Scripting-News: a blog backup target and a comment box, nothing else

The readme answers how to use the repo by saying there is no way to use it

The most honest section in this readme is the one headed as instructions. It opens by calling the question a trick, and the answer is that you need to read the blog. The repository is not a tool with an interface; it is a container the blog lives in. What you are invited to do with it is limited to two things. You can post a question in the form of an issue, with the readme's own caveat that a discussion may or may not follow. And you can watch the repository, which puts an icon at the top of every page and gets you an email when something new appears. The tone throughout is conversational and addressed to programmers, which is consistent with the rest of the file.

A discussion starts with a site-scoped search, then an issue

The one workflow the readme does document is described with a heading that reads like a redirect from a search query to an issue tracker. The pattern is: pick a technical subject worth discussing with the blog's readers, find the existing posts about it, and open an issue on the repository to raise it. The link in that heading is not a discussion thread but a web search restricted to the blog's own domain for one specific topic phrase, which shows the intent, since the subject has usually already been written about and the issue is a pointer to that prior work rather than a question about it. The author's stated reason for using issues at all is that the forge already has the collaboration tools and that the interface is one programmers already know.

The nightly backup script does not live in this repository

The backup mechanism is the one genuinely reusable thing here, and it is deliberately kept elsewhere. A script runs on one of the author's own servers, every night at midnight Eastern, and commits whatever is new on the blog into a directory in this repository through the GitHub API. The source of that script is not in this tree: it sits in a separate repository under the same account, in a single JavaScript file, and is released under the MIT licence. The explanation of what the script actually does is not in this repository either; it is linked as a feed file hosted on a personal site. So if you want to copy the arrangement, the parts you need are the script from that other repository and a description you have to fetch separately.

Nine directories at the root, and only one of them is the backup target

The repository root is a readme and nine directories, and the readme only explains one of them, the blog directory that the nightly script writes into. The rest are separate small projects that happen to share an account: a how-to directory, a manifesto, a miscellaneous folder, a podcast directory, a directory of cross-references, one for Twitter-related material, one for publishing through the forge, and an RSS-in-JSON project. Nothing in the readme explains what most of them contain, and the naming is inconsistent enough that a visitor cannot tell whether a directory is a document, a tool or a data set. The repository is a workbench with a backup attached, not a coherent project.

The update log has two entries, and the newer one is a readme draft

The readme carries an update section with a version number, a date and an author initial on each line, and it contains exactly two. The first, version 0.40 dated June 2016, says the project was started. The second, version 0.41 dated June 2017, records a first draft of the readme for the RSS-in-JSON project. So the last documented piece of work in this file is a draft readme for one of the nine directories, and the version numbering stopped at 0.41. The contrast with the repository's actual state is the interesting part: the default branch here is master, and its most recent push is dated September 30, 2026, which means the file describing the project has not been kept current even though the branch is.

No licence here, while the script that fills this repository has one

There is a licensing asymmetry worth pausing on. The backup script that writes into this repository is released under the MIT licence, in its own repository, and the author is explicit about that. This repository, which holds the archived blog content itself, carries no identified licence at all, and no licence file appears at the root. For a reader that makes the content awkward to reuse: the tool is permissively licensed, the output of running it is not. The invitation at the end of the backup section, aimed at other technical bloggers and suggesting they archive in a similar way, does not come with a licence grant for the archives, so anyone wanting to reuse the posts needs to ask the author.

The blog is the product, and the repository is storage plus a contact point

Stepping back, the shape of this project is easy to misjudge from the name. The repository description frames it as a decision to move blog work onto a forge, on the grounds that the forge has good communication and collaboration tools and might as well be hooked up to a blog. The readme then describes three separate things layered on that decision: a nightly archive of the site's own content, a public issue tracker used as a way to raise topics with readers, and a set of unrelated side projects sharing the account. None of the three is software you would install, and the only reusable artefact is the backup script, which lives next door. Read as a publication's source repository rather than a product, most of the rough edges in the readme stop being rough edges.

Editorial conclusion

Use this repository if you run a blog and want a nightly copy of it committed somewhere public, because the mechanism is small, MIT-licensed elsewhere and built on the GitHub API, and the author explicitly invites other bloggers to copy the idea. Do not use it as a discussion forum with any expectation of a reply, since the readme says a discussion may or may not follow an issue, and do not reuse the archived posts without asking, because this repository carries no identified licence even though the backup script does. Before you copy the setup, read the script in its own repository rather than here, and expect the documentation of what it does to live off-site in a feed file rather than in this tree.

Frequently asked questions

Official sources

1. Official README
2. Project repository
3. README
4. Releases