Notes from 22/4–12/5/19
The last 3 weeks have been absolutely crazy thanks to final exams and projects. Took me almost an entire week to even catch up on the newsletters. But anyway SUTD came up with this ridiculous survey on a new degree they have dubbed “Artificial Intelligence for Design Innovation” which sounds so incredulous I made a quick degree name generator with some help. Now YOU can come up with spicy degree names like “Big Decentralised Computing Entrepreneurship and Design”.
Anyway, time for the content.
An extremely meta case study on case study writing. Enjoyed the read, will use it to sum up some of my projects.
- Define your strategy
- Choose the right candidates
- Run a great interview
- Write up the story
- Put your case studies to work(strategically!)
All told, according to CyberSeek, just over 700,000 people were working cybersecurity for U.S.-based businesses and other organizations in 2018, not including 300,000-plus unfilled positions. The data point in 2017 that had people inside the cybersecurity world buzzing was a prediction by Cybersecurity Ventures that by 2021, there will be roughly 3.5 million unfilled cybersecurity jobs across the globe.
“It’s a cool job,” security entrepreneur Allison Wong tells young women exploring options for their tech careers. “If you stay in it for four years and show you’re good, you’ll make in the six figures. And not just the low six figures.” Plus, one more advantage, she tells them: “It’s not a job you can get bored at. If you get bored, you’re doing something wrong.”
Interesting concept of art on the blockchain
Uses a DAG data structure, vertex contains hash of two main vertices.
Timestamp is given to reflect when most of the network members have recieved that very transaction.
Each node receives all transactions for the shard it is in.
Asynchornous BFT consensus.
Has already launched mainnet.
Came across them back in 2018 when learning about blockchain. Did not know they would be doing their own cryptocurrency and a mainnet. Not sure what’s happened so far. Signed up for community testing.
Glad to have gone through most of these. Bookmarking to share
JPM vs Ripple
JPM: internal credit based on blockchain technology. Okay but based on conversations I’ve had with them, this is just the beginning. They want to eventually fold in more banks into their ecosystem.
XRP: ripple company owns 54% of total XRP. what the heck?
This article does not provide recommendations, just lots of desktop research made into a single article.
I think there’s a typo in the article name… probably ‘creating bitcoin trading bots THAT don’t lose money’
Pretty interesting read on creating a bot that makes the typical trading activities and running it on the bitcoin price chart. But why specifically Bitcoin only? Also is this replicable for other coins?
Anyway, worth trying out yourself. Keeping it for the weekend.
Decentralised finance (DeFi) — ability to borrow and repay your loans in a decentralised manner without middlemen. Using far simpler smart contracts to improve the process itself.
Interesting, wonder who will take the lead.
Specifically designed for supply chain applications. Uses zk-proofs. Time to look it up.
It’s empty. Repository here
Great job from Aleth.io, this is the sort of stuff i’d like to try and do. I wonder where they get their datasets though. They have their own datasets that one can make sense of.
Things seem to be progressing along fine?
Doesn’t… say AragonOS have something like this? I suppose it’s an interesting thought experiment to build out.
Useful to act as a sort of template for scraping things!
About time somebody made an effort to consolidate the information. But it’s really mostly consensys’ tools.
Quick explanation into how RSA works.
Many operations in public key encryption involve the (mod p) operation and where we take the modulus involving prime numbers. This is defined as a finite field. In this case we will use Python to generate a random n-bit prime number. In RSA, for example, we take two prime numbers (p and q) and then multiply them together to create a modulus (N). The value of N is then part of the public and the private key. For RSA-2048 we use two 1,024-bit prime numbers, and RSA-4096 uses two 2,048-bit prime numbers.
A-Z glossary for the buzzword lover in you
The general workflow of an AutoML pipeline is fairly straightforward. You define the training dataset (the Richter dataset), a cross-validation scheme (5-fold cross validation), primary metric (weighted AUC), the experiment type (classification) and you are good to go. Depending on where you want to execute the experiment (on your local pc, a VM in the cloud or a large cluster) and where the data resides (local or in cloud), you also have to specify a data and compute configuration.
