I used it to create the Betula logo.

Сравнение трудового законодательства в России и Сербии

The weirdest old-school text adventure on the internet.

The idea is you set up a Stapler Document per project containing related apps, files, folders, etc. Then you can open them all at once by launching the single document. Each document contains a list of aliases that can be managed, inspected, launched using the app. The key time-saver is that if you launch a Stapler Document directly, all the items in its list will automatically be launched. Cool!

I actually wanted that! And this one is a remake of a classic app. Cool.

An alternative to the World Wide Web

Janet is a functional and imperative programming language. It runs on Windows, Linux, macOS, FreeBSD and *nix.

Designed to be embedded. A Lisp!

From table hacks to CSS kludges, these are some of the techniques that helped web developers get their fancy pages on the internet.

This site is best viewed in Lynx. In other browsers, it tries its best to simulate that experience.

Нередко приходится приводить этот сайт в пример мощного веб-дизайна, сохраню на всякий.

Info on HTTP signatures.

Я заметил, что люди постоянно задают одни и те же вопросы в чатах, интервью и радиоэфирах, и приходится объяснять снова и снова. При этом мне и самому было интересно разобраться, узнал для себя несколько новых вещей, пока вникал. Например, почему даже гонять полупустой автобус с гармошкой — не самая плохая идея.

An app that lets you add various links and displays them in a column layout. You can create a unified dashboard of several social networks, news feeds, and such with it.

This is an extension to clip websites and download them into a readable markdown file. Please keep in mind that it is not guaranteed to work on all websites.

A site generator using the file system as the interface.

One can use gzip to classify data.

Deep neural networks (DNNs) are often used for text classification due to their high accuracy. However, DNNs can be computationally intensive, requiring millions of parameters and large amounts of labeled data, which can make them expensive to use, to optimize, and to transfer to out-of-distribution (OOD) cases in practice. In this paper, we propose a non-parametric alternative to DNNs that’s easy, lightweight, and universal in text classification: a combination of a simple compressor like gzip with a k-nearest-neighbor classifier. Without any training parameters, our method achieves results that are competitive with non-pretrained deep learning methods on six in-distribution datasets.It even outperforms BERT on all five OOD datasets, including four low-resource languages. Our method also excels in the few-shot setting, where labeled data are too scarce to train DNNs effectively.

Our method is a simple, lightweight, and uni- versal alternative to DNNs. It’s simple because it doesn’t require any preprocessing or training. It’s lightweight in that it classifies without the need for parameters or GPU resources. It’s universal as com- pressors are data-type agnostic, and non-parametric methods do not bring underlying assumptions.

Without any pre-training or fine-tuning, our method outperforms both BERT and mBERT on all five datasets.

Questioned:

• https://kenschutte.com/gzip-knn-paper/

Нередко на статье о какой-нибудь очередной инновации я вспоминаю, что такое уже анонсировали лет 15—20 назад, но найти этого не удаётся, — все новостные издания, которые об этом писали, давно исчезли.

This is about Mycorrhiza, for sure.