From IndieWeb

IndieWebCamp Nuremberg 2023 Demos was a session at IndieWebCamp Nuremberg where participants demonstrated and described what they made & built during the weekend.

Felix De Montis

  • https://felix.dm/
  • First task was to add Swarm checkins. That has not been added, currently the last location is here, the tollwerkstatt. It also shows stickers.
  • Second he wanted to move his shirts, but realised why he did not do that before. The images are very large, and does not want them in his Kirby content folder (which also syncs to git)

Sebastian Feldmann

  • https://maker.sfeld.eu/
  • Tried the automatic migration to Nikola CMS, but that did not work
  • It migrated posts, but no pages or pictures, whihc all needs to be done by hand. 99% by hand, 1% automatic, for him
  • Was able to publish a new version of his site: https://make.sfeld.eu/
  • The new one has slightly different styles. Same content. Looks almost the same. But now there is a English/Deutsch switch up top in the header
  • The articles are not translated yet. But UI things like “Read more” are done.
  • Has been writing a blog post about the migration, that has yet to be published. No more tech todos now.


  • Angie
  • Shows a locally made index.html showing text and photos taken

Sonja Weckenmann

Sophie Brunner

Joschi Kuphal

  • Joschi Kuphal https://jkphl.is/
  • Also has a new website. He “switched off” his old website, and has a much simpler new landing page: https://jkphl.is/
  • Enough microformats to be able to log back into the wiki
  • It works with GitHub as the auth provider, though not demoable on the demo computer
  • All the old links to posts still work

Roman Komarov

Sven Knebel

  • Sven Knebel https://www.svenknebel.de
  • He had the ability to have posts private, depending on audiences
  • But those posts where only available through weird ways
  • Getting a token, then curl with the token
  • No way to see it in the browser, not even by him as the author
  • So he has changed that now.
  • Introduced a login field for password credentials
  • When logged in the private post shows up, and it has extra styling (big red border) to make it visible as a private post
  • A little padlock icon can be used to see who is allowed to read a post
  • The private posts also show up in subsections of the site, like the category view
  • The login page only works for him, but for the future might allow others to also login and be able to see private posts


Björn Stierand

  • https://bjoern.stierand.org/
  • found out that his Known site is broken
  • Got IndieAuth setup
  • shows local Obsidian client
  • Uses a plugin for Digital Garden inside Obsidian
  • Is able to publish through a GitHub repo, where Vercel will take it from and make it publicly available on the web
  • Live creating a new note in his local Obsidian
  • There is a dg-publish (digital garden publish) flag that can be toggled
  • Then by telling Obsidian to deploy it will kick off the process
  • Shows the GitHub repo with the notes
  • https://bjoern.stierand.org/iwc went live automatically
  • The connected graph of pages is also made available in the browser
  • There are interlinks to pages that are not published, and those result in missing links
  • Wants to see if he can filter out these broken links

Jeremy Cherfas

  • Jeremy Cherfas
  • Very grateful to Jeremy Keith for the hint of doing the easy thing first
  • Fixed the sidebar on the search result page.
  • Then tried very hard the rest of the day to get Known to use the new source for stamen map tiles, but could not get it to work
  • So frustrating that he spent the final time of the day to try and get an export of all the known data (which first went wrong even though an RSS export was supposed to work)
  • Tried a wget to get a full scrape of the website instead.
  • Looking for ways to get the source of all the posts out of Known, and potentially put it somewhere else.
  • Maybe micro.blog / probably.
  • Thanks everyone who helped today


  • https://webrocker.de/
  • First task was easy: filed an issue on the WordPress plugin and “pestered”  Matthias Pfefferle
  • Did get his blog linked to the fediverse
  • @blog@webrocker.de has a presence on the fediverse now
  • Falls back to the default WordPress logo, and now the guitar that he wants
  • Spent the last 2 hours trying to find out where the information needs to come from.
  • mastodon.social seems to have a slow caching, so even when he found where to put the header image he had not seen it yet

Jan Sauer

  • Jan Sauer https://jansauer.de
  • Got a lot of stuff started
  • Now has a website live again, with the last Strava activities on it
  • Now that the basics are there he is fine with having to play further with it
  • lots of things to do

Martijn van der Ven

  • Martijn van der Ven https://vanderven.se/martijn/
  • first easy task: update profile picture
  • fix IndieAuth/selfauth so he can log into the indieweb wiki again
  • started to redesign: deleted all the CSS
  • updated h-card to include more details for his birthday (esp location)
  • now marked up as a h-event
  • microformats do not all handle objects under dt-* properties well
  • and the spec is not well-defined for some of that

Jeremy Keith

  • Jeremy Keith https://adactio.com
  • added autolinking for @-@ mastodon urls
  • Regex he found to match @-@ failed
  • code he tried broke on unicode chars in domain
  • As one of the latest posts mentioning briansuda had icelandic characters
  • Probably on the train tomorrow he will blog about what he has done, and share the regex
  • Related posts was the other task
  • Opens a “boring webdevelopment post” tagged with boring javascripty things
  • Responses come first after the post, but after responses there is a new section related
  • Comes up with a bunch of other javascripty articles
  • At least 3 tags had to match
  • Ordered by score: the most similar tagging
  • Not efficient at all, lots of separate SQL queries, but as soon as someone visits a page it gets cached so he does not do it all the time
  • Would like to show related links, out of his link section
  • The other thing he was working on was linkrot
  • Worked to implement Remy’s idea https://remysharp.com/2023/09/26/no-more-404
  • Scrolls to links from 2005
  • Within 1 second of clicking, it will try to ping the URL when clicked, if it returns something that does not look correct, it will check the Internet Archive instead
  • If the link clicked is inside an h-entry, grab the dt-published, so it can check internet archive for a copy of the link at the right time
  • Sometimes the most recent version of a URL, even on the Internet Archive, might be a page that is missing the old content
  • If the script is uncertain about anything, it will let you go through to the original link

Paul Robert Lloyd

Calum Ryan

  • Calum Ryan https://calumryan.com
  • Did manage to connect his site to fediverse in parts
  • Now a banner on his site that also is the cover image on e.g. Mastodon
  • Also pulls in profile, but has not gotten the posts to show up yet
  • Has tried to get his gallery to show a few more images. Still needs some more work. But a longrunning issue with images not resizing was fixed
  • All the images should now be resizing nicely after getting the imagemagick config correct

Sara Jakša

Sebastian Greger

  • Sebastian Greger https://sebastiangreger.net/
  • Has achieved something with both topics.
  • Launched a little update. Changed the journal page to be less of a feed and looks more like content. Dates are very small, content foreward.
  • Got photo posts to work in there
  • Something he wants to work towards is a bookshelf, now has an early stage of bookmarking a book with ISBN information
  • Quote posts added there as well.
  • Went deep on linkrot, documented his thinking at https://sebastiangreger.net/2023/10/link-rot-and-schrodingers-url
  • Reviewed and shared code snippets, e.g. code for how he pings the Wayback Machine.
  • This also checks whether there is a recent snapshot
  • Also uses a library called Graby2 to store the main content of a newsarticle locally
  • Used Linkchecker to check all the links. But noticed a bunch of both false positives and negatives
  • Because of the false matches, using Kirbycodes to manually be able to do something
  • Conclusion: always keep the original link, but start showing archival links. Potentially after a specific time.
  • Schrödinger’s URL: a web resource either exists or does not exist until the link is opened.

See Also