"card_file_box Open source self-hosted web archiving. Takes URLs/browser history/bookmarks/Pocket/Pinboard/etc., saves HTML, JS, PDFs, media, and more... " ~10k stars on GitHub.

ISI NLP course taught by Jonathan May.

ABC episode on Intellectual property. Samir Chopra, Professor of Philosophy, interviewed.

> Following Abba Lerner’s influential formulation, this component is often referred to as functional finance (Lerner 1943). Our goal here is not to make an assessment of MMT as a whole, but to ask what is the relationship between the functional finance approach to government budgets and conventional economic analysis. Because we are interested in the logic of the functional-finance position rather than in MMT as a body of thought, we make only limited references to MMT literature here. Our primary interest is in the merits of the functional finance position in the abstract. On the mainstream side, we are focused on what might be called “orthodox policy macroeconomics”—the practical heuristics that guide policy makers and are reflected in undergraduate textbooks, as opposed to DSGE and related models of intertemporal optimization that are the basis of most current macroeconomic theory. The relevant question is not whether MMT is consistent with models of this type, but whether it is consistent with the simpler, more reduced form models that are (explicitly or implicitly) drawn on by public officials and public and private-sector forecasters, as well as by academic economists when engaged in public discussions.

fasttext paper

Careful identification of missing data (now corrected) in the Baumgartner reddit corpus.

Some kind of reddit alternative. Seems to have survived for some time.

"Students taking this module will read novels from all around the world, from the 19th century to the present, within the framework of recent debates over world-literature, modernity, and the still unfolding process of globalisation. We will consider how widening our comparative and international perspective enables us to read and interpret novels (even the most seemingly ‘local’ or ‘national’ productions) as irresistibly ‘global’. The module analyses how certain novelistic forms, themes and issues ‘travel’. Students will discover ways in which certain works contain traces, adaptations and importations of ‘core’ (or global) issues, and adapt, remodel or reject them in accordance with local/national forms and expressions. Issues such as work/war/resources/imperialism/climate/finance/disaster/modernisation and others will be traced as we work to establish each novel's 'globality'. We will consider a number of elements in form and subgenre - magic realism/blog novels/graphic novels/speculative fiction/annalistic wiritng, but we will also look at each work's mediation, its commercial and cultural production and its geopolitical conditioning. You'll also pick up a solid understanding of the crucial issues and debates in globalisation. The module will read novels from Scotland to China, Saudi Arabia to Colombia, England to the U.S, and demonstrate why a global perspective is a necessary requisite for literary studies in the 21st century."

Looks a bit like riseup. Has been around since 2005. "We are a democratically-run, not-for-profit cooperative of movement organizations and activists in the United States and Mexico. We've been around since 2005 and our 850 members (mostly organizations and mainly in the U.S. and Mexico) host over 10,000 email address and over 2,000 web sites on our collectively owned and secured hardware that run exclusively on encrypted disks."

Self-published books by John Swartzwelder, an Emmy-winning writer for the Simpsons. I don't know anything about him nor have I read the books. I find it interesting that he has been self-publishing for so long.

"In a previous post I applied a Bayesian ordered-probit model to movie ratings and showed how the results differ from treating the data as if they were metric. The metric model used frequentist t tests (because that's what most applied researchers would do). In this post, I re-analyze that data as if they metric but using a Bayesian model that has the same hierarchical structure as the Bayesian hierarchical ordered-probit model I used before. Here we can compare ordered-probit to metric treatments with all else held constant. Spoiler: Same conclusions, of course. The ordered-probit model fits the data much better than the metric model. Don't treat ordinal data as metric."

Xavier Marquez's year-end list of recommended reading.