# draft Published articles for draft. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## Nx 22.7 Is Here: Task Sandboxing, 7x Less Memory, and Worktree-Aware Caching DevFeed: [Nx 22.7 Is Here: Task Sandboxing, 7x Less Memory, and Worktree-Aware Caching]() Original publisher: [Read original article]() Author: Juri Strumpflohner Published: 2026-04-28T00:00:00Z Content type: release Language: en Sources: [Juri Strumpflohner]() Topics: [Caching](), [releases](), [Development](), [configuration](), [tracing](), [Git]() Tags: [agentic](), [cache](), [caching](), [draft](), [git](), [nx](), [parallel](), [release](), [releases](), [tracing]() ### AI overview Nx 22.7 consolidates changes from Nx 22.4 through 22.7, including task sandboxing with I/O tracing, reduced daemon memory usage, worktree-aware caching, more precise cache invalidation, resolved-configuration inspection, and agentic development improvements. ### Source excerpt Everything that landed in Nx across 22.4 through 22.7: task sandboxing, a 7x reduction in daemon memory, worktree-aware caching, agentic mode improvements, and more. ## The Verse Calculus: a Core Calculus for Functional Logic Programming DevFeed: [The Verse Calculus: a Core Calculus for Functional Logic Programming]() Original publisher: [Read original article]() Published: 2022-12-12T16:23:17Z Content type: article Language: en Sources: [Lambda the Ultimate]() Topics: [Programming](), [Math and Logic]() Tags: [draft](), [functional](), [lambda-calculus](), [languages](), [logic](), [paper](), [programming](), [rewrite](), [semantics]() ### AI overview The draft paper introduces the Verse calculus (VC), a core calculus for functional logic programming. Its main contribution is a small-step rewrite semantics that enables reasoning about VC programs through successive rewrites, as with lambda calculus. The authors note that the design is still a work in progress and that the current rewrite rules may not be confluent. ### Source excerpt The Verse Calculus: a Core Calculus for Functional Logic Programming https://simon.peytonjones.org/assets/pdfs/verse-conf.pdf LENNART AUGUSTSSON, Epic Games, Sweden JOACHIM BREITNER KOEN CLAESSEN, Epic Games, Sweden RANJIT JHALA, Epic Games, USA SIMON PEYTON JONES, Epic Games, United Kingdom OLIN SHIVERS, Epic Games, USA/li> TIM SWEENEY, Epic Games, USA Functional logic languages have a rich literature, but it is tricky to give them a satisfying semantics. In this paper we describe the Verse calculus, VC, a new core calculus for functional logical programming. Our main contribution is to equip VC with a small-step rewrite semantics, so that we can reason about a VC program in the same way as one does with lambda calculus; that is, by applying successive rewrites to it. This draft paper describes our current thinking about Verse. It is very much a work in progress, not a finished product. The broad outlines of the design are stable. However, the details of the rewrite rules may well change; we think that the current rules are not confluent, in tiresome ways. (If you are knowledgeable about confluence proofs, please talk to us!)We are eager to enagage in a dialogue with the community. Please do write to us. ## Silent Duels--Constructing the Solution part 2 DevFeed: [Silent Duels--Constructing the Solution part 2]() Original publisher: [Read original article]() Published: 2022-03-24T07:00:00Z Content type: tutorial Language: en Sources: [Jeremy Kun]() Topics: [Mathematics](), [implementation](), [Python]() Tags: [draft](), [implementation](), [mathematics](), [python]() ### AI overview A draft tutorial documents partial progress implementing a general construction of optimal strategies for silent duels. The first serious test case fails, and the author remains unsure how to fix it. ### Source excerpt Previous posts in this series: Silent Duels and an Old Paper of Restrepo Silent Duels--Parsing the Construction Silent Duels--Constructing the Solution part 1 Since it's been three years since the last post in this series, and the reason for the delay is that I got totally stuck on the implementation. I'm publishing this draft article as partial progress until I can find time to work on it again. If you haven't read the last post, please do. ## Defining a new HTTP method: HTTP QUERY DevFeed: [Defining a new HTTP method: HTTP QUERY]() Original publisher: [Read original article]() Author: HTTP Toolkit; Tim Perry Published: 2021-04-12T15:00:00Z Content type: article Language: en Sources: [HTTP Toolkit]() Topics: [HTTP](), [API](), [Internet Engineering Task Force (IETF)]() Tags: [api](), [apis](), [draft](), [http](), [ietf](), [request](), [responses](), [server](), [standards]() ### AI overview This article introduces HTTP QUERY, a new HTTP method for safe requests that include a request body. It explains why the method is needed and compares it with existing HTTP methods such as GET and POST. The article notes that HTTP QUERY was recently adopted as an IETF draft standard and that its specification evolved from the earlier name SEARCH. ### Source excerpt Nothing is ever finished or perfect, and HTTP is no exception. HTTP QUERY is a new HTTP method, for safe requests that include a request body. It's still early & evolving, but it was recently adopted as an IETF draft standard, and it's going to add some great new tools for HTTP development everywhere. What does that mean, why do we need a new HTTP method, how would HTTP QUERY work? Update: This post previously called this method SEARCH, but since it was originally published the spec has been updated, and the method is now called QUERY. This post has been updated accordingly. HTTP methods today Today, there are 5 main HTTP methods you'll see in modern APIs. To understand how each one works, it's important to remember that HTTP is defined in terms of resources. A resource might be a document, or a photo, or a specific customer, or the whole list of customers, and it's identified by a URL like example.com/customers (all customers of example.com) or example.com/customers/123 (one specific customer). GET A GET request asks the server for a resource. This is frequently used to request HTML pages, read data from an API, or load images. These are intended to be 'safe' requests, which purely read data. They shouldn't change the state of the server, they shouldn't have side effects, and so they can be cached in many cases (which means that many client GET responses will come from a cache, and never hit the real server). GET requests can be parameterized by their URL, which might contain a path and/or query parameters, but they can't have a request body. It's not specifically banned, but it is defined as being completely meaningless, and many existing implementations will ignore the body or reject the request entirely if you try to send one. They can also use Accept, Accept-Language and Accept-Encoding headers to request a specific content type, language or encoding ('give me customer 123 as XML please'), and use Range headers to request only part of a document ('give me the f ## Headless Preview in Craft CMS DevFeed: [Headless Preview in Craft CMS]() Original publisher: [Read original article]() Author: andrew@nystudio107.com (Andrew Welch) Published: 2019-08-13T02:40:00Z Content type: tutorial Language: en Sources: [nystudio107 | Articles on modern web development.]() Topics: [Content Management System](), [Single-page application (SPA)](), [API](), [Front end](), [Database]() Tags: [api](), [cms](), [content](), [craft](), [database](), [draft](), [editor](), [exploration](), [frontend](), [headless](), [here-s](), [implement](), [insights](), [introduced](), [preview](), [react](), [svelte](), [vue](), [works]() ### AI overview This article explains how Craft CMS 3.2 introduced headless content preview for pages rendered by a separate single-page application frontend. It describes the tokenized preview workflow, using auto-saved drafts, database records, tokenized URLs, and API requests to connect the editor with the frontend renderer. ### Source excerpt Craft CMS 3.2 introduced headless content preview. Here's an exploration of how it works, and how you can implement it