Orion's arm: a far future, deeply transhumanist, hard sci-fi collective world building effort. LWers will find themselves immediately at home, and at every page you will find something interesting or surprising. Eliezer is explicitly quoted as a source for their mindspace diagram.
On me it exerts an attraction even stronger than tvtropes, so be warned!

I interviewed Robin Hanson about The Age of Em on my podcast Future Strategist.

Touhou:

• "Trojan Asteroid Jungle" (akikiki; {2015}) [classical]

Vocaloid:

• "Heliosphere" (Jizel {2013}; a tribute to the Voyager probes)

Stellaris - a 4X sci-fi game with interesting pro's and con's.

I especially like the very detailed race creation, with choices ranging from basic morphology to meta-ethics and government type. I also like that the game accounts for ethical drifting within your population, different models of FTL travel and malignant events such as UFAI fooming (called synth rebellion in the game).

However it's a young and complex production, and some of its strategic aspects are poorly balanced, read the negative Steam reviews for those, I've not played it enough and with the right focus to have an opinion on those.

As a pure sci-fi 4X, it's not the best: you will probably enjoy Endless Space or Beyond Earth much more.
But as a meta-civ simulator, I think it's amazing. I play Stellaris as if I'm playing a space version of Democracy 3: coming up with ludicrous empires and having fun evolving them in space, without much focus on winning.

The (Dis)Honesty Project hopes to create a safe space where we can explore the complicated truth(s) of the matter and improve our own behavior and that of the world around us.

http://thedishonestyproject.com/

• Zero Escape 3 - Zero Time Dilemma (video game with positive review on LW and ANN I instead watched a YouTube 100% walkthrough of, skipping the room-escape puzzles. I watched the first few such puzzles but found them rather tedious-looking, so that's definitely a reason to prefer watching Zero Time Dilemma to playing it. This way, ZTD can be considered a very low-budget anime done using 3D CGI lasting perhaps 10 hours. The CGI itself is... not great, by any means, but one understands what's going on, and it's not quite as tedious as reading a visual novel like Umineko even if Umineko had much better 2D artwork. That leaves the story itself, which is ambitious SF, trying to work together time travel, parallel universes & shifting, the anthropic principle, the Prisoner's Dilemma, the Sleeping Beauty problem, and amnesia drugs, in the framework of people trapped in a bomb shelter by a psychopathic mastermind. The 'decisions' made in each game provide a garden of forking paths the player navigates between to explore the consequences of particular choices, trying to navigate to a good end, where good is relative to the consequences of other possible universes. Eventually the characters themselves start understanding the situation and start shifting as well. Most choices end in death, so the exponential increase doesn't get out of hand. Some of the plot twists are dramatic and surprised me - for example, the architecture twist surprised me because while I had wondered how they could all have access to the same exit elevator if they were sealed off, I never put the remaining pieces together. Probably I would've enjoyed it more if I had watched the previous games to get the backstory, but it was still a fun SF series. That said, I felt some of the story was half-baked and it did not work as well as Steins;gate: the characters do disappointingly little to work around the amnesia despite ample opportunity; I was surprised by the ending because although I had guessed Zero's motive long before, at the anthropic principle/dice game, some of the events involving x-codes had seemed to specifically disconfirm it and the ultimate product seems kind of trivial compared to Zero's efforts to obtain it; the availability of both sliding and a space-time teleporter felt inelegant and the possibilities underused; the revelation of who Zero is was a complete asspull which - shades of Umineko - relies on a 'treachery of images' trick which simply doesn't work for both narrative & animation reasons, and worse, was unnecessary; and the narrative pacing is weak as the first part of the game spends what feels like an eternity on the repetitive starting Prisoner's dilemma game and then rushes through the rest.)
• La Maison en Petits Cubes (a simple concept, similar to the later memory sequence in Pixar's Up, but like Up, effective)
• Yurikuma Arashi (another peculiar Ikuhara production. It carries over much of his style in coloring and background and inscrutable symbolism from Revolutionary Girl Utena & Mawaru Penguindrum. YA is far more inscrutable through to episode 5, where finally enlightenment begins; then everything is explained to an extent unusual for Ikuhara. Stated baldly, his thesis of 'true love saving the world' is not nearly as interesting as the fancy garments he cloaks it in. Combined with the irritatingly flat characters, I suspect this may partially account for why YA was unsatisfying & unpopular.)

Harry Potter and the Cursed Child is available now. I devoured the script in four hours and I will only say: it's powerful and beautiful.

