53rd Monthly Technical Session (MTS) Report
Monthly Technical Session (MTS) is HENNGE’s mini conference. As the name implies, MTS is held monthly and its talks are (mostly) about technology.
The 53rd MTS was held on December 14th, 2018.
“What About Drones Anyways?” by Stephen
Drones are unmanned aircrafts or ships. They are increasing in number and it is more and more common to see one in our daily lives.
There are several kinds of drones, such as military, commercial, and personal. Military usage of drones is arguably the biggest driver in advancements of drone technology. Commercial usage of drones is still young but has a huge potential. For example, Amazon released a proof of concept video about using drones for delivery. Personal usage of drones keeps increasing in recent years, especially for photography, journalism, and other such activities.
Drone usage is facing quite a lot of regulatory hurdles. There are concerns regarding the safety of using drones, technical capabilities of drones, usage of drones for cyber terrorism, etc.
“PiterPy 2018” by Jonas and Yuri
PiterPy 2018 was held on November 2nd and 3rd at Saint Petersburg, Russia. About 250 people attended the conference.
There were 20 talks in 2 parallel tracks. All of the talks were available in both English and Russian thanks to simultaneous translation. Some of the talks that Jonas found interesting includes How to Write Deployment-Friendly Applications by Hynek Schlawack and asyncio in Practice: We Did It Wrong by Lynn Root.
“A Gentle Introduction to Computer Vision with Python” by Kelvin
Computers represent images as arrays of values. A black and white image can be represented as a two dimensional array of 0s and 1s. Colorful images can be represented as such an array of RGB values. Transparency can be accounted for by adding alpha values, resulting in an array of tuples of RGB and alpha values.
Computer vision deals with how computers understand images. Kelvin used the support vector machine approach to identify images of numbers. According to his tests, this application he developed is 98.5% precise.
“re:Invent 2018” by Shi Han
AWS re:Invent 2018 was held on November 25–30 in Las Vegas, USA. re:Invent is Amazon Web Services’ global customer and partner conference. People go to re:Invent to learn new things, looking for solutions to specific problems, and to network.
In this talk Shi Han shared some dos and don’ts he learned from his visit to re:Invent 2018. Some of them are:
- Don’t do long layovers
- Do take good care of your health
- Do pay attention to the agenda
- Don’t prioritize Sessions (this refers to the session type called Sessions)
- Do Workshops
- Don’t be intimidated by session levels
- Do walk up to sessions that you were unable to reserve a seat of
- Do prepare a personal AWS account
- Do try the labs (Hands-On and Spotlight)
- Do visit the Expo
- Do visit the Quad
- Do enjoy!
“The Cases of Sales in MTS” by Hinky and Okutani
Hinky and Okutani are members of HENNGE’s Sales Team. They volunteered to talk in this MTS so that other teams know Sales Team better.
They explained quite a lot about their work, including their workflow and some case studies. They shared cases of both successful and unsuccessful sales.
“Creativity by Restriction on the PICO-8” by James
PICO-8 is a fantasy console to make, share, and play tiny games and other computer programs. Programs made for PICO-8 are distributed on .png images that look like cartridges, complete with labels and a fixed 32k data capacity. PICO-8 limitations encourage small but expressive designs.
PICO-8 provides tools for editing code, music, sound, sprites, and maps. PICO-8 cartridges can not only be saved in a special .png format but also exported as standalone HTML5, Windows, Mac, or Linux apps. Any cartridge can be opened in PICO-8, letting anyone peeks inside to study or modify them.
James likes PICO-8 for a variety of reasons. It is restrictive but it includes all the things one needs to create a tiny game. It is easy and fun to see what you’re working on. Its restrictive nature prevents scope creep. It encourages focusing on the core of a game and then adding as many details as possible.
“Solving Equations” by Jason
It’s not always easy to solve mathematical equations by hand. Jason explained the tools he has used to solve equations.
WolframAlpha is an answer engine. It solves queries about mathematics, science & technology, society & culture, and everyday life. Solving equations is but one of its many capabilities.
Jason also explained Python libraries to use for solving equations, such as z3 and angr. z3 is a satisfiability modulo theories (SMT) solver. Jason used it to solve NP-complete problems and Einstein’s Riddle. angr is built on top of z3, and it works specifically for binaries. Jason used it to solve mazes.
“The Evaluation of Language Learning Tools” by Jennifer
Jennifer has been studying Japanese for over 15 years. In this talk she explained the different methods and tools she used during that time.
Back in 2001, the main resources to learn Japanese were books, cassette tapes, and CDs. In 2006 Jennifer started to use electronic dictionaries. Unfortunately they left much to be desired, mostly because they can only look up one kanji at a time, can not be updated, and are expensive. Some years after that Jennifer found a dictionary software for the Nintendo DS. Apparently it was very good for the time, as it allows her to write to look up whole words.
Online resources increased as time goes by. Jennifer mentioned spaced repetition tools (e.g. The Mnemosyne Project and Anki), community sites (e.g. Mixi and Ameba), and language sharing sites (e.g. Lang-8 and HiNative). Obviously there are also numerous language learning tools for smartphones, such as iKnow and Duolingo.
Unfortunately it is still quite hard to become proficient at a new language, despite all the tools that are available today. According to Jennifer this is because most tools only support passive learning, they lack context in teaching words and phrases, they don’t group things to learn together, and they don’t enable cumulative learning.
“Snakify (Part 2): How to Get First Users to Your B2C Product” by Vitaly
Snakify is a service Vitaly developed that teaches Python 3 and web design. This talk is a sequel to his talk in the 52nd MTS.
In the early days of Snakify, Vitaly tried to get its first users by reaching out to people who are interested in learning Python. In time he learned that reaching out to Python teachers is a more effective approach of getting Snakify users. He also started to develop Snakify presence on social media such as Facebook and Twitter. All of these combined to allow Snakify to grow based on positive word of mouth.
Vitaly makes money off of Snakify mostly through ads. He eventually implemented subscriptions as well, but ads remain to be the larger part of the revenue stream.
The day of the 53rd MTS was also the last day of our Global Internship Program (GIP) interns: James, Jason, Jennifer, and Vitaly. We held a small event for them and gave them some souvenirs. In turn, they shared their impressions of working with us.
Thank you very much, friends!
As usual, we had a party afterwards :)
