# Journal is published using Common Markdown

Now we have a dotnet version of the publisher that uses an in-house and a CommonMarkdown parsers. The CommonMarkdown is almost identical to the original parser, based on AWK version. However, some differences still need to be fixed:

• Inline `code` is not appearing.

Make the first (final?) version in Python using MarkdownIt to make accomplished and return to backlog to check how it goes on solving other issues.

The idea is to once each sub-issue is accomplished it will be already being used in the main build, as it is being done for metadata.

The linked sub-issues must apply.

Generate months pages

Start to use jinja to render html. This first step demonstrates how to use jinja to read a template file and generate months files. Need to complete this first example with remaining data and probably doing so complements metadata from parser.

IMPORTANT

Need to change template folder files and provide required info in posts variable being populated.

Generate tags pages

Generate index

Generate remaining pages

# Subtitles inside text

Scenario: I want to journal everyday and sometimes it is better to inside one text to quote different content like a movie I watched, a recipe I thought of or something work related. Usually I would need to write one post per subject which would generate separate texts without cohesion.

Idea: it would be nice to have a way to index subtext so I can write about several subjects inside the same text and still be able to search or reference the subjects for the reader. Let's draw an example:

# Main title
date: 2025-11-20
tags: journal
Today I worked about some #csharp code that had me spend some hours because it
needed to configure #windbg in a certain way that I will describe here.
I also watched [The Title of the Movie] today, which I think does not deserve a review,
but could be searchable inside this text if someone is curious to know if I wrote
something about it. #movies
Notice that I could add tags as hashtags inside each paragraph, which could
count as a separate piece of text searchable in tags. Also in the case of
The Title of the Movie it will be listed as a different post in the blog or as a
different title in some e-book indexed by title. And as it has #movies tag it
will also be listed a a movie.
Consider that The Title of the Movies does not exist and it is not referenced. If
there was an entry then we could just reference and then the reversal references
implementation (TBD) would work. Also need to be decided if a text that does
this kind of empty link (no reference) could in the future reference some
new entry just because, without the need to enter the reference in the old
text. This would be cool, but has to pay attention about other signals
like `[1]` (this need to be entered as code, or it is a bottom reference?).

Ideas to be implemented here and could be divided in other issues:

• Reference paragraphs.
• Tag paragraphs.
• Consider subtitles as able to be listed and search (including by its internal tags).
• When a new text with same slug is created reference old text with same pseudo-slug.
• Reverse reference (paragraphs that reference other titles and subtitles).
• Consider bottom references ([1]) when thinking about this.

# Apply tags based on metadata

This issue is to allow to separate tagging based on metadata from tagging based on content. Tagging based on content should be by example the author tags something like "movies" or "cooking" because the content is related to these subjects. Tagging based on metadata should be by example to tag some post as "miniviews" because the content has only one paragraph and it is tagged as "movies" already. Or to tag something like "quotes" because there is only one paragraph and it is a quote.

This tagging by metadata form should not be in the original text, it should be configured as metadata analysis in a separate config form. The author should be able to create different tags based on metadata for different purpose. Tagging based on content should remain simple. The author should not think anymore about to tag content based on its metadata, only in its content that cannot be deduced without IA or author purpose.

# Awk ebook publisher not linking posts (brackets stay the same)

Could have be inserted during brackets change? Nothing changed at all for ebooks, just variable monitoring.

Apparently some links were really broken since my deleting a lot of posts review approach: https://caloni.com.br/2023-10.html#o_que_vem_ai_na_mostra_2023_47a

# Awk publisher tags some posts as private even not being private

I think it is related to be the last issues or something.

# Tags hierarchy by frequency

How about a new future issue where not only the tags are related to each other, but the main ones are deduced by their volume, and the smaller tags are hidden from the overview? This would make navigation simpler for those looking to find something through them.

At the same time, this would allow for granular categorization. For example, a level-3 tag (rarely used) could be designated as a keyword (like the keywords in a research paper) or something along those lines.

The configuration could be based on the percentage of total posts, regardless of how many levels there are. There wouldn’t be direct control from the author, but as they start writing more about a particular topic, the tag would automatically be promoted to a category (main tag).

# Better Kindle navigation

Today we can go to the beginning of book and the index with some huge groups, where the most useful is the letters and tags. We can do better:

• Months and sections (see #2) listing titles.
• ~Better tags listing (all at once as the script output).~
• Link to the real beginning and end of texts, chronologically.

# Separate config from scrip

Today the scripts contain personal blog and book information such as links, author, description.

We are going to move all that to configuration and style (proto theme) settings.

# Batch change posts

• rename tags and entire file
• select, move and delete posts

# Footnotes

Support to footnotes in ebook or even in the blog.

# Sister tags

The idea is to remove necessity of posts to contain every possible tags they fit in to just the main ones (hopefully one). The sister tags will be shown in each tag page so the user can click on the main tag and navigate to other similar tags. By example, a post has "movies" as tag and inside "movies" tag page we see other tags to navigate to its page as "cinema", "miniviews" and so on.

This information can be deduced by the posts and all the tags in common. By example, if one post have the tags movies, cinema and anime then when clicking in each of one of these tags it will list the other tags as a option to navigate.

Eventually there will be a post about a particular group to nest all these tags but no singular post will have to have all those.

# Multiple links to single text

Keep the original position of the text, but once updated with a second date link this date into the second month with the content pointing to the original position. By example, in this post:

# Title
[date]:# "2012-02-01 2025-03-02"Paragraph1
Paragraph2

We put the post content inside 2012-02.html month page but in 2025-03.html month page we see an entry for the same post, but it is pointing to the 2012-02.html page.

# Sections as month/year

Today the blog is structured in "chapters" that are the month and year when each post is labelled. They are put together into the same HTML page as a chapter tied by date.

The idea here is to continue to use date as a marker, but also support other kinds of markers to group posts together as a classic chapter from a book. It should be possible to create, maybe inside the tags, values that are parsed as new groups despite the date each one belongs.

The final view will be posts organized by these new markers and not appearing anymore to the months market. It will be shown in the blog just as the month/year pages, but with another name to group.

# O corpo interior

O corpo interior está na fronteira entre a forma e a essência, que é a sua verdadeira natureza. Nunca perca o contato com ele.

Eckhart Tolle (O Poder do Agora, 1997)