A.01, Markdown and RSS

A few days ago I saw the following paragraph in Dave Winers blog. "I’d like to see someone like Substack or Medium, for example, who says _“Write your stuff in your favorite writing tool, export it in Markdown, and give us the link. We’ll take it from there.”[Winer2022] At this point I had several brief thoughts that don’t really have anything to do with Dave Winer’s entry. However, I will briefly reproduce them here, as they include some points that have been lying around in my Zettelkasten for a long time and were triggered by the term Markdown. It’s all very spontaneous and unsorted. I have a problem with data silos and have tried at above services. Unfortunately, I didn’t like them at all and I also don’t like data silos because my work disappears behind closed doors there and is hard to access for other people.

Then I noticed in the last years that I actually only write my texts in Markdown and I would like to have a very minimalistic writing tool in my browser in Firefox. Just a text window with export to pdf and basic stuff. Open safe to disk and functions like that. No buttons for Markdown functions because I dont need that. I find it overall ultra exhausting to render everything out in HTML. Why isn’t there actually an addon for Firefox (or a Markdown browser) on a mobile device that immediately displays online .md files in a rendered form? Then I could save the intermediate step to HTML when creating. Then, you need a browser. Mozilla has been working on a brand new browser for a while - Servo. That team got laid off last week, which sucks. That project includes standalone Rust crates for font rendering, and there’s a world-class Rust Markdown implementation, and a growing set of amazing application frameworks. Could you build a pure-Markdown-browsing browser that goes straight through this pipeline? Maybe?[MacWright2020]. This thought then also applied to RSS. Why can’t RSS be published as a new version in Markdown? In my opinion xml is more suitable for programmers and Markdown better for content creator. I find Markdown just clearer and I would super like to be able to publish a feed as e.g. rss.md file. Actually I just want to put my .md file on a server and the browsers do the rest. If you want, you can publish the link and put it in a data silo like Dave suggested.


A.01.01, Markdown and Firefox Addons

In A.01 I talked about how good I would like to see a Firefox addon that renders Markdown directly in the browser. Such an addon exists, but only for desktop computers. As I found out, the offer for Android is quite limited. As I am not an addon developer I have no idea if this is decided by the addon developer or a Mozilla team what addons are released on Android. But this would also fit well with a Markdown News Feed system. Overall I like the idea more and more not to have to learn HTML, XML, JavaScript and CSS first. Just write the text in Markdown, because I really don’t want to have to take care of everything myself. I’am a content creator, not a programmer. I imagine this in the Firefox browser as it has been solved so far with the read mode button. But I don’t want to press the button myself, I expect the browser to recognize it automatically. If it is a Markdown feed I can subscribe to it with an additional button. In the first weeks you could still solve this with the bookmark function as a workaround.

After this train of thought, I did some deeper research to find out why there are only a total of nine addons for the mobile version of the Firefox browser. And unfortunately, this is once again typical of this software. They had a good idea and then didn’t follow it up. As always, when the user base likes something, Mozilla’s managers act like any other management team at IBM, Nokia or my housing association. Don’t try anything, rather stop the project before something else could work. Especially with design decisions and not important additional features I observe this very often with Firefox. More Recommended Extensions will be enabled on release in the coming weeks as they are tested and optimized. We are also working on enabling support for persistent loading of all extensions listed on addons.mozilla.org (AMO) on Firefox for Android Nightly. This should make it easier for mobile developers to test for compatibility, and for interested users to access add-ons that are not yet available on release. You can follow our progress by subscribing to this issue. We expect to have this enabled later this month.[Neiman2020]

In the article, the criticism is addressed, but that was 2020. As so often comes nothing more from Mozilla and therefore I think the development of Firefox is a dead end. Too bad. I had looked at a few other browsers in between and Vivaldi had something with Markdown, but it is only about notes in the sidebar. In nearly every way, Markdown is the ideal syntax for taking notes. Not surprisingly, it has been a perfect match for our built-in Notes, one of the signature features in Vivaldi.[Vivaldi2018] The other browsers are too fishy for me or have too many privacy or data protection inconsistencies. Lets face it. Firefox is no role model there either. So whats next? I want to avoid having to program anything at all. Why work when other people can do it for me. Or should I just do everything on my own, like I do with my own server?


A.01.02, Markdown and Internet Archive

I solved the first little problem with my Markdown in the browser idea. Since I have been managing my private archive at the Internet Archive for a very long time and offer some of my videos, texts and photos. I have created a new account for my idea. In this account I will upload all my Markdown articles in the future and make them available to the public. Also my other files, like the HTML computer games and the corresponding screenshots will be published there. The Internet Archive has the advantage that all files are stored unchanged. So if I upload a .md file, it remains such a file. Also the link of the file is directly available, which is better for what I want to do. Below in the source directory I have posted an example link. So I have already found a file hoster for my Markdown files. I could also have rented a special hoster, but I like the philosophy of the Internet Archive and I also like librarians. Librarians are great and really nice people.

It is planned that I will set up my own server again like in the 90’. I think that this will become a new niche market in the future. Sure, nowadays you can quickly click together a Docker for every small project or buy a new VM, but over time you have pretty high costs that you have to carry every month. I don’t have those costs for my own server and I don’t trust the cloud anyway. This is bullshit network technology. Also, there is a security aspect for me personally. If I host my files somewhere and by, say a software error, they are deleted I am powerless. I could not do anything before but trust the hoster. After that, in the worst case, you can never recover the files.

