Deep JavaScript Foundations, v3 vs Learn SQL
Same Bayesian formula, same rubric — so the difference in scores reflects the difference in the courses, not the difference in how we evaluated them.
Frontend Masters (Kyle Simpson) · Web Development
Deep JavaScript Foundations, v3
Codecademy · Web Development
Learn SQL
Per-criterion
Reviewers consistently describe the curriculum as the deepest treatment of core JavaScript available in video form. The three pillars — types and coercion, scope and closure, and the this/prototype object system — are explained directly from the language specification rather than from surface behaviour. Gurpreet Singh's testimonial captures the consensus: "I know of no other resource that goes this deep into fundamentals of JS." The main critique is that some sections, particularly the OLOO delegation pattern in the object-oriented module, are presented as settled best practice when many experienced developers disagree.
Kyle Simpson, author of the You Don't Know JS book series, is widely praised for clarity and depth. Lara Karki called him "not only brilliant, but an excellent and articulate teacher" and Hatem Hassan noted he "has a unique way of explaining deep technical CS concepts in a simple and straightforward manner." The score is held below the instructor scores of less polarising teachers because his opinionated, prescriptive style draws genuine criticism — one HN commenter found his persona "very off-putting" and others object to his advocacy for == coercion.
The course is bundled into the Frontend Masters subscription at $39/month or $390/year, which also unlocks the entire library including Kyle's scope, closures, and asynchronous JavaScript courses. Lara Karki called the membership "the best $40 I'll spend this month, by far." Value is excellent for committed learners but the subscription model means a single 10.5-hour course is not buyable standalone, which frustrates learners who only want this one title.
The course makes developers materially better at reading and reasoning about production JavaScript — Kevin O'Shaughnessy noted it "blew my mind how much there was going on in 10 lines of code." However, it is conceptual rather than project-based: there is no application built end to end, no framework, and no tooling. Reviewers who wanted to ship something tangible note that this is a foundations course, and its payoff shows up later as fewer bugs and clearer mental models rather than a portfolio piece.
Frontend Masters provides downloadable exercise files, transcripts, and a hosted course platform with adjustable playback speed, but there is no graded auto-grader, no mentor, and no certificate. Learners rely on the public GitHub note repositories and exercise mirrors that other students have published. This is a lecture-and-exercise format, not a guided cohort, so self-directed learners do best.
Learn SQL is organised into four tightly scoped lessons — Manipulation, Queries, Aggregate Functions, and Multiple Tables — followed by five hands-on projects and four quizzes, with an estimated five hours to complete. The curriculum covers the genuine fundamentals of working with a relational database: creating and updating records, filtering and ordering result sets, computing aggregates with GROUP BY and HAVING, and joining related tables. For a complete beginner with no prior exposure to databases, this is a well-sequenced path that moves from individual statements to multi-table reasoning without overwhelming detail. The standout pedagogical feature, cited repeatedly across reviews, is the visual treatment of how data moves and combines. Multiple learners single out the animations that illustrate how rows transform during joins as the clearest explanation of that concept they had encountered, including some who had previously studied SQL at university and still preferred Codecademy's visual approach. The recurring criticism is depth. The course teaches SQL against a SQLite engine and, as one detailed blog reviewer put it, could fairly be retitled "Learn SQLite." It introduces primary and foreign keys but stops at uniqueness; it does not cover database normalisation, schema design, indexing, transactions, or window functions. Several reviewers noted topics that ended with an acknowledgement that more depth exists but would not be covered. The content is excellent for using an existing database and insufficient for designing a new one — an honest scope limitation rather than a quality failure.
The entire Learn SQL course — every lesson, project, and quiz — is free on the Basic plan. There is no paywall on the learning content itself, which over a million learners have taken. Payment is only required for a certificate of completion and Pro-exclusive features, available through Plus (around $14.99/month billed annually) or Pro (around $19.99/month billed annually), which add personalised feedback, career paths, and interview prep. For a learner whose goal is to become competent at querying data, the free tier delivers essentially the full value at zero cost — a strong proposition for a five-hour, fully interactive course. The certificate carries modest professional weight; reviewers and comparison articles broadly agree the real value is the skill gained rather than the credential, which is considered less rigorous than exam-based alternatives such as DataCamp's. The one caveat is platform-level: Codecademy's subscription scores poorly on Trustpilot (around 2.4/5), with complaints typically about billing and the upsell pressure toward Pro rather than the quality of this specific course. Learners who only want the free content should be deliberate about not auto-upgrading.
The skills taught — selecting, filtering, aggregating, and joining data — are exactly the day-one SQL competencies expected of analysts, developers, and data-adjacent roles, and the most common positive theme is immediacy. Learners describe being able to query data "the next day" and apply the syntax directly to work tasks within the same week. Codecademy positions the course as a building block of its Data Analyst Career Path, which adds Python, statistics, and visualisation around this SQL foundation. Because the course is purely query-focused, it transfers cleanly to any relational database a learner will encounter at work — the SELECT, JOIN, and GROUP BY patterns are standard across PostgreSQL, MySQL, and SQL Server. This makes it a reliable first rung whether the end goal is data analysis, backend development, or simply being literate when reading a colleague's query. The applicability ceiling is real-world database operations beyond querying. The course does not cover connecting to a production database, performance tuning, schema design, or the operational concerns of managing data at scale. One reviewer's analogy — that learning SQL here is like fencing against a hologram, with the theory but not the live feedback — captures the gap a learner must close with their own projects after finishing.
This is the course's defining strength. Every concept is taught inside Codecademy's browser-based coding environment, so learners write and run real SQL from the first lesson rather than watching video. There is nothing to install, no local database to configure, and the integrated console returns results immediately — a setup that removes the single biggest friction point that derails self-taught beginners. The five projects reinforce this. Guided projects such as "Create a Table" and the "New York Restaurants" exercise walk learners through applied scenarios, while the "Analyze Hacker News Trends" project pushes them toward more independent query writing against a realistic dataset. Reviewers consistently describe the learn-by-doing format as a "game changer" for self-study and credit it with making concepts stick far better than passive material. The trade-off, raised by more experienced learners and one forum thread, is that the sandbox can feel too forgiving: exercises sometimes surface full solutions or accept answers without the friction of debugging against a real production database. The practice is excellent for building correct mental models, but it does not replicate the messy feedback of connecting to and querying a live system.
Codecademy's model is interactive text instruction with inline exercises rather than a single charismatic video instructor, so "teaching quality" here means the clarity of the written lessons, the helpfulness of hints, and the pacing of exercises. On the whole this lands well: reviewers describe the explanations as touching "the essential with just enough context" and praise the gradual ramp that lets total beginners feel they "learned months in a week." The visual explanations of joins and table transformations are the most praised teaching element, repeatedly called the clearest treatment of the topic learners had seen. The embedded quizzes and immediate feedback loop keep engagement high and confirm comprehension after each lesson. Criticism centres on inconsistency in the exercise scaffolding. A minority of reviewers and forum posts found certain instructions terse or the hints unhelpful, and some exercises were felt to be padded with formatting that made them longer than the underlying concept warranted. These are isolated rough edges in an otherwise smooth and beginner-respecting teaching flow.
Scoring methodology applies identically to every course on the site — see the formula.