OCR- optical character recognition
Uses pytesseract library
Elasticsearch — searching documents on keywords
Kibana — for visualisation
Got to try it myself when my laptop is back.
Would be interesting to have that for… something that is mostly available in image form and not searchable yet
I remember having to go through the entire GAN field almost a year back as part of a project. I think this article helps to summarise some of the major findings including some sample code.
Common pitfalls and solutions to having users complete and understand new product features. I’ve not actually done a product tour, much less run one myself but the concepts are quite similar to some sort of onboarding experience.
So this is the first time I’ve heard of what a modal is. Basically it’s a pop-up box.
Interesting and useful list!
Time to add this to my list. Got to finish it!
- Lambda functions so you don’t have to write an entire function
- Maps to apply a function to a single list or multiple list
- Filtering to return elements that only return
- Itertools to perform interator operations without all that range(len(x)) stuff
- Generators to declare a function that behaves like an iterator but it generates it as you need. Less RAM used
There are so many packages out there to visualize data and no clear best package. Hopefully after reading through this review, you can see how some of the various aesthetics and code lend themselves to different situations, from EDA to presentation.
Went through networkx, pygal, plotly, bokeh, ggplot
PICK UP CLOJURE?! or at least
So he says he’ll come up with some sort of tutorial to try this. Can’t wait then!
Some things to pick up… in a weekend?
- ClojureScript Koans, learn the syntax of the language interactively on the browser.
- Clojure Brave and the True, best place to start to learn the concepts of Clojure and functional programming.
- 4Clojure, entertaining exercises and learn from others’ solutions.
- re-frame documentation, the best source to understand the framework, and really enjoyable to read.
- purelyfunctional.tv, many exceptional post blogs to understand reagent (wrapper for
ReactJS) and re-frame.
- Simplicity matters, by the creator of Clojure himself.
- ClojureScript: Lisp’s Revenge, by one of the key architect of ClojureScript.
I’ve actually used docker before? This one is very much just a list of commands and the overall architecture. But yeah good for absolute beginners.
Another thing to add to the learning list!
- switch function — it’s like an if
- shortcut keys for RStudio — i mostly just use the one that runs code
- flexdashboard — for Shiny and also lets you use Rmd
- req and validate — to act as error handling
- styler — sort of like linting?
- Parameterise your Rmd — then you can just replace the report with whatever variables you need
- revealjs — now this is really interesting. People can simply follow your presentation off their own phones!
- praise — praises people when they do something
REALLY USEFUL. SURPRISINGLY USEFUL. And also well structured. Much to learn here.
Things to google: time to write a follow up article?
OOCSS, BEM, SMACSS, SCSS
OOP Angular functions, prototypes in action, OOP and FP patterns, SOLID principles with OOP and FP
visual design principles, material design, awwwards, Smashing Magazine, fluid design, adaptive design, mobile first design
git, git workflow commands, learn git
unit testing principles, what unit testing is not (to help dismiss common misconceptions), Jasmine, Mocha, Jest, differences between unit testing frameworks and libraries, unit testing patterns
Node.js CRUD, Node connection to database, Node API patterns, MongoDB, mySQL design and architecture
AWS/Azure/Google Cloud web stack/deployment, Heroku and Docker, Docker deployments, continuous delivery pipeline for AWS/Azure/Google Cloud/Heroku
This genius made a COMIC. READ IT HERE
Decoding Difficult Conversations: My Interview with Negotiation Expert, Sheila Heen [The Knowledge Project Ep. #57]
When I least want to listen and when I’m most frustrated, I need to actually lean into the conflict
Your first negotiation is really with yourself
Interests are feelings
The way to introduce feelings, you can do two things: listen for the feeling or share how you’re feeling
If a conversation feels difficult, there’s often something about identity
Be a nobody, experience the Sublime.
Be a nobody, be free from societal expectations.
Be a nobody, understand that you are the sum of your struggles.
Okay I think it’s a bit try-hard. You can be somebody but still be humble. I think i’ll stick with be humble.