If I set up my own server and something happens to it, it’s simply my own fault. Even if it means more work for me in the first moment. I should have been more careful then. But I can also fix everything much faster, because I have much more information about the problem and I will try harder to fix it, because it is my data. It’s like having a rented apartment and having my own house. I’ll work a lot faster if my own roof is broken. Seriously. I live in a student flat right now and here the balcony door has been broken for three years. I’ve called the building society four times and two times a technician was on site. The door is still broken today. Instead of the door being repaired, we received a questionnaire by mail about how we see the housing association. Unreliable and untrustworthy.

Overall, I had another interesting sounding idea on the side. Actually I could imagine the Internet Archive as a web hoster or social network. I just upload my Markdown files and they are rendered. I imagine it to be similar to Gitlab or Github Pages, where you just post your files to a repository and they are automatically converted for the viewer. The Internet Archive would then also be a data silo but I could trust this service better than, for example, the Microsoft group that bought Github. It would also be interesting to see how the Wayback Machine could be integrated into the project, because a version system for a blog is certainly quite practical. As a software developer I know that this can’t be implemented within a weekend, but maybe it is possible to let single modules work better together in the future.

Technically, almost everything you need for a blog service is already there. Login, forms for user input and a system for comments are already there. The Wayback Machine for single versions and the possibility to upload single files into a directory structure. Now we just need to create the function to recognize an index.md file as a starting point and to parse Markdown as HTML in the browser. Finished is a rudimentary blog service. An interesting idea would be to allow only free licenses so that the blogs could continue to be used by the community even after the blog owners leave or die. Also, one can then scale this further in which one would further expand the video and image service. An obstacle I see so far only with the slow servers. There one would have to look how one can solve this problem with a minimum expenditure. That the Internet Archive has a pretty hard 90` frontent is ok, would be then everything like in the early MySpace or Facebook phase.

A.01.03, Markdown as a HTML Replacement

Since a few days I’m working on some ideas that have to do with Markdown or with lightweight markup lingos. All in all, I would prefer to be able to write my articles simply in Markdown without needing another layer to transfer the article into HTML. It’s not about criticizing any technological revolutions or necessarily wanting to change, but just my laziness. Parsing something takes time and work and I am extremely stingy with both. I don’t want to waste unnecessary work, nor do I want to waste my extremely important working time. First, you need a minimal, standardized markup language for sending documents around. You might want to start with a lightweight markup language, which will ironically be geared toward generating HTML. Markdown’s strict specified variation, Commonmark, seems like a pretty decent choice. That’s the language I’ve written all my blog posts in, and the most popular language in its family. There are lots of great parsers and a big ecosystem of tools for Markdown.[MacWright2020] Here, I am concerned with the following points. I want to keep things simple. Markdown is much easier to write because the syntax is more adapted to a human’s readability. Here is an example.

<a href="file.html">file</a>

When you create a blog or build your first website, you usually learn to link to another document in HTML. This is not difficult and after some time you will learn many other great things you can do with HTML. HTML was primarily invented to create complex documents and hierarchies within a closed system. For example, a web page that resides on its own server. It was only with the spread of the web that people realized that hyperlinks could be used to link to other projects and thus to other servers. Something that seems quite obvious to us today was a really great technical insight in the early days of the web. Linking is one of the most important elements of the structure of the web and without linking, all websites, blogs and projects would just consist of individual little digital bubbles. My Zettelkasten, for example, lives from the fact that I include other people’s thoughts in my thought structures and link them here. In contrast to this, I reference the thoughts of other people who have worked on the same topic (or a completely different topic). This allows me to look at complex problems, ideas and thoughts from another persons point of view and thus come up with new solutions or ideas faster. In Markdown I can easily set a link.


Two important points stand out. The Markdown syntax is kept much clearer and can be read better by people. Especially when the article or the web page offers a lot of content, everything is incredibly clear. This is also something that I have already criticized with ‘xml’ and will go into more detail at another point. Bloggers are mostly not programmers, but people who like to write their thoughts in articles. HTML is contrary to this, because it has to be learned in a very complicated way. Markdown is much easier to learn than HTML and I can confirm that as a trained programmer and computer scientist. Over the years I use less and less HTML and write blog articles or compact web pages only in a static site generator. I had even written my own site generator with the help of Pandoc, but it has yet to be completely rewritten. Even though I love SSG, I would like to do without this intermediate step. So the question I’ve been asking myself for months is why browsers, for example, don’t adopt this system. This can have several reasons.

The web and web browsers have become Lovecraftian horrors of an unprecedented scale. They’ve long since left “scope creep” territory and entered “oh my god please just stop” territory, and are trucking on through to hitherto unexplored degrees of obscene scope. And we don’t want what they’re selling. Google pitches garbage like AMP and pushing dubious half-assed specs like Web Components. Mozilla just fired everyone relevant to focus on crap no one asked for like Pocket, and fad nonsense like a paid VPN service and virtual reality tech[DeVault2020]. This complexity of web browsers also shows how rigid and inflexible the web has become. Stadards can not be easily implemented and most people do not want that. There are also many reasons against a replacement for HTML. Markdown is much simpler and therefore developers have less possibilities if you want to build an extremely complex frontend. HTML offers much more possibilities. But I prefer a web that focuses on content and not too much on design and features. So how can we work together to turn this complex problem into a simple idea?