<title>hashtagbibs</title>

Abstract

hashtagbibs (or simply: ‘bibs’) are tiny tags, allowing mere mortals to connect words to other words using pencil, voice or keyboard.
They allow polyglot tagging by expanding into multiple (reversed) BibTags, JSON or XML snippets.

For example, using OCR, scanned paper can now easily ‘connect’ with online graphs.
Think of it as the brother of hashtags: a command for tagging-this-with-that.

The goal of this spec is three-fold:

specify bibs: a plaintext polyglot tag (DSL) which expands into various languages
specify bibs as URI fragment: a way to hint browsers to filter a document on certain tag(s)
specify bibrulers: to specify bibtags microformats like visual-meta

What are bibs

bibs allow non-technical humans to specify: THIS POINTS TO THAT. In other words: compressed (reversed) BibTags, JSON or XML like so:

John please get out the laundry 

#laundry@chores@todo

This basically means:

laundry points to chores and todo

Or put in another way: laundry is tagged tag laundry and tag chores

and expands into the following formats:

BibTeX             JSON                                   XML
======             ====                                   ===

@chores{laundry}   { "this":"laundry","that":"chores"}   <tag this="laundry" that="chores"/>
                                                        
@todo{laundry}     { "this":"laundry","that":"todo"  }   <tag this="laundry" that="todo"/>

the word and john laundry can now be highlighted in the human text (or 3D object can be shown when their objectname matches). bibs are basically one step up from socialmedia hashtags, allowing mere mortals to connect words to other things using pencil, voice or keyboard.

Bibs generalizes the BibTex author/title-semantics (author{title}) into this points to that (this{that}).
There’s no precise predicates or properties, just simply ‘this points to that’, which empowers citizen annotation (an essential precursor of RDF).

NOTE: in the rest of this article, we use focus on BibTex for convenience (as it is the most terse, easiest to write/repair/speak outputformat).

format

#<textpattern>[@tag[@anothertag[...]]]

syntactically, bibs are hashtags with (concatenated) emaildomains without an extension

language	example
javascript regex	`/(#[a-zA-Z0-9_+@\-]+(#)?)/g`
shell grep	`cat textwithbibs.txt \| grep -oE '/(#[a-zA-Z0-9_+@\-]+(#)?)/'`
shell awk	`cat textwithbibs.txt \| xargs -n1 \| awk '/(#[a-zA-Z0-9_+@\-]+(#)?)/ { print $0 }'`

to qualify as a bib, a word should start with a hashtag, and (optionally) contain on or more @ characters
last bib wins: overlapping bibs overwrite eachother (last tag(s) win)
spaces are not allowed, maximum by using + to represents spaces (the+bill@todo e.g.)

#great+gatsby@book@readinglist
#great+gatsby@book

would only expand to:

@book{great+gatsby}

hashtagbib mimetypes

mimetype	expand bibs to format	hides in document
`text/plain;charset=utf-8;bib=^@`	BibTex	any BibTex
`text/plain;charset=utf-8;bib=^{`	JSON	any JSON
`text/plain;charset=utf-8;bib=^<`	HTML	any HTML

This mimetype indicates that bibs and their expanded format occuring in plain text, are automatically hidden and expanded by browsers.

For example bib=^@ means that:

lines starting with @ (regex ^@) will not be rendered verbatim by default, in order for software to:
expand bibs (starting with a hashtag) into, and interpret BibTex

automatically create/detect links between textual/spatial objects within the document (see XR Fragments)
detect opiniated bibtag microformats (visual-meta e.g.)

This significantly expands expressiveness and portability of human tagged text, by postponing machine-concerns to the end of the human text in contrast to literal interweaving of content and markupsymbols (like markdown/HTML/XML/JSON etc)

Example: extra properties

Properties are not the scope of bibs, as bibs are a one-way-street (from bibs to X).
However, its output can be further decorated with properties using the following notation:

BibTeX             JSON                XML
======             ====                ===

@article{john,     { "that":"john",     <tag that="john"
  note={foo}         "this":"article",       this="article"    
}                    "note:"foo"             note="foo"/>
                   }

Example: an textual kanban using tags

buy milk
finish paper
contact John
buy the great gatsby

#milk@todo
#contact@doing
#contact@done
#gatsby@done
#finish+paper@doing

If this text would be written on a paper, it could be scanned by a computer and represented spatially like so:

todo	doing	done
buy milk	finish paper	contact John
	buy great gatsby

One could argue that tagging a word like buy would create conflicts, but for most purposes this is really easy to spot / workaround. For serious, large bodies of text use (unique) expanded BibTags instead.

Example javascript bibs expander

Tiny but powerful implications:

expandBibs = (text) => {
    let bibs   = { regex: /(#[a-zA-Z0-9_+@\-]+(#)?)/g, tags: {}}
    text.replace( bibs.regex , (m,k,v) => {
       tok   = m.substr(1).split("@")
       match = tok.shift()
       if( tok.length ) tok.map( (t) => bibs.tags[t] = `@${t}{${match},\n}` ) 
       else if( match.substr(-1) == '#' ) 
          bibs.tags[match] = `@{${match.replace(/#/,'')}}`
       else bibs.tags[match] = `@${match}{${match},\n}`
    })
    return text.replace( bibs.regex, '') + Object.values(bibs.tags).join('\n')
}

t = expandBibs(`john, could you feed the cat?

