Interesting site I built on top of Github Pages

[veltzer]

Interesting site on top of GitHub Pages

I did an interesting site on top of GitHub pages. The interesting part is not mkdocs (which is quite common) but the media link there where you can browse my own personal data of various sorts.

I really don't know what I want to do with it, but thought I would post the link here so that people could comment and maybe get ideas from me and contribute some back.

Also I wanted to hear what the GitHub engineering team thought of it.

The URL is https://veltzer.github.io/media

[IklilIlm]

This is a really cool personal archive. I love the level of detail—especially the "Device" and "Location" fields. It adds a weirdly nostalgic layer that most people wouldn't think to track.

A few thoughts:

The ratings are interesting. You've got a bunch of 7s and 8s, some 6s, and a lot of unrated ones with "Progress: X/24". It looks like you're using this as both a catalog and a way to track unfinished courses. That's actually a really nice use case—like a personal learning backlog.

Potential feature ideas:

A stats page—total listening time, average rating by lecturer, most-used device, etc. People love seeing the numbers add up.

Tagging or filtering by "location" (Night Walks, Gym, Car). That's a pretty unique dimension—it tells a story about when and where you learn.

Export to something like a CSV or JSON. If this is a personal data archive, having a way to get the data out would make it feel more "future-proof."

What I'm curious about:

Is this manually entered, or do you have a script generating it from some source?

The "Progress" field suggests you revisit courses—do you update entries over time?

For the GitHub team perspective:
This is a great example of using GitHub Pages for personal knowledge management. It's not a blog, not a portfolio—it's a personal data dashboard. There's a growing trend of people using static sites + GitHub as a "personal data warehouse" (fitness logs, reading lists, etc.), and this fits right in. Might be worth featuring as an example of what Pages can do beyond documentation.

Curious to see where you take it. Even if you "don't know what you want to do with it," just having the data in a clean, searchable format is already valuable.

[veltzer]

The actual data is in yaml files. I mark off anything I finish (course, series, book) and the code does the rest. The yaml files have a structure like this:

name: "The Concerto"
lecturers: ["Robert Greenberg"]
date_utcz: "2016-11-18T08:53:26Z"
date_timezone: "Asia/Jerusalem"
device: "Sansa Clip 2"
location: "Car"
rating: 8
review: "Great lecturer"
great_courses_id: 7270
great_courses_slug: "the-concerto"
The build system does the rest.

A stats page—total listening time, average rating by lecturer, most-used device, etc. People love seeing the numbers add up.
There is a statistic page for each topic but not overall one. Good idea. Will do.

Export to something like a CSV or JSON. If this is a personal data archive, having a way to get the data out would make it feel more "future-proof. -> There is no need for this. The "Source of Truth" is already in YAML.

The "Progress" field suggests you revisit courses—do you update entries over time? Yes. Every time i finish a chapter.

Thanks.

[Deeppatelll]

Really interesting idea using GitHub Pages as a personal media browser is clever. From the initial view, the page feels very minimal, with just a sort control and a page-visits loading indicator, and your site is set up so media.html is served as a static page through MkDocs Material.

A few improvements that would make it much stronger: add a clear title/subtitle explaining what this media collection contains, show thumbnails or cards instead of only a plain list, and include richer metadata like type, date, tags, and file size. Search plus filters for category/date would make browsing much easier, and a better loading/error state for the visits counter would make the page feel more polished. If the content is personal data, a short note on what is public and what is private would also build trust. Overall, the concept is strong a bit more structure and discoverability would make it much easier to use and contribute to.

[veltzer]

Will take the notes into consideration...:) Just put them in my TODO list.

Thanks...