Exordium

The goal of this project is to provide a static website generator published using indieweb microformats, suitable for both a personal site or a special interest group.

I will also make use of convention over configuration and the choice of reasonable defaults, which can be overridden when necessary.

I will start by extracting the common elements from andrewsullivancant.ca and kwruby.ca.

Integrate with existing projects as much as possible:

URL Design

This is a map of possible URLs for an exordium website, and so might change over time. URLs may or may not get implemented as time permits.

/
  # figure out a common structure for hompeage based upon
  # https://indieweb.org/homepage
  #   * include vcard with optional additional details
  #   * optional stream with configuration
  #   * think about what other stuff to include
/[yyyy]
  /[mm]
    /[dd]
      /article/[title]
      /notes/[count for day]
      /reply
      /photo
      /rsvp
      /event
      /commit
      /request
      /read
      /pull_request
      /request
        # entry in an issue tracker or ticketing system
        # No sure if issue and pull_request should be seperate.
        # https://indieweb.org/issue_tracker
/articles
/events
/request
/posts/
  # This could contain all posts
/wiki
  /[article name]
    /changes
/tag
  /[tag name]
/git
  /[repo_name]
    /log
    /tags
    /branches
      /[branch_name]
    /[file path]
      /[oid]
/posts
  # feed of all posts
/[each post type]
  # each is a collection for seeing just that post type
  # 
/photo
  # possible endpoints for providing things like galleries and collections of
  # photos
/about
  # Is there any value in putting the identity into is own resource?
  # Yes, because the homepage provides a summary and the this contains more
  # details.
  # I do n0t totally like 'about', but it is so commonly used that it makes
  # sense to continue using it. (alternative names: identity, me, id)
  /disambugation
    # Should this be a seperate page, or just in the about page.
  /avatar
    # display the list of available avatars
    /current
    /[image_hash]
      # default to 80x80
      /[size]
    /[email_hash]
      # provider either re-directs or copies to the images
      # default to 80x80 again
  /gpg
  /resume
    is cirriculum_vitae at better name for this?
      probably because it is not as common as usage
    allow various formats of resume (pdf, word, opendoc)

Questions?

where should subscrition links go?
- RSS feed + indieweb idea of a link which is readable and subscribe-able
www vs no-www for the conical URL
- https://stackoverflow.com/questions/1109356/www-or-not-www-what-to-choose-as-primary-site-name
- http://www.yes-www.org/why-use-www/

Research

Git repository displays

GitHub
Gitlab
cgit
stagit
sourcegraph
indieweb has some note about this
- http://indiewebcamp.com/issue_tracker
- http://indiewebcamp.com/code
microformats to create
- hTicket
- hRepository
- http://indieweb.org/git
- http://indieweb.org/code
- http://indieweb.org/issues
- http://indieweb.org/issue
patch work
- a git tool for managing mailing list patches on the web as well
- http://jk.ozlabs.org/projects/patchwork/
git-bug
git-appraise

