presentation we describe the proposed approach and current progress of a recently funded two year project. During this project, we are developing a proof of concept system showing how semantic based machine translation can, in conjunction with robust computational grammars, be used to interact with students, better understand their language errors, and help students correct their grammar through a series of useful feedback messages and guided language drills.Ultimately, we aim to prove the viability of a new integrated rule-based MT approach to disambiguate students’ intended meaning in computer-assisted language teaching. This is a necessary step to provide accurate coaching on how to correct ungrammatical input, and it will allow us to overcome the current bottleneck in Computer Assisted Language Learning (CALL) – an exponential burst of ambiguity caused by ambiguous lexical items.

The Translational Molecular Imaging Lab at the Stanford Canary Center for Cancer Early Detection applies machine learning algorithms to extensive collections of ultrasound data to improve pancreatic and breast cancer diagnostics. I will present an overview of the MeVisLab software that we use to analyze and segment 3D ultrasound images, and then dive into the machine learning techniques that are used to improve spatial and temporal resolution for the detection of molecularly targeted ultrasound contrast agents. These techniques, applied to big data and combined with in vitro blood testing, have significantly improved the accuracy of cancer diagnostics, and present the future of early cancer detection. As a concluding remark, I will discuss the potential of open-sourcing medical imaging data and the benefits it will bring to the medical field as a whole, telling my story of how I originally got involved with an open source project called CancerBase.org and how that developed into an opportunity to work at a professional lab.

Since the last FOSSASIA, the Deep Learning Workshop repo (on mdda's GitHub) has been extended substantially. Depending on the time allotted, we'll be able to tackle 1 or 2 'cutting edge' topics in Deep Learning. Participants will be able to install the working examples on their own machines, and tweak and extend them for themselves. Like last year, the Virtual Box Appliances will be distributed on USB drives : The set-up has been proven to work well. Since this is hands-on, some familiarity with installing, running and playing with software will be assumed. Depending on demand, I can also do a quick intro about Deep Learning itself, though that would be pretty well-trodden ground that people who are interested would have seen several times before.

*  Participants should bring their laptop AND have Oracle's VirtualBox already installed and working.

Loklak is an amazing project which scrapes data from various social networking sites, especially twitter. Twitter Wall is one interesting project to engage your social media audiences at your event. Any tweets related to an event will pop up on our wall within seconds of the actual tweet. We built a social twitter wall for our college tech fest, using various API's provided by loklak. Majorly used the search API to extract the data based on the hashtags. In my talk, I'm going to demonstrate my project along with quick steps to implement your own twitter wall.

A twitter wall for “Class Representative elections” was built efficiently using the Loklak Search API, which acts as a ‘web app’. At the Front-End a beautiful User Interface, designed for students to view the nominated candidates list with leading polls (depends on number of tweets on one’s name) while, at the back-end Loklak API is implemented to query the results from Twitter directly (result is declared by calculating number of tweets on that name)

https://docs.google.com/document/d/1A9GU5p6927fs0DU4fMV5u9I_5RZYNZwJ_Gpr02MsI0Y/edit

Loklak is an open source P2P distributed system that scrapes and stores twitter data and provides open non restricted access to tweets. The power of social networks can be immense and can result in building AI systems and machine learning on this data can give interesting insights to data scientists. This workshop will focus on visualizing the data with Kibana, building applications using the loklak SDKs and an introduction to Ask Susi , the AI powered by Loklak.

With the advent of micro-services, dozens of releases per day, logs are the bread and butter for any successful real-time technology platform like OlaCabs. In this talk, I would be presenting our logging pipeline and the challenges we faced while doing it at Ola scale.

Managing server will monitoring all things; disk, network, i/o, CPU, RAM, log, etc. Monitoring that all things will be easy if the number of server is small. We just take manual action, login to the server and watch the status. But, how to do monitoring if the number of server will increasing; dozens, hundreds? We need tools.ELK Stack specially can be use for monitoring log. Centralized log. Centralized logging can be very useful when attempting to identify problems with our servers or applications, as it allows us to search through all of our logs in a single place. It is also useful because it allows us to identify issues that span multiple servers by correlating their logs during a specific time frame.In this workshop, we will go over the installation of ELK (Elasticsearch, Logstash, Kibana) Stack. We will also show how to configure it to gather and visualize the syslogs of systems in a centralized location, using Filebeat. Logstash is an open source tool for collecting, parsing, and storing logs for future use. Kibana is a web interface that can be used to search and view the logs that Logstash has indexed. Both of these tools are based on Elasticsearch, which is used for storing logs.It is possible to use Logstash to gather logs of all types, but we will limit the scope of this tutorial to syslog gathering.The goal of the workshop is to set up Logstash to gather syslogs of multiple servers, and set up Kibana to visualize the gathered logs.

