💾 Archived View for bacaliu.de › 20220710-meta.md captured on 2023-07-10 at 13:39:40.
-=-=-=-=-=-=-
# Table of Contents
1. [Orgmode makes it all possible](#orgmode-makes-it)
1. [Org-publish](#org-publish)
2. [Why semantics matter](#why-semantics-matter)
3. [Is this designed to last?](#this-designed-last)
4. [Targeting Org or HTML](#targeting-org-or)
2. [Bibliography](#bibliography)
- [Nav](#nav)
- [Footer](#footer)
<a id="orgmode-makes-it"></a>
# Orgmode makes it all possible
<a id="org-publish"></a>
## Org-publish
I write this website in Orgmode. It's a markup-language like Markdown, but with way more capabilities. Just consider, this entire Document is exported out of it. The text I write looks like this:
:PROPERTIES:
:ID: 5e5e2b79-18ab-4bce-9036-b02fa0b19c91
:END:
#+title: Meta
#+subtitle: About this page
#+keywords: Meta, Orgmode, Bacaliu.de
#+filetags: Meta Orgmode
#+description: Why is this site so clean? Why does it need no cookies?
#+language: en
#+bibliography: ../ref.bib
#+cite_export: csl ~/Vorlagen/csl/rub-eng.csl
* Orgmode makes it all possible
** Org-publish
I write this website in Orgmode. It's a markup-language like Markdown, but with way more capabilities. Just consider, this entire Document is exported out of it. The thig I write looks like this:
#+caption: The org-file
#+begin_src org
...
Through Org-publish (<a href="#citeproc_bib_item_1">“Publishing (The Org Manual)” 2022</a>) I create HTML, txt and gemini out of it. This is sent over to the server with git (over GitLab). I can even put programm-code inside this document and run it. I just type the code-part and the results are generated automaticly!
for i in range(3, 19, 4):
print(f"The number {i} is one more than {i-1}")
The number 3 is one more than 2
The number 7 is one more than 6
The number 11 is one more than 10
The number 15 is one more than 14
<a id="why-semantics-matter"></a>
## Why semantics matter
Orgmode supports different markup:
- **`*bold*`:** **bold**
- **`/italics/`:** *italics*
- **`=keybindings=`:** `keybindings`
- **`~code~`:** `code`
- **`+removed+`:** <del>removed</del>
- **`_underline_`:** <ins>underline</ins>
If you see the html-verison you will notice: underlined text is green. I don't see the need of having *tree* different emphasis options. How should I decide if an important word should be bold, italic or underlined? By creating a complement to struck through text, I can correct <del>erorrs</del> <ins>errors</ins> in this intuitive way. The css and my Emacs-config makes the underlined text green. But also on other exports it is somewhat intuitive that *struck-through* text folowed by *underlined* text means first was replaced by second.
This whole thing shows a problem with most modern content. Semantics and aesthetics are often very bound together. I have to write my own css to make sure the content looks like I want. By making headings bigger and colored I don't mean the words are literally big and colorful. I mean they are a heading and you should see it as a heading. At least html-tags are somewhat semantic-driven. When your browser reads a `<h1>Heading</h1>` it can use it to mark it big by default. But it can also give this information to a screen-reader in case you are blind or something.
<a id="this-designed-last"></a>
## Is this designed to last?
While reading in Karl Voids Blog I came across the concept of designing a website to last (<a href="#citeproc_bib_item_2">“This Page is Designed to Last: A Manifesto for Preserving Content on the Web” 2022</a>). In short: By removing possible failing-points like getting JavaScript via a CDN or hotlinking images the probability of a site loosing its functionality and information is strongly reduced.
Is my site designed to last? At least I hate CDNs for exactly this reason: I would have to trust a third party when I could just download and serve the JS myself. And it's not much anyway. Just the interactive things like weather-forecast depend on it. Anything else is just formatted text.
<a id="targeting-org-or"></a>
## Targeting Org or HTML
When writing Orgmode-Documents as notes, mostly for university, I write them with the knowledge that I won't search through the documents itself. Instead I export it to HTML and view it in a browser. Some upsides about this:
- no accidental editing
- nicely formattet $\LaTeX$ equations
- dark-mode dimms images by `img.invertable{hue-rotate: 180deg; filter: invert();}`
On the other hand I cut actually profit from Orgmode's capabilities when e.g. searching for something, outline foldung, and such stuff.
In any case when writing *this* it would be bad to target the HTML-output, because the parallel output as
- **HTML:** <https://www.bacaliu.de/index.html>
- **txt:** <https://www.bacaliu.de/index.txt> - with maybe broken links
- **gmi:** - i browse it with the cli-program `amfora`
For some reason Markdown made problems, I wish that would do it too.
<a id="bibliography"></a>
# Bibliography
<style>.csl-entry{text-indent: -0; margin-left: 0;}</style><div class="csl-bib-body">
<div class="csl-entry"><a id="citeproc_bib_item_1"></a>“Publishing (The Org Manual),”. 2022. June 19, 2022, URL: <a href="https://orgmode.org/manual/Publishing.html">https://orgmode.org/manual/Publishing.html</a>, retrieved on July 10, 2022.</div>
<div class="csl-entry"><a id="citeproc_bib_item_2"></a>“This Page is Designed to Last: A Manifesto for Preserving Content on the Web,”. 2022. July 1, 2022, URL: <a href="https://jeffhuang.com/designed_to_last">https://jeffhuang.com/designed_to_last</a>.</div>
</div>
<a id="nav"></a>
# Nav
- Tags: [Meta](./tags/Meta.md) - [Orgmode](./tags/Orgmode.md)
<!-- BEGIN insert Backlinks (but there are no) -->
- Formats: [md](./20220710-meta.md) - [txt](./20220710-meta.txt) - [html](./20220710-meta.html) - [gmi](./20220710-meta.gmi)
<a id="footer"></a>
# Footer
License: CC BY-4.0
[Impressum und Datenschutz](./impressum-datenschutz.gmi)