Ideas

every feed to posts should include
- HTML5 indieweb microformat feed
- RSS
- iCal
https://github.com/statonjr/middleman-sitemap
gpg publishing
- looks like it can work with indieauth
- http://indiewebcamp.com/pgp
- hawkpost.co, a system for sending GPG encrypted messages from people who do not have GPG setup, uses OpenPGP.js
static file encryption with GPG
- http://git-annex.branchable.com/encryption/
- http://git-annex.branchable.com/design/encryption/
- http://obnam.org/encryption/
- https://github.com/ahoward/middleman-gibberish
- can the webauthn API help with using GPG de-cryption in the browser?
separate command/control services from static publishing service
- authenticate publicly accessible command services with client certificates
  - http://blog.nategood.com/client-side-certificate-authentication-in-ngi
- consider including a micropub interface
- consider too to help with content creation
  - automated spelling, grammar, proofreading, reading quality
  - meta data suggestions (e.g., HashRobot
- https://github.com/yabwe/medium-editor
create helpers for setting site license (CC, PD, all reserved)
create a consistent identity structure
- integrate with namecoin
- handle multiple and deprecated email addresses
- disambugation pages
- xfn related things
- webfinger
- indieauth
- openid
avatar support and POSSE
- save all avatars which have been used, and provide a browsing page
- specify the current avatar for the site
- POSSE, default or an overridden, avatar to
  - gravatar
  - twitter
  - linkedin
- provide libravatar support
add public GPG key publishing
- keep all keys including revoked onces with their revokation certificates
- associate with appropriate email address
- point to the appropriate default key for communication
implement POSSE to various other services
- Twitter
- Meetup
- LinkedIn
- medium
  - http://www.elezea.com/2016/02/medium-as-rss-reader/
- image sizes on various services
- initially implement to be executed from a workstation/laptop
  - this will avoid having to worry about how to store credentials on the network
  - investigate a secure way to do this in the future
- SNAP wordpress extensions
investigate storing POSSE and webmention information in data-branches
get push notifications working
- PubSubHubbub
- http://christopheducamp.com/w/PubSubHubbub?christopljw_session=mke14ssbovehvdo6tq94bpff44
- other formats to use?
easy SSL with lets encrypted
adding an easy option for deploying to various sites:
- internet archive Wayback Machine
- P2P networks (e.g., IPFS)
- S3
- github pages
- CloudFlare
- rsync
- ftp
- google amp
  - Amplify for Jekyll
- facebook instance articles
- google shield
- [Scuttlebut social network]
make it easy to add a .onion address
write a plugin for doing draft blog management in branches
add link checking and wayback machine support for content and links
- check links regularly
- link to the wayback machine in content if the original page disappears
- can I send pages I link to to the wayback machine?
  - yes http://lifehacker.com/manually-archive-web-pages-by-submitting-them-to-the-wa-1580713309
add support for auto filling different kinds of links
- investigate if this would work in Markdown or now
- easy links to things like
  - internal pages
  - wikipedia
  - github
  - public contact/person information
publish contacts and membership with XFN and rel tags
- http://microformats.org/wiki/rel-group
- think about what level of privacy would be appropriate?
  - who should be able to access what
  - how to authenticate (indieauth with the related page, GPG encryption)
search
- use a client side javascript search
- http://manas.com.ar/blog/2015/10/22/middleman-search-client-side-search-in-your-middleman-site.html
- https://stork-search.net/, Rust and WASM based client side search
allow code examples to be stored in separate files and inlined into articles
- this should make it easier to check and execute the code
- would this then need to include a gist like system?
look into integrating the mermaid diagraming tool
add support for including code snippets as separate files which are include into articles
make it easy to do mentions in articles/wiki/etc
- https://seblog.nl/2017/06/14/6/at-mentioning-people
things to PESOS
- OpenStreetMap
- MusicBrainz
- Wikipedia
figure out how to publish events and session
- both allow events to be created, this could be PESOS from other places
- example https://aaronparecki.com/presentations
- and POSSE to other sites
  - https://www.slideshare.net/
  - https://slidr.io/aspleenic/rails-still
include vcard list with exordium, for linking to others
- https://indieweb.org/nicknames-cache
- use ppl to manage it
support posting and subscription over email
- how to make it easy
  - have a paid which is the default
  - and easy support for something self-hosted
- consider using gitlab CI
- allow posting by email, maybe something like matrix as well
pick a method for doing slides with works well with this
- https://github.com/geraldb/talks/blob/master/slideshow-s9.md
describe at least one method for CI/CD/automatic deploying
- GitlabCI
- https://github.com/alestic/aws-git-backed-static-website
follow up on posting my own editings to OpenStreetMap
- https://indieweb.org/edit
work on bring icky style or bookmarking by URL
- https://indieweb.org/geo-wiki
  - this describes a similar idea
  - except the wiki is associated with URL/URNs instead of lat/long
  - actually has similar issues of wanting to link to nested pages
  - e.g.,
    - a page for a less precise lat/log should link or embed a page for a more precise lat/log up and down
    - a URL page should similary link to other URLs up and down
  - /wiki/
  - /uri/
  - /geo/
    - I think that I like this better it makes things more clear
    - it is actually consistent with something like /photo/, where a photo might get similary updated with information over time
  - https://tools.ietf.org/html/rfc3986#section-1.1.2
  - convert the icky formath to markdown + yaml annotations
  - make it possible to then link to photos
add linting to the project and keep adding tools
- webpagetest.org
- lighthouse
figure out indiemark and then work through the levels
- https://indieweb.org/IndieMark
verify that HTML header information is fully populated
- https://gethead.info/
add ccRel link formatting to everything
- follow the W3C linking standard for it
- document using it in indieweb stuff
book reading and review
- https://macwright.com/2022/02/24/indie-bookshelves.html
- https://macwright.com/2020/12/24/the-new-reading-stack.html
create a useful 404 page with dynamic javascript
- https://ben.balter.com/2022/06/30/helpful-404s-for-jekyll-and-github-pages/
middleman has some tools for processing files with frontmatter
- this could be used for reading and writing those files
- ::Middleman::Util::Data.parse
- https://stackoverflow.com/questions/36948807/edit-yaml-frontmatter-in-markdown-file/36984468#36984468

HTML5/Microformats

for git repository
- consider flag to indicate intended usage (e.g., for use, clone only, deprecated)
- want to integrate web comments on all parts of the repo
  - e.g., commits, lines, PRs
- would like to integrate display of related wiki and issues
- also include an intent flag
  - prototype
  - clone
  - deprecated
  - for_use
  - what else?
- generate arbitrary diffs in the browser using web worker + git.js
for projects
- want to be able to go all the way from idea to full running project
- embed list of links and tickets
- and then allow links out to the same later on
- this will be useful for
- http://microformats.org/wiki/h-product
- http://microformats.org/wiki/project
- https://github.com/ewilderj/doap/wiki
- http://microformats.org/wiki/project-formats
use a HTML5 structure with main and related structure tags
- https://developer.mozilla.org/en-US/docs/Web/Guide/HTML/Using_HTML_sections_and_outlines
proposal for listing book information library JSON

Wiki

add category support, like wikipedia

#Inspiration This to review for ideas:

https://github.com/crtvhd/middleboy
https://github.com/gregory/middleman-settings
Locking the Web Open: A Call for a Distributed Web
- javascript (e.g., client side search engine)
- bitcoin
- IPFS/bittorrent
- Namecoin
- public/private keys
investigate pushing git repos or projects to aggregators
- openhub
- github
- gitlab
- bitbucket
- rubytoolbox
- article on this topic by flameeyes, Project Memory
Autotune Vox Media's middleman based CMS
Ain't CMS
The Future of Websites: Headless CMS

acant/exordium