• The Man'yoshu (review)

I'm halfway through Italo Calvino's Cosmicomics. They're science-based creation myths, and they're breathtaking.

• The Theory of Special Operations, McRaven 1992 (review)
• The Genius Factory, Plotz 2005 (review)
• Average is Over, Cowen 2013 (review)
• The Pillow Book of Sei Shonagon (review)

Everything is heritable:

• "Predicting educational achievement from DNA", Selzam et al 2016 (supplement; polygenic scores now predict 3.5% of intelligence, 7% of family SES, and 9% of education)
• "Chinese scientists to pioneer first human CRISPR trial: Gene-editing technique to treat lung cancer is due to be tested in people in August"
• "Genetically-Mediated Associations Between Measures of Childhood Character and Academic Achievement", Tucker-Drob et al 2016
• "The continuing value of twin studies in the -omics era", van Dongen et al 2012
• "Estimate of disease heritability using 4.7 million familial relationships inferred from electronic health records", Polubriaginof 2016
• "Evolution Is Happening Faster Than We Thought" (Coyne commentary)
• "Diagnoses in Siblings of Probands With Autism Spectrum Disorders", Jokiranta-Olkoniemi et al 2016 (see also "Different neurodevelopmental symptoms have a common genetic etiology", Pettersson et al 2013)
• "Autism As a Disorder of High Intelligence", Crespi 2016 (Seems to overemphasize the role of common variants and underplay all the severe de novo mutations discovered in autism cases.)
• "Detecting Genome-wide Variants of Eurasian Facial Shape Differentiation: DNA based Face Prediction Tested in Forensic Scenario", Qiao et al 2016
• "Healthy ageing of cloned sheep", Sinclair et al 2016

Politics/religion:

• "Preparing for the Possibility of a North Korean Collapse", Bennett 2013 (I didn't know many of those details about how the East German collapse was able to go so smoothly, or that Kim Jong-Il had intimidated his subordinates during the '90s famines with footage of impoverished East German elites.)
• "Police, Prosecutors and Judges Rely on a Flawed \$2 Drug Test That Puts Innocent People Behind Bars"
• "How Successful Was Christianity?"
• "Opium Made Easy"
• France's untouchables
• "Seeds of Doubt: An activist's controversial crusade against genetically modified crops"
• "Hell is the Absence of God", by Ted Chiang
• "The shadow commander: Qassem Suleimani is the Iranian operative who has been reshaping the Middle East. Now he's directing Assad's war in Syria"

AI:

• "Alignment for Advanced Machine Learning Systems", Taylor et al 2016
• "Assessing Human Error Against a Benchmark of Perfection" (Difficulty of a chess endgame position predicts mistakes far better than time constraints or chess ability. Not necessarily surprising - as chess AIs demonstrate, humans only span a fraction of the possible range of objective chess playing ability, so individual human differences in ability won't matter much.)
• "Accelerating Eulerian Fluid Simulation With Convolutional Networks" (neural network all the things)

Statistics/meta-science:

• "Preventing future offending of delinquents and offenders: what have we learned from experiments and meta-analyses?", Mackenzie & Farrington 2015
• "Why It Took Social Science Years to Correct a Simple Error About 'Psychoticism'"
• "Can Results-Free Review Reduce Publication Bias? The Results and Implications of a Pilot Study", Findley et al 2016 (on result-blind peer review)
• "Laplace the Bayesianista and the Mass of Saturn"
• "Our Fathers of Old", Rudyard Kipling

Psychology/biology:

• Do Portia spiders have a mind? (commentary)
• "The Medical Mystery Behind the Itch"
• "United States of Paranoia: They See Gangs of Stalkers"
• "New Drug Development: Estimating entry from human clinical trials", Adams & Brantner 2003
• "The Efficacy of Psychological, Educational, and Behavioral Treatment: Confirmation From Meta-Analysis", Lipsey & Wilson 1993 (incidentally, published studies yield mean effects 0.14 SDs larger than unpublished studies.)
• "Could Women Be Trusted With Their Own Pregnancy Tests?"
• "Breast-feeding the microbiome"
• "Dengue fever cases drop 91% in neighbourhood of Piracicaba, Brazil, where Oxitec's Friendly Aedes were released"
• "Genetically Engineering an Icon: Can Biotech Bring the Chestnut Back to America's Forests?"
• "A Cavity-Fighting Liquid Lets Kids Avoid Dentists' Drills"
• Have a nice day