The presentation will include the most recent events and updates regarding PetaBencana.id and the monsoon season in Indonesia. PetaBencana.id is a community-led, open source risk map for Indonesia; the project is supported by the Humanitarian Infrastructures Group of the Urban Risk Lab at MIT, with the aim of further developing CogniCity Open Source Software as a free, transparent, enterprise-level platform for emergency response and disaster management in megacities in South and Southeast Asia. As these cities become increasingly complex systems of people and interconnected infrastructure, extreme weather events and long term environmental changes pose acute challenges for disaster response and humanitarian action.Although rapidly urbanizing environments are often thought to be “data scarce,” by enabling residents to share emergency data openly and in real time, PetaBencana.id makes a wealth of information available that promotes resilience to extreme weather events. By gathering, sorting, and visualizing data, PetaBencana.id transforms the noise of social and digital media into critical information for residents, communities, and government agencies. The platform adopts a “people as sensors” paradigm, where confirmed reports are collected directly from the users at street level in a manner that removes expensive and time-consuming data processing. This framework creates accurate, real-time data which is immediately made available for users and first responders. The CogniCity OSS was designed, tested, and deployed as an operational platform in Jakarta during a three-year (2013-2016) applied research project. As a “proof of platform,” the free web-based flood map PetaJakarta.org was used by hundreds of thousands of residents during monsoon flooding. It is was also adopted by the Jakarta Emergency Management Agency (BPBD DKI Jakarta) to monitor flood events, to improve response times, and to share emergency information with residents. CogniCity OSS proved the value of crowdsourced social and digital media for emergency response through the PetaJakarta.org platform, but this use of open data for humanitarian disaster response is only a first step. Additional CogniCity OSS modules are in development at the MIT Urban Risk Lab to address other hazards, including earthquakes, volcanoes, tsunamis, typhoons, and terrorism. CogniCity OSS is also being designed to integrate and coordinate additional data sources, including various instant messaging services. And, as CogniCity OSS is deployed as a disaster risk management system in other Asian megacities, additional platform extensions and innovations will provide additional support for humanitarian response to urban disasters.

The Shogun Machine learning toolbox provides a wide range of unified and efficient Machine Learning (ML) methods. The toolbox seamlessly allows to easily combine multiple data representations, algorithm classes, and general purpose tools. This enables both rapid prototyping of data pipelines and extensibility in terms of new algorithms. We combine modern software architecture in C++ with both efficient low-level computing backends and cutting edge algorithm implementations to solve large-scale Machine Learning problems (yet) on single machines.One of Shogun's most exciting features is that you can use the toolbox through a unified interface from C++, Python, Octave, R, Java, Lua, C#, Ruby, etc. This not just means that we are independent of trends in computing languages, but it also lets you use Shogun as a vehicle to expose your algorithm to multiple communities. We use SWIG to enable bidirectional communication between C++ and target languages. One of the focus point of the talk is the introduction of the meta-example framework developed as part of the toolbox: a user can write an example use-case for a method using our 'meta-example' syntax, that would then be auto-translated into all of the supported languages of the library. This not only allows us to easily scale with the number of new methods in the sense of having the example use-cases of the method being exposed to all of the target languages, but it allows us to have those methods tested throughout of all the languages.Originally focussing on large-scale kernel methods and bioinformatics, the toolbox saw massive extensions to other fields in recent years. It now offers features that span the whole space of Machine Learning methods, including many classical methods in classification, regression, dimensionality reduction, clustering, but also more advanced algorithm classes such as metric, multi-task, structured output, and online learning, as well as feature hashing, ensemble methods, and optimization, just to name a few. Shogun in addition contains a number of exclusive state-of-the art algorithms such as a wealth of efficient SVM implementations, Multiple Kernel Learning, kernel hypothesis testing, Krylov methods, etc. All algorithms are supported by a collection of general purpose methods for evaluation, parameter tuning, preprocessing, serialisation & I/O, etc; the resulting combinatorial possibilities are huge.

The motive is not to criticize Artificial Intelligence as the past, as the name of the talk suggest. Instead, it's to summarize the advancement made in the recent past, mainly since 2016 Q3. However, the summary may never be exhaustive. So, the focus will be on impact rather than implementation. The talk will give sneak peek into the interesting projects that AI gave to the world through large corporations or individuals. Majority of them are developed using Machine Learning than anything else in AI.

Securing you web application - (testing and Verification)introduction:- Functional testing vs Security testing - Security testing tipsStrategy for securing your web application: Secure at the Source-SDLC (secure software development Lifecycle) -introduce Microsoft open framework of secure development lifecycle Testing Methodology- type of tests black, white and grey boxOpen Web Application Security Project (OWASP) is an online community which creates freely-available articles, methodologies, documentation, tools, and technologies in the field of web application security-OWASP ASVS (Application Security and Verification Standard)-3 main parts of OWASP-19 Security requirements Areas for ASVS 3.0:-OWASP Testing GuideProcessRecon --> mapping --> discovery --> exploitationtesting tool e.g.-Nmap & Zenmap-Burpsuite-Cookies Manager+-etc Verifying identified vulnerabilities by attacking and exploiting them•Go after the data or functionality that real attackers would go after•Successful exploitation is a stepping stone and should open up a new round of mapping and discoverymaybe a demo on injection and cross site scripting