Here is the latest bag of tweets^*, which covers October 2018.

(*) These are interesting news that I found on Twitter and that I archive periodically. Note that I have slightly updated reporting and fetching methods so that you will get between 80 and 120 of my most recent favorites from my Twitter account (@chlalanne).

• 2018-10-01 @clojuregrams: ; clojure.core (defn nthnext [coll n] (loop n n xs (seq coll) (recur (dec n) (next xs)) xs)))
• 2018-10-01 @roydanroy: New book on probabilistic programming on arXiv. I’m sure the authors @hyang144 @jwvdm @frankdonaldwood will welcome feedback. https://t.co/O6rUtYSgnB
• 2018-10-01 @ThomasMailund: Functional lists and queues in #rstats https://t.co/V9xVljw24h https://t.co/Ww9I8CPLCm
• 2018-10-01 @hrbrmstr: If/when you upgrade to macOS Mojave, you can have light and dark apps live side-by-side in harmony (if you’ve set 10.14 to ‘Batman’ (Dark) mode) with this handy utility. Esp gd for taking screen[sho|cas]ts in “normal” mode for apps // Gray https://t.co/JQdEuFSKlK
• 2018-10-01 @grrrck: One of my favorite #rstats moves is reading a directory of CSV files into a single data frame with just a few lines code, thanks to {fs}, {purrr}, and {readr}. https://t.co/4olLrcNlTI https://t.co/JPiT4AQhNW
• 2018-10-01 @jsbois: New term @Caltech with my course on data analysis in the biological sciences: https://t.co/PHSO0h9TQj. Lots of @mcmc_stan and @ProjectJupyter, and of course influence from @betanalpha. Will be good fun!
• 2018-10-01 @jtleek: Our new MOOC program: (1) Is pay what you want (including nothing!) for everyone. (2) Can be done on any computer with a web browser and an internet connection. (3) Assumes only high school math and reading as pre-requisites. https://t.co/PFfcLef9Xh
• 2018-10-01 @ericnormand: 😎 Functional Programming Tip ∙ Clojure ✔︎ Clojure 1.9’s new swap-vals! lets you easily implement mutable stacks and queues in an atom. https://t.co/8uNk43b78O
• 2018-10-01 @mbostock: I’m experimenting with notebooks on introductory topics in computer science, starting with linked lists. https://t.co/ITW4kdFGCH
• 2018-10-02 @newsycombinator: Sourcegraph is now open source https://t.co/ZsWRRtSTgk
• 2018-10-02 @EmacsReddit: A faster way to tangle org-mode – A command line tool to tangle code blocks in org file to source code file. https://t.co/34zpNSZAu2
• 2018-10-02 @ThomasMailund: Promises, their environments, and how we evaluate them in #Rstats https://t.co/gB5dltzvE5 https://t.co/D8kOwg1DAU
• 2018-10-02 @robertstats: I wrote this blog post for @statauk @TimberlakeLtd on threshold regression in @Stata 15. There’s data on goldfinches and the weather in the beautiful English backwater of Kenley! You can download the data and do-file and follow along. https://t.co/9pFvpMFnmz
• 2018-10-02 @zevross: A detailed new blog post to help you navigate what I consider the most underrated #rstats package, tmap. Colors, layouts, scale bars, interactivity and more. Complete with data collection code using tidycensus | https://t.co/Ns3HeelIuG https://t.co/HqLUocWWGQ
• 2018-10-02 @beesandbombs: https://t.co/6FZ1V0u7wc
• 2018-10-02 @mauricio_szabo: So far, the best experience when writting plug-ins for Atom for me is to drop JS and write in #clojure/#ClojureScript with #ShadowCLJS https://t.co/VQzMm455dT
• 2018-10-02 @newsycombinator: Should you learn C to “learn how the computer works”? https://t.co/xZYrFlQSLx
• 2018-10-02 @math_rachel: Excited about the release of fastai v1! fastai is the 1st deep learning library to provide a single consistent (& state-of-the-art) interface to: - vision - text - tabular data - time series - collaborative filtering https://t.co/shHEz1tEsR
• 2018-10-02 @jeremyphoward: Wow @karpathy just made my day at the @PyTorch conference: “I tell everyone to do https://t.co/GEOZuodrZj. There’s nothing better out there.” :) I still refer to his CS231n course notes all the time to figure out how best to explain things… https://t.co/njsjKrYIgB
• 2018-10-03 @clojuregrams: (defn vec-invert [v] (into {} (map-indexed #(vector %2 %1)) v))
• 2018-10-03 @seabbs: Just finished Generalized Additive Models - An introduction with R (https://t.co/N2pt7F0eHJ). Overall a great combination of #rstats and GAMs. Any suggestions for a follow up read focussing more on applications and interpretation?
