VentureBeat: PolyCoder is an open source AI code-generator that researchers claim trumps Codex. “…while research labs like OpenAI and Alphabet-backed DeepMind have developed powerful code-generating AI, many of the most capable systems aren’t available in open source. For example, the training data for OpenAI’s Codex, which powers GitHub’s Copilot feature, hasn’t been made publicly available, preventing researchers from fine-tuning the AI model or studying aspects of it such as interpretability. To remedy this, researchers at Carnegie Mellon University — Frank Xu, Uri Alon, Graham Neubig, and Vincent Hellendoorn — developed PolyCoder, a model based on OpenAI’s GPT-2 language model that was trained on a database of 249GB of code across 12 programming languages.”
EurekAlert: New data science platform speeds up Python queries. “Researchers from Brown University and MIT have developed a new data science framework that allows users to process data with the programming language Python — without paying the ‘performance tax’ normally associated with a user-friendly language.” The new platform is free.
The Register: The nightmare is real: ‘Excel formulas are the world’s most widely used programming language,’ says Microsoft. “Microsoft will let users create custom functions in Excel using the number wrangler’s own formula language….Dubbed LAMBDA, the feature (currently rolling out to beta customers) will be a lifesaver for anyone charged with maintaining herds of increasingly complicated spreadsheets, who have doubtlessly been wondering how it could be that Excel was missing such a seemingly obvious ability for so many decades.”
Ars Technica: The unreasonable effectiveness of the Julia programming language. “I’ve been running into a lot of happy and excited scientists lately. ‘Running into’ in the virtual sense, of course, as conferences and other opportunities to collide with scientists in meatspace have been all but eliminated. Most scientists believe in the germ theory of disease. Anyway, these scientists and mathematicians are excited about a new tool. It’s not a new particle accelerator nor a supercomputer. Instead, this exciting new tool for scientific research is… a computer language.”
Phys .org: Not a ‘math person’? You may be better at learning to code than you think . “New research from the University of Washington finds that a natural aptitude for learning languages is a stronger predictor of learning to program than basic math knowledge, or numeracy. That’s because writing code also involves learning a second language, an ability to learn that language’s vocabulary and grammar, and how they work together to communicate ideas and intentions. Other cognitive functions tied to both areas, such as problem solving and the use of working memory, also play key roles.”
Hongkiat: 10 Websites to Test Your Codes Online . “In this article I want to outline 10 interesting web apps for testing your code online. All of these apps require an Internet connection, and some of the more advanced editors offer pro plans to upgrade your account features. But most of these tools will surely come in handy when you’re scrambling to debug a block of JavaScript or PHP.”
ZDNet: Google reveals new Python programming language course: Scholarships for 2,500. “There are six courses in the Google IT Automation with Python Professional Certificate, including an introductory ‘crash course on Python’, learning how to use Python to manipulate files and processes on a computer’s operating system, a course in using Git and GitHub, troubleshooting techniques, learning how to automate and manage fleets of computers in the cloud, and automating real-world tasks with Python.” The article notes that the course is not free.
Make Tech Easier: Web Code Playground Tools You Should Try. “If you’re at all into web coding, you’ve probably used at least one code playground in your time. These aren’t your usual code editors, but places where you can test your code without worrying about the backend server setup. Codepen and JSFiddle are the two most popular, but, perhaps unsurprisingly (given that the target audience for web code playgrounds is people who have the skills to build code playgrounds themselves), there are many alternatives.”
Computer Business Review: GitHub Adds 10 Million New Users, Reveals 10 Most Popular Languages. “Over the past year, developers collaborated in a staggering 370 primary languages on GitHub. Among the top 10 programming languages, C#, Python and Shell climbed the list this year, while Ruby and Java fell in popularity. That’s according to the code repository’s annual Octoverse report, which also reveals a colossal 532 percent increase in the use of Google’s Dart language, as interest surges in the company’s Flutter SDK – built using Dart.”
Make Tech Easier: Seven Coding Games to Help You Build Your Programming Chops. “These coding games cover plenty of languages, age ranges, and skill levels, so whether you’re a complete beginner or looking for something on the next level, there’s a game out there to help you learn coding the best way: by doing it.”
DevClass: GitHub trials machine learning so you can mind your language. “GitHub is trialling a machine-learning powered system to identify the babel of languages across the code repo platform. You might wonder whether this is a big deal – surely people know what language they’re using, and that’s all that matters.”
Amit Agarwal: Google Apps Script for Developers. “In this video tutorial, you’ll learn how to develop Google Apps Script projects locally on your computer inside Visual Studio Code. You can write your code in modern JavaScript, neatly organized in modules, and the build environment will use Babel and Webpack to transform your code into a version of JavaScript that is compatible with Apps Script.”
Wolfram Alpha: Version 12 Launches Today! (And It’s a Big Jump for Wolfram Language and Mathematica). “OK, so what’s new in 12.0? There are some big and surprising things—notably in chemistry, geometry, numerical uncertainty and database integration. But overall, there are lots of things in lots of areas—and in fact even the basic summary of them in the Documentation Center is already 19 pages long…”
Quartz: What’s the best way to learn the programming language R? (Preferably, for free). “As data becomes an ever larger part of work, for many people spreadsheets just are not enough. Programs like Microsoft Excel and Google Sheets are powerful tools, but they have limitations in terms of the amount of data you can work with, the kind of analyses you can do, and the types of charts you can make. When data users reach these limitations, the obvious next step is learning a programming language.”