  #john
  #laundry@chores@todo
  #some-scope#
`)

BibTeX OUTPUT:

 john, could you feed the cat?
 
 
 @john{john,
 }
 @chores{laundry,
 }
 @todo{laundry,
 }
 @{some-scope}

Bibs & BibTeX combo: lowest common denominator for linking data

Eventhough Bibs can expand to JSON and XML as well, it’s worth noting that Bibs & BibTex are closest to human thought:

Unlike XML or JSON, BibTex is typeless, unnested, and uncomplicated, hence a great advantage for introspection.
It’s a missing, lowbarrier, sensemaking precursor to extrospective RDF.

“When a car breaks down, the ones without turbosupercharger are easier to fix”

BibTeX-appendices are already used in the digital AND physical world (academic books, visual-meta), perhaps due to its terseness & simplicity.
In that sense, it’s one step up from the .ini fileformat (which has never leaked into the physical world like BibTex):

frictionless copy/pasting (by humans) of (unobtrusive) content AND metadata
an introspective ‘sketchpad’ for metadata, which can (optionally) mature into RDF later

characteristic	UTF8 Plain Text (with BibTeX)	RDF
perspective	introspective	extrospective
structure	fuzzy (sensemaking)	precise
space/scope	local	world
everything is text (string)	yes	no
voice/paper-friendly	bibs	no
leaves (dictated) text intact	yes	no
markup language	just an appendix	~4 different
polyglot format	no	yes
easy to copy/paste content+metadata	yes	up to application
easy to write/repair for layman	yes	depends
easy to (de)serialize	yes (fits on A4 paper)	depends
infrastructure	selfcontained (plain text)	(semi)networked
freeform tagging/annotation	yes, terse	yes, verbose
can be appended to text-content	yes	up to application
copy-paste text preserves metadata	yes	up to application
emoji	yes	depends on encoding
predicates	free	semi pre-determined
implementation/network overhead	no	depends
used in (physical) books/PDF	yes (visual-meta)	no
terse non-verb predicates	yes	no
nested structures	no (but: BibTex rulers)	yes

Merging (BibTagged) overlaps

When a bib (#great+gatsby@book is copy-pasted into another document (a PDF or Textfile with a visual-meta appendix e.g.):

the editor should check for the existence of @book{great+gatsby
if exist: do nothing, leave target document as is
if not: create the expanded BibTag
optionally, the editor can offer to add properties (as bibs are propertyless)

propertyless bibs, are a great way as a ‘process later’-medium (“I wrote down great+gatsby@readinglist on a papertowel/email, to scan/copy-paste it later)

What are bibs in a URI fragment

Just like regular URI Fragments they hint the browser to focus anything (Bib)Tagged:

https://website.com#@chores@todo
https://mastodon.io/myprofile/#@chores@todo
://xrworld.org/3dscene.gltf#@chores@todo

Format: #@<bibtag>[ + @<bibtag> + [ ... ] ]

Rulers (microformats)

The following allows for microformats:

Apples, cinnamon, sugar and dough make a great applepie.

#recipe-start#
#applepie@todo
#recipe-stop#

expands into the following:

BibTex             JSON                                HTML
======             ====                                ====

${recipe-start}    { "ruler":"recipe-start" }          <ruler that="recipe-start"/>
@todo{applepie}    { "that":"todo", "this":applepie"}  <tag that="todo" this="applepie"/>
${recipe-stop}     { "ruler":"recipe-stop" }           <ruler that="recipe-stop"/>

BibTex rulers have been pioneered by the visual-meta microformat as a means of organizing BibTags:

@{visual-meta-start}
@{visual-meta-header-start}
@visual-meta{
 version = {1.1},
 generator = {Author 7.6.2 (1064)},
}
@{visual-meta-header-end}
@{visual-meta-bibtex-self-citation-start}
@book{2021-12-08T10:57:03Z/TheFutureo,
author = {Frode Alexander Hegland},
editor = {Frode Alexander Hegland},
title = {The Future of Text ||},
...

Eventhough it looks like ‘nested data’, this snippet would actually get decoded to this unnested array:

[
  { ruler: `{visual-meta-start}` },
  { ruler: `{visual-meta-header-start}` },
  { k: "visual-meta", v: { version: "1.1", generator: "Author 7.6.2 1064)" } }
  { ruler: `{visual-meta-header-end}` },
  { ruler: `{visual-meta-bibgtex-self-citation-start}` },
  { k: "book{2021-12-08T10:56:03Z/TheFutureo", v: { author: "Frode ...", ... } }
  ...
]

Why not a nested tree-structure? This kneejerk reaction should always be considered ‘a temporary option’.
Don’t forget that Bibrulers are simply rulers (not blocks):

don’t try to re-invent XML or JSON
don’t promote traversing graphs (instead: just a list with tags)
they promote rather dumb, unnested, streamable lists (unlike HTML/XML/JSON) adhering to JSONLines/CSV
they are much faster/simpler to lookup, implement, (de)serialize across low- and highlevel languages.

Contact

leonvankammen|gmail.com

IANA Considerations

This document has no IANA actions.

Acknowledgments

TODO acknowledge.

coderofsalvation / hashtagbibs