• 2018-10-03 @trebor74hr: “IPython 7.0, Async REPL” by @Mbussonn https://t.co/160UeWWRNt #python
• 2018-10-03 @nikitonsky: NO https://t.co/fcwJR9Hp51 https://t.co/J5ppMwhcmb
• 2018-10-03 @awhstin: Visualizing temperature variability over the years with the #rstats ggridges 📦https://t.co/4uZ9oGqixU https://t.co/w3RXswKbGn
• 2018-10-03 @greghendershott: racket-mode for @racketlang and @emacs – overview how it works, recent updates, and possible future direction https://t.co/izueZhUDMb
• 2018-10-03 @robjohnnoble: Non standard eval is my least favourite thing about #rstats. https://t.co/kHG4M4bPhN
• 2018-10-03 @davidjayharris: Suppose I want logistic regression on large categorical data (10^7 categories). Do I need to use dimensionality-reducing “tricks” (eg random projections, hashing trick), or can I do it directly? If I do reduce dim, what are the risks? @beenwrekt @tslumley
• 2018-10-03 @RikaGorn: I’ve started a repo for collecting all my fav Twitter #rstats #R4DS tips and resources and categorized by keyword (e.g. clean, viz, git, report, etc) it’s pretty messy but way better than looking through my likes and bookmarks https://t.co/bcSjYUwTzj
• 2018-10-04 @colindcarroll: ArviZ has a release up on PyPI! If you do Bayesian things in Python, please give it a try and leave some feedback. Share, visualize, and criticize models from @mcmc_stan, @pymc_devs, emcee, pyro, and anything else you can turn into a dictionary of arrays: https://t.co/Rq8DNieIgn
• 2018-10-04 @zim2918: Painted model https://t.co/tGzynzly1O
• 2018-10-04 @_wilfredh: A thoughtful post arguing that the read function is more important than homoiconicity in lisps: https://t.co/MekDfRuY0o
• 2018-10-04 @zim2918: https://t.co/yqXI43tgk0
• 2018-10-04 @eddelbuettel: littler 0.3.5 on CRAN: Other kids broke my toys Everybody’s favourite way to rock R at the command-line and in scripts https://t.co/Ri9xnZiVtn #rstats https://t.co/95BVO4B3BF
• 2018-10-04 @twostraws: Array performance: append() vs reserveCapacity() – https://t.co/Nh7zTGwaUe #swiftlang https://t.co/2IEJIQBqpf
• 2018-10-04 @Jose_A_Alonso: Haskell study plan (an opinionated list of resources for learning Haskell). ~ Gil Mizrahi (@_gilmi). https://t.co/SwDEmXdzPR #FunctionalProgramming #Haskell
• 2018-10-04 @tslumley: As an example of the dangers of over-generalisation in machine learning, it’s hard to improve on the mushroom dataset (https://t.co/yCPlkHh4of). If you train on North American Agaricus and Lepiota species, you can easily get a rule that thinks Amanita phalloides is safe!
• 2018-10-05 @hrbrmstr: This is a super-helpful addition to @RStudio. I prototyped https://t.co/j3LAjb6ar6 using it for the San Diego Union-Tribune’s Immigration Piece Interactive // r2d3 - #rstats Interface to D3 Visualizations | RStudio Blog https://t.co/lhQhSiB9us https://t.co/IiSBpvMQBs
• 2018-10-15 @mf_viz: Explaining (and understanding) the Central Limit Theorem is hard, so I built this interactive visual explanation to introduce my data science students to inferential statistics. Made with #d3js and #React. Enjoy! https://t.co/Tb7AdeCXUE
• 2018-10-17 @StatGarrett: Hands-On Programming with R is now available for free online, https://t.co/GNEGOVeM8h! Please help me spread the word. Thank you O’Reilly for the CC! #rstats #bookdown @rstudio @dataandme https://t.co/HKs1eKybBw
• 2018-10-18 @snoyberg: I’m happy to announce that I’m starting a Rust crash course. Introduction is now available on my blog, first lesson will be available next week! https://t.co/wohbiiLSwa
• 2018-10-19 @binitamshah: Interactive example of how Facebook leaked 50M access tokens : https://t.co/u2RsmyG9Xg More : https://t.co/EovKoA9CYZ
• 2018-10-23 @rlmcelreath: Multithreading works in rstan 2.18.1 @mcmc_stan now. Need a new flag in Makevars: $ cd ~/.R $ echo “PKG_CXXFLAGS += -DSTAN_THREADS” >> Makevars And in R session, set thread count: > Sys.setenv(STAN_NUM_THREADS=4) https://t.co/jy09yRDvFK
• 2018-10-23 @amcrisan: How do researchers visualize data in #genomic #epidemiology? Myself, @jennifergardy, and @tamaramunzner wanted to know. So - we conducted a systematic review of #datavis in #genepi that’s now out in @OxfordJournals Bioinformatics: https://t.co/4DzNR3gxW8 (1/19)
• 2018-10-23 @rlmcelreath: I just pushed a ton of commits up to the Experimental branch of rethinking. This is all paving the way to 2nd edition of textbook. >> https://t.co/2RgOpZoWqt
• 2018-10-23 @Jose_A_Alonso: Principles of algorithmic problem solving. ~ J. Sannemo. https://t.co/eTbRqLTFv2 #eBook #Algorithms #Programming #CompSci
• 2018-10-23 @yonchovski: Emacs is OS which lacks Java Integrated Development Environment no longer true. #emacs #spacemacs #dapmode https://t.co/OIFmXqQoTA & https://t.co/9vhln60Zrl https://t.co/q4VDheUckQ
• 2018-10-23 @ucfagls: Introducing gratia, my new #rstats pkg for working with and plotting GAMs estimated using #mgcv https://t.co/234gZvtkTM
• 2018-10-24 @CompSciFact: “We introduce the zip tree, a form of randomized binary search tree.” https://t.co/RfuKfpOvBa
• 2018-10-24 @ThomasMailund: Ok, I think I know why c(x,y) is faster for numerics than lists — they are simpler objects to copy, I guess. But I wonder why subscripting is so much faster for building lists in #Rstats. https://t.co/0RQHCWQGxe
• 2018-10-24 @maartenzam: A new and very strong contender in the “most amazing population maps” category: Human Terrain by @matthew_daniels from @puddingviz https://t.co/vzT4oHwMfE (love the “people residing on screen” feature) cc @undertheraedar @tjukanov @MaxCRoser https://t.co/UGzrLDfKYh
• 2018-10-25 @realpython: Adding a pipe operator to Python – Hacker Noon https://t.co/z2ne65gf2H
• 2018-10-25 @newsycombinator: End-to-end implementation of a machine learning pipeline (2017) https://t.co/1XMfMPbPgX