ebook-for-education, a Thai course material archive kept in Git

One directory per course, and a table that links them out

The repository description is a single Thai sentence that means roughly, distribute teaching documents for study. The README is a table with three columns: the course, the topic, and a download link, and every row pairs a PDF book with a YouTube playlist for the same subject.

The catalogue covers introductory programming and web work. There is a C programming course, a Docker course, Git and GitHub, JSON, the Flask framework, OpenCV with Python for image processing, PHP, Node.js, object-oriented programming in Java, object-oriented programming in Python, a fundamentals course covering HTML, CSS and JavaScript, Visual Studio Code, React version 17, and a Prompt Engineering folder. Some of the entries are marked as free and some are marked as updated most recently, which is a hint that the material is revised from time to time.

The top-level directory listing matches the table, which is a good sign of a repository that is kept in order. Each course has a directory named after the subject in Thai, and the root also carries an assets/ folder, a learning animation, and an image that appears to be promotional material for the author's courses.

Around the repository there is a larger personal ecosystem. The README is a page of badges pointing at two YouTube channels, one for official announcements and one for tutorials, plus a Facebook page, a Udemy instructor profile, a Shopee storefront, a Medium publication, a CodePen profile, a TikTok account, and a GitHub profile for the author's organisation. The repository is one artefact of a teaching brand rather than a project with a maintainer and a roadmap.

Download links that point at different revisions of the tree

The most concrete technical detail in the repository is visible in the download links, and it is the kind of thing that only shows up when you follow them.

Some of the links in the table point at the main branch, for example the Docker, Flask, OpenCV and Python, Node.js and web fundamentals entries. Others point at a specific commit, with a long hash in the path, and the C, Git and GitHub, JSON, PHP, Java and Python object-oriented courses are linked that way. One entry, the Python object-oriented book, points at a third hash, a different one again.

So the same table offers two different kinds of link. A link to main follows the branch, which means the document a reader gets today is whatever the branch currently holds, and it changes if the material is updated. A link to a commit is frozen, which means the reader gets the document as it was at that revision, and it never changes.

Neither is wrong, and there is a reasonable reading in which the commit-pinned links are deliberate: a course video refers to a specific book, and the student following along should get the same pages the instructor was holding. But the mixture is undocumented, and there is no note explaining why some rows are pinned and others are not, so a reader cannot tell which behaviour a given row has without reading the URL.

The practical consequence is that two people following the same table can end up with different documents, and that a link which was correct when the README was written may now serve something the course no longer matches. For a repository whose entire purpose is to be the authoritative copy of a set of books, that is the one thing worth fixing.

PDFs in Git, with no releases and no versioning

The distribution method is a folder in a Git repository, which is a deliberate choice with consequences.

The good part is that everything is versioned. A teacher can fork the repository, adapt a course book, and the change is a pull request away. A student can check out the version that matches their cohort. A translation or a local adaptation is a branch rather than a separate download. There is also a guarantee that the material has not been swapped out from under a link, because the history is right there.

The cost is that there is no concept of a release. The repository has no GitHub releases, no tags, and no version numbering for the material. The last push was on 2026-09-30, and the README offers no changelog and no statement about which revision corresponds to which course run. For a course that runs once a year, that may not matter. For a course that is revised every term, the absence of a version marker means the maintainer of the material has to track what changed by reading the history.

The binary-in-Git question also applies. PDFs are large, and every revision of a revised book is a new blob in the history, which makes the repository grow faster than a text-only project would. The layout suggests this has been manageable, since the courses are slide-deck and booklet scale rather than scanned books, but the approach has a ceiling that a release-based distribution with external file hosting would not have.

No licence, which is the real obstacle to reuse

The repository has no licence recorded, and the README does not state one. The top-level listing contains no licence file among the course directories and the assets.

For a personal teaching repository that exists to support its author's own courses and channels, that is a defensible position. The material is there so that students can follow along and so that the courses can be advertised, and the videos on YouTube are the real product.

For anybody else, it is a blocker with no workaround other than asking. Without a licence, the default position under copyright is that nobody may copy, redistribute or adapt the work, and a fork that a school or a company makes for its own students is a legal question rather than a technical one. The same applies to the PDFs inside the directories rather than only to the repository as a whole, since the documents are the substantial part of it.

A single licence file would settle the question for every course at once, which is why its absence is the most consequential gap in an otherwise simple repository. It is also the cheapest gap to close, and the sort of thing a contributor could propose in a pull request if the project wanted to invite that.

The catalogue as it stands, and where it is dated

The course list tells you what the teaching covers and, by looking at the labels, how current it is.

The web fundamentals are grouped as a single HTML, CSS and JavaScript course, and the three entries in the table are labelled as revised editions from 2020, with the HTML5, CSS3 and JavaScript rows each carrying its own playlist and its own download link into the same shared directory. The React entry is for version 17, which is a long way behind the current line of that framework. Node.js and React are both marked as free, which suggests they are offered as an introduction to a wider set of paid courses.

The other material is less tied to a version: C, Java, Python, PHP, Docker, Git and GitHub, JSON, Flask and OpenCV are all subject areas where the fundamentals move slowly, so the books in those subjects stay usable for longer. The two directories whose names say the Python material is the updated version are the exception, and they are the ones most likely to have been revised in a way a reader would notice.

The Prompt Engineering folder does not appear in the README table, which suggests it is either newer than the table or handled separately. Either way, a reader browsing the repository directly sees more than the table lists, and a reader using the table sees less.

Where this repository does not fit

The first thing to say is what it is not. There is no code, no library, no API, no build, no test suite and no issue tracker discipline, and it will never be a dependency of anything. It is a document repository with a README that does the work of a catalogue.

That makes the wrong-tool question mostly about expectations. A developer looking for a course on Git and GitHub will find a booklet and a playlist, and that may be exactly what they wanted. A team that wants a maintained curriculum with versioned slides they can pin per cohort will find that the pinning is inconsistent and that the material has no version markers, and they will have to build that discipline themselves on a fork.

The second is language. The material is in Thai, the course names are in Thai, and the directory names are in Thai, which is entirely appropriate for the audience and completely unusable for a reader who does not read Thai. The repository does not offer a translation, and the badges point at English-language platforms, but the content itself is aimed at a Thai-speaking classroom.

The third is the absence of a licence, discussed above, which is the practical blocker for any reuse beyond the author's own teaching.

The fourth is the link consistency, which makes the table a slightly unreliable index of the repository's own contents.

The fifth is the way the material is distributed. Hosting PDFs in Git works, and it works well for a small set of booklets. It does not scale to a course with video, and the readme's own ecosystem, with YouTube channels, an Udemy course and a storefront, is where the heavy lifting is done. The repository is the handout, not the product, and judging it as if it were the product would be a mistake.

Editorial conclusion

The repository fits a teacher or a student who wants the slide decks and booklets that accompany a Thai-language programming course, and who values having the material versioned in Git so it can be forked and adapted. It does not fit a team looking for a maintained component, because there is no code, no release and no licence, and it does not fit a course that needs current material, since the web topics are labelled as 2020 revisions. Before relying on it, check the licence file yourself, because none is recorded, and note that the download links point at different revisions of the tree, so a link that worked last year may fetch a different document from the one a colleague gets today. The last push was on 2026-09-30.

Frequently asked questions

Official sources

1. Issues
2. kongruksiamza/ebook-for-education on GitHub
3. README