Using redundant phrases = insecure
Maybe that’s true but I’ve found that you need to cushion some of your words because others take it too heavily.
Lift open source code and rent it to others.
Build safety, share vulnerability, establish purpose.
Know your role in the team, signal the right sort of culture you wish. Embody it.
Also, know why this team even exists. Know it’s purpose and why it was formed.
Team structure -> Team culture. Some ideas:
Weekly demo meetings, teaching slots for each team member, cupcakes for launches
Become a “Google Analytics for space”.
The company earned $1.8B in revenue in 2018, with 88% of that revenue tied to memberships. That 88% number includes enterprise clients, which now comprise 32% of WeWork’s total membership. It does not include revenue from WeWork’s ancillary offerings:
- In-space services and add-ons like printing, phone service, WiFi, coffee
- Daily, non-recurring space rentals
- Perks and subscriptions sold through the WeWork services store
WeWork’s business model is built on principles of compounding returns. By combining scalable office space solutions with a growing array of services and amenities, WeWork attracts an ever-expanding ecosystem of solopreneur, startup, small business, and enterprise clients. As membership grows, WeWork is able to leverage the size and quality of this network to offer better lease terms to its landlords, which in turn leads landlords to offer WeWork better deals on space. Better deals lead to more space, more space leads to more members, and more members lead to better deals. The result is a self-reinforcing cycle that builds momentum as the company grows, allowing WeWork to unlock network effects, efficiencies of scale, and, eventually, profit.
In January 2019, the company announced that it had officially rebranded as The We Company, with a self-proclaimed mission to “elevate the world’s consciousness.” The move gestures at a vision for the company that expands way beyond co-working, spanning living spaces, fitness spaces, schools, a startup accelerator, and more.
WeWork has built a brand on big vision and big expectations. With the rebranding as The We Company, and its upcoming arrival on the public market, it seems that vision is about to get bigger still. But so much about the company — from the profitability of the core co-working business to the financial viability of its new ventures to its ability to deliver value that can justify that $47B valuation — is still hypothetical.
If the company can maintain its growth trajectory and successfully leverage the treasure trove of data it’s gathering in a bid to become “the Google Analytics of space” and survive threats to its survival in the form of capital constraints, competition, and a possible market downturn, it may yet be able to live up to the promise of its valuation.
Here are Google’s leadership evaluation questions:
- My manager gives me actionable feedback that helps me improve my performance.
- My manager does not “micromanage” (get involved in details that should be handled at other levels).
- My manager shows consideration for me as a person.
- The actions of my manager show that he/she values the perspective I bring to the team, even if it is different from his/her own.
- My manager keeps the team focused on our priority results/deliverables.
- My manager regularly shares relevant information from his/her manager and senior leaders.
- My manager has had a meaningful discussion with me about career development in the past six months.
- My manager communicates clear goals for our team.
- My manager has the technical expertise (e.g., coding in Tech, selling in Global Business, accounting in Finance) required to effectively manage me.
- I would recommend my manager to other Googlers.
- I am satisfied with my manager’s overall performance as a manager.
Then Google employees are asked to complete two other questions:
12. What would you recommend your manager keep doing?
13. What would you have your manager change?
We cannot untangle ourselves from the past. Self-mastery requires us to understand how it has shaped our circumstances, our attitudes, beliefs and our values.
Your passions will lead you to obstacles and resistance, but that, to Nietzsche, is a good thing. Overcoming resistance is gaining strength. Just as lifting weights build muscle, the problems we face when we follow our passions will make us grow and flourish as individuals.
Those who choose easy contentment and material happiness will never come close to achieving the exuberance of the Overman. The secret to finding true happiness, as far as Nietzsche presents it — to rejoice in one’s strength and exuberance — is to risk happiness.
Not sure if this was meant to be a cure for negativity but more of a way to live fulfilled. I suppose one could be fulfilled and still be negative.
Review of platforms, guides, frameworks, practice environments and even an entire approach to this. Will need to pick this up in order to work on serve.sg if i’m to communicate with the security folks.